Research Article Estimation of the Mechanism of Adrenal Action of Endocrine-Disrupting Compounds Using a Computational Model of Adrenal Steroidogenesis in NCI-H295R Cells Ryuta Saito, 1,2,3 Natsuko Terasaki, 4 Makoto Yamazaki, 3 Naoya Masutomi, 4 Naohisa Tsutsui, 4 and Masahiro Okamoto 1 1 Graduate School of Bioresource and Bioenvironmental Sciences, Kyushu University, Higashi-ku, Fukuoka 812-8582, Japan 2 Biology Research Laboratories, Mitsubishi Tanabe Pharma Corporation, Toda-shi, Saitama 335-8505, Japan 3 DMPK Research Laboratories, Mitsubishi Tanabe Pharma Corporation, Toda-shi, Saitama 335-8505, Japan 4 Safety Research Laboratories, Mitsubishi Tanabe Pharma Corporation, Kisarazu-shi, Chiba 292-0818, Japan Correspondence should be addressed to Ryuta Saito; [email protected] Received 19 November 2015; Revised 20 January 2016; Accepted 21 January 2016 Academic Editor: Steven J. Bursian Copyright © 2016 Ryuta Saito et al. is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Adrenal toxicity is one of the major concerns in drug development. To quantitatively understand the effect of endocrine- active compounds on adrenal steroidogenesis and to assess the human adrenal toxicity of novel pharmaceutical drugs, we developed a mathematical model of steroidogenesis in human adrenocortical carcinoma NCI-H295R cells. e model includes cellular proliferation, intracellular cholesterol translocation, diffusional transport of steroids, and metabolic pathways of adrenal steroidogenesis, which serially involve steroidogenic proteins and enzymes such as StAR, CYP11A1, CYP17A1, HSD3B2, CYP21A2, CYP11B1, CYP11B2, HSD17B3, and CYP19A1. It was reconstructed in an experimental dynamics of cholesterol and 14 steroids from an in vitro steroidogenesis assay using NCI-H295R cells. Results of dynamic sensitivity analysis suggested that HSD3B2 plays the most important role in the metabolic balance of adrenal steroidogenesis. Based on differential metabolic profiling of 12 steroid hormones and 11 adrenal toxic compounds, we could estimate which steroidogenic enzymes were affected in this mathematical model. In terms of adrenal steroidogenic inhibitors, the predicted action sites were approximately matched to reported target enzymes. us, our computer-aided system based on systems biological approach may be useful to understand the mechanism of action of endocrine-active compounds and to assess the human adrenal toxicity of novel pharmaceutical drugs. 1. Introduction Because steroid hormones play an important role in a wide range of physiological processes, the potential to disturb endocrine effects is a major concern in the development of novel pharmaceutical drugs such as etomidate and aminog- lutethimide [1]. e adrenal gland is the most common target for toxicity in the endocrine system in vivo, because steroid hormones are primarily synthesized through enzymatic reac- tions in the adrenal cortex [2–5]. Indeed, in these studies based on chemically induced endocrine lesions observed in vivo, the most frequent site of reported effects was the adrenal gland. erefore, the prediction of human adrenal toxicity based on the mechanism of on- or off-target actions in the early stages of drug development is important. e NCI-H295R human adrenocortical carcinoma cell line has been used to elucidate mechanisms of adrenal steroidogenic disrupting compounds [1, 6]. e H295R cell line was established by Gazder and his collaborators in 1990 [7], which expresses all key steroidogenic enzymes and steroidogenesis-related proteins [7–9]. H295R cells have the physiological characteristics of zonally undifferentiated human fetal adrenal cells and the ability to produce steroid hormones found in the adult adrenal cortex [1, 7, 9]. In vitro Hindawi Publishing Corporation Journal of Toxicology Volume 2016, Article ID 4041827, 19 pages http://dx.doi.org/10.1155/2016/4041827
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Research ArticleEstimation of the Mechanism of Adrenal Action ofEndocrine-Disrupting Compounds Using a ComputationalModel of Adrenal Steroidogenesis in NCI-H295R Cells
Ryuta Saito123 Natsuko Terasaki4 Makoto Yamazaki3 Naoya Masutomi4
Naohisa Tsutsui4 and Masahiro Okamoto1
1Graduate School of Bioresource and Bioenvironmental Sciences Kyushu University Higashi-ku Fukuoka 812-8582 Japan2Biology Research Laboratories Mitsubishi Tanabe Pharma Corporation Toda-shi Saitama 335-8505 Japan3DMPK Research Laboratories Mitsubishi Tanabe Pharma Corporation Toda-shi Saitama 335-8505 Japan4Safety Research Laboratories Mitsubishi Tanabe Pharma Corporation Kisarazu-shi Chiba 292-0818 Japan
Correspondence should be addressed to Ryuta Saito saitouryuutamcmt-pharmacojp
Received 19 November 2015 Revised 20 January 2016 Accepted 21 January 2016
Academic Editor Steven J Bursian
Copyright copy 2016 Ryuta Saito et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited
Adrenal toxicity is one of the major concerns in drug development To quantitatively understand the effect of endocrine-active compounds on adrenal steroidogenesis and to assess the human adrenal toxicity of novel pharmaceutical drugs wedeveloped a mathematical model of steroidogenesis in human adrenocortical carcinoma NCI-H295R cells The model includescellular proliferation intracellular cholesterol translocation diffusional transport of steroids and metabolic pathways of adrenalsteroidogenesis which serially involve steroidogenic proteins and enzymes such as StAR CYP11A1 CYP17A1 HSD3B2 CYP21A2CYP11B1 CYP11B2 HSD17B3 and CYP19A1 It was reconstructed in an experimental dynamics of cholesterol and 14 steroids froman in vitro steroidogenesis assay using NCI-H295R cells Results of dynamic sensitivity analysis suggested that HSD3B2 plays themost important role in the metabolic balance of adrenal steroidogenesis Based on differential metabolic profiling of 12 steroidhormones and 11 adrenal toxic compounds we could estimate which steroidogenic enzymes were affected in this mathematicalmodel In terms of adrenal steroidogenic inhibitors the predicted action sites were approximately matched to reported targetenzymes Thus our computer-aided system based on systems biological approach may be useful to understand the mechanismof action of endocrine-active compounds and to assess the human adrenal toxicity of novel pharmaceutical drugs
1 Introduction
Because steroid hormones play an important role in a widerange of physiological processes the potential to disturbendocrine effects is a major concern in the development ofnovel pharmaceutical drugs such as etomidate and aminog-lutethimide [1]The adrenal gland is the most common targetfor toxicity in the endocrine system in vivo because steroidhormones are primarily synthesized through enzymatic reac-tions in the adrenal cortex [2ndash5] Indeed in these studiesbased on chemically induced endocrine lesions observed invivo themost frequent site of reported effects was the adrenal
gland Therefore the prediction of human adrenal toxicitybased on the mechanism of on- or off-target actions in theearly stages of drug development is important
The NCI-H295R human adrenocortical carcinoma cellline has been used to elucidate mechanisms of adrenalsteroidogenic disrupting compounds [1 6] The H295R cellline was established by Gazder and his collaborators in1990 [7] which expresses all key steroidogenic enzymesand steroidogenesis-related proteins [7ndash9] H295R cells havethe physiological characteristics of zonally undifferentiatedhuman fetal adrenal cells and the ability to produce steroidhormones found in the adult adrenal cortex [1 7 9] In vitro
Hindawi Publishing CorporationJournal of ToxicologyVolume 2016 Article ID 4041827 19 pageshttpdxdoiorg10115520164041827
2 Journal of Toxicology
bioassays using the H295R human cell line have been able toevaluate the effects of chemicals on steroid hormone produc-tion [10ndash15] steroidogenic enzyme activities [11 16 17] andthe expression of steroidogenic genes [11 18] In transcrip-tome studies the mechanisms of action of many steroido-genic disrupting compounds have been qualitatively assessedin terms of adrenal toxicity However gene expression doesnot always reflect the production of steroid hormones [19]Furthermore measuring a few specific steroid hormonesmay not be a useful approach to study the mechanisms ofsteroidogenic disrupting effects in complex pathways suchas adrenal steroidogenesis To systematically understand howexogenous compounds affect adrenal steroidogenesis simul-taneous determination of all detectable steroid hormones andintegrative analysis of these complex data would be impor-tant As an exploratory approach to analyze complex dataToxClust developed by Zhang and colleagues in 2009 is ableto visualize concentration-dependent response relationshipsin the characteristics of chemically induced toxicologicaleffects [20] However this exploratory approach is unable toprovide a quantitative understanding of the mechanism ofaction of adrenal toxicants or reveal systematic informationabout the effect of each enzymatic reaction interactions andfeedback in the adrenal steroidogenesis pathway
Systems biology based on computational models of bio-logical processes and the comprehensive measurement ofbiological molecules is the most powerful approach to quan-titatively understand the influence of each factor in complexbiological pathways In recent studies by our collaboratorsa computational model of adrenal steroidogenesis has beendeveloped in NCI-H295R cells including the steroidogenicdisrupting effects of metyrapone to inhibit enzymatic reac-tions of CYP11B1 [21 22]Themodel reproduces the dynamicsof adrenal steroidogenesis in NCI-H295R cells and theinfluence of metyrapone A current computational model ofadrenal steroidogenesis was incorporated with a reaction ofoxysterol synthesis as a bypass to consume cellular cholesterol[22] In addition all reactions in this model are describedby a kinetic equation of the first-order reaction [22] Itis difficult to quantitatively evaluate the influence of eachprotein in the complicated system of adrenal steroidoge-nesis using the reported models because it is simple andany biochemical and cellular biological information is notsufficient For example to clearly understand the causeof the change from the differentially dynamic patterns ofsteroid hormones it is necessary to consider the substrateinhibition of steroidogenic enzyme because most of steroido-genic enzymes recognize multiple steroids as the enzymaticsubstrate However the substrate inhibition of steroidogenicenzyme cannot be described by the mathematical modelbased on kinetic equations of first-order reaction that doesnot consider Michaelis constant 119870
119898expressing the affinity
of the substrate To quantitatively estimate the mechanismof steroidogenic disrupting compounds from comprehensiveexperimental data of adrenal steroidogenesis in NCI-H295Rcells the reportedmodel should be improved according to thefollowing two points First the kinetic equation of enzymaticreactions should be exchanged from the first-order equationto a steady-state kinetic equation based on the mechanism of
the enzymatic reaction Because a mathematical model orga-nized by first-order equations operates in a simple structure-dependent manner it does not show complex behavior basedon molecular interactions feedback or regulation Secondintracellular localization processes of cholesterol should beincorporated as a considerable mechanism Because intra-cellular cholesterol molecules are stored as cholesterol estersor widely distributed as membrane components only a fewcholesterol molecules localized on the mitochondrial innermembrane are available for the adrenal steroidogenesis path-way [23 24]Moreover cholesterol-trafficking processes fromthe outer to inner mitochondrial membranes which are reg-ulated by steroidogenic acute regulatory (StAR) protein areone of the rate-limiting steps in adrenal steroidogenesis [24]By overcoming these limitations in the reported steroido-genesis model systems analysis of adrenal steroidogenesisin H295R cells may be able to quantitatively estimate themechanismof action of steroidogenic disrupting compounds
In the present study to quantitatively estimate thetoxicological mechanism of endocrine-active compoundsin adrenal steroidogenesis and to predict human adrenaltoxicity of novel pharmaceutical drugs in the drug discov-ery phase we developed a novel computational model ofsteroidogenesis in NCI-H295R cells It includes cholesteroltransport into intracellular regions from the extracellularspace the cholesterol translocation system in intracellularregions including oxysterol synthesis themetabolic pathwayof adrenal steroidogenesis and transport of steroid hor-mones Global sensitivity analysis of this adrenal steroido-genesis model is able to evaluate the influence of eachsteroidogenic enzyme and related protein for each steroidhormone observed in an in vitro steroidogenesis assay ofNCI-H295R cells Furthermore the mechanisms of actionof steroidogenesis disrupting compounds for steroidogenicenzymes can be estimated by the optimization method tosolve the reverse problem from the concentration changesof 12 steroid hormones measured by liquid chromatogra-phymass spectrometry in the steroidogenesis assay of NCI-H295R cells in vitro Using this developed model of adrenalsteroidogenesis and the analytical approach the in vitrosteroidogenesis assay of NCI-H295R cells can assess thehuman adrenal toxicity of a novel pharmaceutical drug basedon quantitative understanding of its toxicological mechanismin adrenal steroidogenesis
2 Materials and Methods
21 The Experimental Part
211 Cell Culture NCI-H295R human adrenocortical carci-noma cells were purchased from the American Type CultureCollection (Cat CRL-2128 Manassas VA) and cultured at37∘C in a humidified atmosphere with 5 CO
2 The cells
were maintained in a 1 1 mixture of Dulbeccorsquos modifiedEaglersquos medium (DMEM GIBCO Life Technologies Carls-bad CA) and F-12 medium (MP Biomedicals Inc IrvineCA) supplemented with 15mM HEPES (Dojindo Labora-tories Kumamoto Japan) 000625mgmL insulin (Sigma-Aldrich Inc St Louis MO) 000625mgmL transferrin
Journal of Toxicology 3
(Sigma-Aldrich Inc St Louis MO) 30 nM sodium selen-ite (Wako Pure Chemical Industries Ltd Osaka Japan)125mgmLbovine serumalbumin (BSA Sigma-Aldrich IncSt Louis MO) 000535mgmL linoleic acid (Sigma-AldrichInc St Louis MO) 25 Nu Serum (Becton Dickinson andCompany Franklin Lakes NJ) 100UmL penicillin (MeijiSeika Pharma Tokyo Japan) and 100mgL streptomycin(Meiji Seika Pharma Tokyo Japan)
212 Adrenal Steroidogenesis in Human Adrenal Corticocar-cinoma NCI-H295R Cells NCI-H295R cells were stimulatedwith adrenocorticotrophic hormone (ACTH) forskolin andangiotensin II to initiate steroidogenesis Changes in steroidconcentrations over time were measured after stimulationin both cells and culture medium to construct a simulationmodel
The cells were seeded at 6 times 105 cellswell in 6-well platesAfter 3 days of culture the culture medium was changed tostimulationmediumconsisting ofDMEMF-12 (1 1)mediumsupplemented with 000625mgmL insulin 000625mgmLtransferrin 30 nM sodium selenite 125mgmL BSA000535mgmL linoleic acid 10 fetal bovine serum(GIBCO Life Technologies Carlsbad CA) 100UmL peni-cillin 100mgL streptomycin 50 nM ACTH (Sigma-AldrichInc St Louis MO) 20120583M forskolin (Sigma-Aldrich Inc StLouis MO) and 100 nM angiotensin II (Calbiochem MerckMillipore Darmstadt Germany) Culture media and cellswere collected at 0 8 24 48 and 72 h after stimulation Thecells were collected in 100120583L distilled water and sonicatedto produce a cell lysate The cultures were conducted in fourwellstime point (119873 = 4)
The concentrations of 12 steroids pregnenolone (PREG)17120572-hydroxypregnenolone (HPREG) dehydroepiandroster-one (DHEA) progesterone (PROG) 17120572-hydroxyprogester-one (HPROG) androstenedione (DIONE) testosterone(TESTO) 11-deoxycorticosterone (DCORTICO) 11-deoxy-cortisol (DCORT) corticosterone (CORTICO) cortisol(CORT) and aldosterone (ALDO) in the medium and celllysate were measured by LCMS Concentrations of estrone(E1) and 17120573-estradiol (E2) were measured by enzyme-linkedimmunosorbent assays (Wako Pure Chemical Industries LtdOsaka Japan) In addition the concentration of cholesterolwas measured using a commercial kit (Wako Pure ChemicalIndustries Ltd Osaka Japan) based on the cholesterol oxi-dase method
213 Liquid Chromatography A LC-VP series (ShimadzuKyoto Japan) consisting of an SIL-HTc autosamplerLC-10ADvp Pump CTO-10ACvp column oven and DGU-14AM degasser was used to set the reverse-phase liquid chro-matographic conditions The column was a Cadenza CD-C18 column (100 times 2mm id 3 120583m Imtakt Corp KyotoJapan) used at 45∘C The mobile phase included waterace-tonitrileformic acid 955005 (vvv Solvent A) andwateracetonitrileformic acid 3565005 (vvv Solvent B)The gradient elution programs were 0 B (0-1min with anisocratic gradient) 0ndash40 B (1-2min with a linear gradient)40 B (2ndash7min with an isocratic gradient) 40ndash100 B(7ndash12min with a linear gradient) 100 B (12ndash14min with
an isocratic gradient) 100ndash0 B (14-15min with a lineargradient) and 0 B (15-16min with an isocratic gradient) ata flow rate of 03mLmin The autosampler tray was cooledto 45∘C and the injection volume was 5120583L HPLC gradeacetonitrile and formic acid were purchased fromWAKO
214 Mass Spectrometry A triple quadrupole mass spec-trometer API4000 (Applied BiosystemsMDS Sciex Con-cord Canada) coupled with an electrospray ionization sourcewas operated in the positive ion mode The optimized ionsource conditions were as follows collision gas 6 psi curtaingas 40 psi ion source gas 1 50 psi ion source gas 2 80 psiion source voltage 5500V ion source temperature 600∘CNitrogenwas used as the collision gas in themultiple reactionmonitoring (MRM) mode The conditions of declusteringpotential collision energy and collision cell exit potentialwere optimized by every steroid The transitions in MRMwere as follows PREG 119898119911 317 rarr 299 HPREG mz 315 rarr297 DHEA mz 289rarr 271 PROG mz 315rarr 109 HPROGmz 331 rarr 109 DIONE mz 287 rarr 97 DCORT mz 331 rarr123 DCORTICOmz 347rarr 161 CORTICOmz 347rarr 100CORT mz 363 rarr 309 ALDO mz 361 rarr 343 and TESTOmz 289 rarr 109 Mass spectroscopic data were acquired andquantified using the Analyst 142 software package (AppliedBiosystemsMDS Sciex Concord Canada)
215 Estimation of the Cell Volume Cell volume was esti-mated from the number of cells in the well and the averagediameter of the cells Cells were detached from the well using0025 trypsin (MP Biomedicals Inc Irvine CA) in a 002EDTA solution (Dojindo Laboratories Kumamoto Japan)at the start of preculture start of stimulation and at 2448 and 72 h after stimulation The numbers and diametersof the cells were measured by a cell counter Vi-cell XR201 (Beckman Coulter Krefeld Germany) after trypan bluestaining Parameters of the cell volume and number of cellswere estimated to fit experimental time-course data usingexponential curves
216 Test Compounds in Validation Study NCI-H295Rcells were exposed to seven well-characterized inhibitors ofsteroidogenesis and then the concentrations of the steroidsin the culture medium were measured to estimate theenzyme inhibition to evaluate the performance of the simula-tion model The adrenal steroidogenic inhibitors includedaminoglutethimide (AGT Bachem AG Bubendorf Switzer-land) op1015840-DDD (DDD Sigma-Aldrich Inc St Louis MO)spironolactone (SP Sigma-Aldrich Inc St Louis MO)metyrapone (MP Sigma-Aldrich Inc St Louis MO) keto-conazole (KC Wako Pure Chemical Industries Ltd OsakaJapan) miconazole (MC Wako Pure Chemical IndustriesLtd Osaka Japan) and daidzein (DZ Sigma-Aldrich IncSt Louis MO) The cells were also exposed to four adrenaltoxicants whose adrenal toxicity is not mediated throughsteroidogenesis inhibition The toxicants were acryloni-trile (AN Wako Pure Chemical Industries Ltd OsakaJapan) salinomycin (SM Sigma-Aldrich Inc St LouisMO) thioguanine (TG Tokyo Chemical Industry Co LtdTokyo Japan) and fumaronitrile (FN Wako Pure Chemical
4 Journal of Toxicology
Industries Ltd Osaka Japan) All chemicals were dissolvedin DMSO (Wako Pure Chemical Industries Ltd OsakaJapan) and added to the culture medium at 1 1000 dilutions
217 Validation Study Using Adrenal Toxicants NCI-H295Rcells were cultured for 3 days in 6-well plates and thenstimulated with the above-mentioned compounds Upon thestart of stimulation various concentrations of test chemicalswere added to the cultures After a further 3 days of culturewith the chemicals the concentrations of 12 steroids (PREGHPREG DHEA PROG HPROG DIONE DCORTICODCORT CORTICO CORT ALDO and TESTO) in theculture medium were measured by LCMSMSThe test con-centrations of the chemicalswere determined by dose-findingcytotoxicity assays The cytotoxicity assay was conducted in96-well plates using ATP content in cells as an endpoint(CellTiter-Glotrade Luminescent Cell Viability Assay Promega)Concentrations that caused more than 20 cytotoxicity werenot used in the steroidogenesis assay The test concentrationsof adrenal steroidogenesis inhibitors and other compoundsare shown in Table 1
218 Statistical Analysis Comparisons were performed bythe two-sample Welchrsquos 119905-test with Bonferroni multiple test-ing correction for each steroid hormone species Statisticallysignificant steroid hormones were considered at adjusted119901 values of less than 001 Differential metabolic steroidprofiles were classified by hierarchical cluster analysis Pair-wise distances between all compounds and all steroids werecalculated by standardized Euclidean metric This distancematrix was analyzed with Wardrsquos method for hierarchicalclustering Statistical analysis was performed usingMATLABsoftware (MathWorks Inc Natick MA)
22 The Computational Part
221 Mathematical Modeling of Adrenal Steroidogenesis inNCI-H295R Cells Steroid hormones secreted from humanadrenal corticocarcinoma NCI-H295R cells are synthesizedfrom cholesterol through the C
21-steroid hormone biosyn-
thesis pathway A mathematical model of adrenal steroido-genesis in NCI-H295R cells was constructed with cholesteroltransport and the intracellular localization pathway theoxysterol synthesis pathway as a bypass of steroidogenesisthe C
21-steroid hormone biosynthesis pathway as the main
steroidogenesis pathway passive transport of steroid hor-mones and cell proliferation (Figure 1) In this model twocompartments the intracellular space and culture mediumwere incorporated as the available region Equations andparameters of the cell proliferation and diffusional trans-port of steroid hormones have been proposed by previousstudies [21 22] Cholesterol transport and the intracellularlocalization pathway including the oxysterol bypass wereintegrated using a part of the ACTH-stimulated cortisolsecretion model described by Dempsher and colleagues [47]The C
21-steroid hormone biosynthesis pathway includes 14
steroid hormones PREG HPREG DHEA PROG HPROGDIONE TESTO DCORTICO DCORT CORTICO CORTALDO E1 and E2 and 17 enzymatic reactions catalyzed by
nine steroidogenic enzymes cholesterol side chain cleavageenzyme (CYP11A1) 17120572-hydroxylase (CYP17H) C
1720-lyase
(CYP17L) 3120573-hydroxysteroid dehydrogenase (HSD3B2) 21-hydroxylase (CYP21A2) 11120573-hydroxylase (CYP11B1) 18-hydroxylase (CYP11B2) 17120573-hydroxysteroid dehydrogenase(HSD17B3) and aromatase (CYP19A1) In this mathematicalmodel of adrenal steroidogenesis inNCI-H295R cells the fluxvelocities ofmolecular transportation and enzymatic reactionrates of steroidogenic enzymes were defined based on thefirst-order reaction and rapid-equilibrium enzyme kineticsrespectively All equations in the mathematical model ofadrenal steroidogenesis of NCI-H295R cells were describedin a supplementary document (see Supplementary Materialavailable online at httpdxdoiorg10115520164041827)The rate constants and the maximum activities were esti-mated by fitting to experimental time-course data of theconcentrations of cholesterol and all steroids Initial valuesof cholesterol and the 14 steroid concentrations were usedin each experimentally measured value and every steroidconcentration was assumed to rapidly reach the equilibriumstate between the culture medium and intracellular space Allfixed values of static parameters and initial values of variableparameters in this model were described in Tables S1 and S2in a supplementary document respectively
222 Modeling and Simulation Environment This compu-tational model of adrenal steroidogenesis in NCI-H295Rcells was developed on the simBio platform which is ageneral environment of biological dynamic simulation andcomputational model development [48] ODEs were solvedby the fourth-order Runge-Kuttamethodwith a variable timestepThe time step (119889119905) was adjusted to refer to themaximumabsolute value of flux velocities or enzymatic reaction ratesat each time point and the range of the time step was from1 times 10minus5 to 10minus2 To confirm whether the range of the timestep was suitable the numerical error ratio was calculated bycertain fixed time steps in the range of the time step whichwas under 1 times 10minus8 in every time step The duration time ofcomputational simulation of adrenal steroidogenesis in NCI-H295R cells was set at 72 h
223 Parameter Optimization To reconstruct experimentaltime-course patterns of the concentrations of cholesterol andthe 14 steroids in the culture medium and intracellular spacewe optimized every rate constant and maximum velocityof the steroidogenic enzymes This parameter optimizationproblem was solved by the Levenberg-Marquardt methodwhich is one of the nonlinear least squares methods [49ndash51] The objective function of optimization was used as thefollowing normalized least squares distance (NLSD)
NLSD = sumℎ
sum
119894
sum
119895
(119883expℎ119894119895minus 119883
simℎ119894119895)
2
119883maxℎ119894
2 (1)
where ℎ is the compartment (culture medium or intracellularspace) 119894 is the molecular species (cholesterol and the 14steroids) 119895 is the time point (0 8 24 48 and 72 h) 119883exp
ℎ119894119895
is the experimentally measured concentration of molecule
Journal of Toxicology 5
Table1Ad
renaltoxicities
andactio
nsof
teste
dcompo
unds
Testchem
ical
Testconcentrations
(120583M)
Pathologicalfeatures
ofadrenaltoxicity
Repo
rted
enzymeinh
ibition
sRe
ferences
Acrylonitrile
(AN)
01110and
100
Hem
orrhagicadrenaln
ecrosis
Not
repo
rted
[25]
Salin
omycin
(SM)
000
00100101
1and10
Dam
agetoadrenalm
edullae
Not
repo
rted
[26]
Thioguanine(TG
)001110and100
Hem
orrhagicadrenaln
ecrosis
Not
repo
rted
[25]
Fumaron
itrile
(FN)
01110and
100
Hem
orrhagicadrenaln
ecrosis
Not
repo
rted
[25]
Aminoglutethim
ide(AG
T)01110and
100
Hypertro
phyvascular
degeneratio
nCY
P11A
1CY
P21A
2CY
P11B1andCY
P11B2
[627ndash30]
op1015840-D
DD(D
DD)
0111025and100
Atroph
yCY
P11A
1HSD
3B2CY
P21A
2CY
P11B1andCY
P11B2
[293132]
Spiro
nolacton
e(SP
)11050and100
Hypertro
phy
CYP17H
CYP
17LCY
P11B1andCY
P11B2
[633ndash35]
Metyrapon
e(MP)
01110and
100
Hypertro
phyvascular
degeneratio
nCY
P11A
1CY
P11B1andCY
P11B2
[62936ndash39]
Ketoconazole(KC)
01110and
100
Hypertro
phy
CYP11A
1CY
P17H
CYP
17LHSD
3B2CY
P21A
2and
CYP11B1
[62940
ndash43]
Micon
azole(MC)
011102550and
100
Hypertro
phy
CYP11A
1CY
P17H
CYP
17LCY
P21A
2andCY
P11B1
[4144
45]
Daidzein(D
Z)01110and
100
Unk
nown
HSD
3B2andCY
P21A
2[46]
6 Journal of Toxicology
CHOC
PREG
PROG
DCORTICO
CORTICO
ALDO
CORT
DCORT
HPROG
HPREG DHEA
DIONE
TESTO
CYP11A1
E1
E2
HSD3B2
CYP1
7L
CYP21A2 HSD17B3
CYP1
9A1
CYP11B2
OXYOxysterol synthesis
PREGHPREG DHEAPROG
HPROG DIONE E1DCORTICODCORT TESTO E2CORTICO
CORTALDO
NCI-H295R cell
Medium culture
CYP11B1
CHOL
StAR
CHOM
CHOSCEH
CHOR
CHOLCholesterol transport
Total intracellular cholesterol
CYP1
7H
Vloc Vacc
VMTR
VCEH
Figure 1 Schematic diagram of the mathematical model of adrenal steroidogenesis in NCI-H295R cells Overview of the mathematicalmodel of adrenal steroidogenesis in NCI-H295R cells including cholesterol transport and intracellular localization oxysterol synthesisthe C
21-steroid hormone biosynthesis pathway passive diffusional transport of steroid hormones and cell proliferation CHOL total
cholesterol in medium culture CHOS stored cholesterol esters in the endoplasmic reticulum CHOC intracellular free cholesterolCHOM mitochondrial free cholesterol CHON mitochondrial free cholesterol close to CYP11A1 enzymes CHOR mitochondrial freecholesterol remote from CYP11A1 enzymes PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosteronePROG progesterone HPROG 17120572-hydroxyprogesterone DIONE androstenedione DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICO corticosterone CORT cortisol ALDO aldosterone TESTO testosterone E1 estrone E2 17120573-estradiol OXYoxysterol CEH cholesterol ester hydrolase StAR steroidogenic acute regulatory protein CYP11A1 P450 side chain cleavage enzymeCYP17H 17120572-hydroxylase of CYP17 CYP17L C
17ndash20 lyase of CYP17 HSD3B2 3120573-hydroxysteroid dehydrogenase CYP21A2 21-hydroxylaseCYP11B1 11120573-hydroxylase CYP11B2 18-hydroxylase HSD17B3 17120573-hydroxysteroid dehydrogenase and CYP19A1 aromatase
119894 in compartment ℎ at time point 119895 119883simℎ119894119895
is the simulatedconcentration of molecule 119894 in compartment ℎ at time point119895 and 119883max
ℎ119894is the maximum concentration of molecule 119894 in
compartment ℎ over all time points Data points under thelower quantitation limit were excluded from the evaluationby the objective function
Effects of every static model parameter for parameteroptimization were calculated from differences of fitting theobjective function using sensitivity analysis
224 Quantitative Estimation of the Mechanism of Action ofAdrenal Toxicants Metabolic steroid profiling and differen-tial patterns of the adrenal steroid hormones by chemicalperturbation were reconstructed to optimize the relativeactivities of the steroidogenic enzymes The input datafor the quantitative mechanistic analysis of adrenal toxiccompounds was a fold change (ratio) of the measured 12
steroid concentrations induced by drug exposure for 72 hThe two-step optimization method of the real-coded geneticalgorithm (RCGA) was adopted as a global optimizationmethod in the quantitative mechanistic analysis of adrenaltoxic compoundsThe operations of the crossover and gener-ation alterationmodel in RCGAwere used for the real-codedensemble crossover (REX) and just generation gap (JGG)[52ndash55] As the initial parameters of RCGA maximum gen-eration population size selection size of parent individualspopulation size of child individuals and termination criteriawere 1000 100 6 25 and under 01 of NLSD respectivelyThe search space for the relative activities of the steroidogenicenzymes was from 1100 to 100 To evaluate the fitness ofeach individual the sum of squared residuals for fold changesof measured 12 steroid concentrations was used as theobjective function Nonlinear least squares optimization bythe Levenberg-Marquardt method was used as a local
Journal of Toxicology 7
search [49ndash51] As the estimated mechanisms of actionsof the adrenal toxic compounds the relative activities ofeight steroidogenic enzymes (CYP11A1 CYP17H CYP17LHSD3B2 CYP21A2 CYP11B1 CYP11B2 and HSD17B3)were optimized by the above-mentioned 2-step optimizationmethod Every optimization calculation was duplicated tocheck the numerical stability of the optimal parameters
225 Global Dynamic Sensitivity Analysis The propertyof every kinetic parameter in this computational modelof steroidogenesis in NCI-H295R cells was evaluated bydynamic sensitivity analysis The sensitivity (119878
119909119910) of kinetic
parameter 119909 for variable parameter 119910 was defined by the fol-lowing equation
119878119909119910(119905)=
Δ119910(119905)119910(119905)
Δ119909119909 (2)
where variable parameter119910was the concentration of a steroidhormone in the cytosolic space of NCI-H295R cells Theperturbation for kinetic parameters was +10 (Δ119909119909 = 01)
3 Results
31 Experimental Data on Adrenal Steroidogenesis
311 Adrenal Steroidogenesis of NCI-H295R Cells and theMass Balance All steroid hormones in the culture mediumwere significantly increased after 72 h of stimulation with50 nM ACTH 20120583M forskolin and 100 nM angiotensinII (Figure 2(a)) Mass balances in steroidogenesis of NCI-H295R cells under nontreatment and control (stimulated)conditions are shown in Figures 2(b) and 2(c) respectivelyUnder stimulation the dynamics of net mass in these exper-iments were unchanged and accumulated cholesterol wasconverted to adrenal steroids
312 Cytotoxicity of Adrenal Toxicants Viabilities of cellstreated with each compound were expressed as a relativevalue to the ATP level of the control Effects of AN SM TGFN AGT DDD SP MP KC MC and DZ on cell viabilitywere determined to be valid under 80 of the relative ATPlevel at 7 days after treatment AN SM TG and FN showedcytotoxicity at over 100 1 10 and 10120583M respectively AGTMP andDZ did not affect cell viability at up to 100 120583MDDDSP KC andMC induced less than 80 of cell viability at over100 50 100 and 25 120583M respectively
313 Differentially Steroid Profiling of Adrenal ToxicantsAfter NCI-H295R cells were exposed to each test compoundduring three days the concentrations of 12 steroid hor-mones in the culture mediumwere simultaneously measuredby LCMSMS All effects of the compounds on adrenalsteroidogenesis were evaluated at the concentration withoutany overt cytotoxicity The differential metabolic steroidprofiles of 11 adrenal toxic compounds were classified andvisualized by using hierarchical clustering analysis (Figure 3)
Four adrenal toxicants without steroidogenic inhibitionAN SM TG and FN did not change the medium concen-trations of all steroid hormones by more than 2-fold Above-mentioned 4 compounds at every condition and 7 adrenalsteroidogenic inhibitors at the low exposure concentrationwere gathered into a big cluster as nonchange group The7 steroidogenic inhibitors at the maximum exposure con-centration showed the characteristic steroid profiles eachbut 100 120583M DZ and 10 120583M SP were classified as a clusterAGT drastically decreased the medium concentrations ofPREG HPREG DHEA PROG DCORTICO CORTICOand ALDO at 100 120583M DDD dose-dependently decreased themedium concentrations of PROG DCORTICO CORTICOCORT and ALDO at gt10 120583M and decreased PREG HPREGDHEA PROG HPROG DIONE and DCORT at the maxi-mum exposure concentration of 25120583M SP increased PREGHPREG and DHEA and decreased PROG DIONE DCOR-TICODCORTCORTICOALDO andTESTOat 10120583MMPdose-dependently decreased CORTICO CORT and ALDOand decreased DHEA HPROG DIONE and TESTO at themaximum exposure concentration of 100 120583M KC drasticallydecreased the medium concentrations of PREG HPREGDHEAHPROGDIONEDCORTICODCORTCORTICOCORTO ALDO and TESTO at 10 120583M MC increased themedium concentrations of PROG and decreased DIONEDCORT CORT and TESTO at 10 120583M DZ increased PREGHPREG and DHEA and decreased DIONE DCORTICODCORT CORTICO CORT ALDO and TESTO at 100 120583M
32 The Mathematical Modeling
321 Optimization of the Mathematical Model of AdrenalSteroidogenesis in NCI-H295R Cells The mathematicalmodel of adrenal steroidogenesis in NCI-H295R cells wasoptimized for several kinetic parameters of cholesteroltransport intracellular localization the oxysterol pathwayand maximum velocity of steroidogenic enzymes to fit theexperimental time-course data All optimized kinetic para-meters are shown in Table S1 in a supplementary documentThe optimized mathematical model was reconstructedwith the experimental dynamic patterns of cholesteroland the 14 steroid hormones in the intracellular space andculture medium The fitness was 0621761 of NLSD valuesas the fitting objective function The simulation results andexperimental data are shown in Figure 4
Optimized kinetic parameters were calculated sensitivi-ties for the NLSD value as the fitting score and are shown inTable S1 in a supplementary document The highly sensitiveparameters for fitting the NLSD score were the extractednine kinetic parameters 119896Cholesterol Transport 119896
acc119891 119896loc119891 119896loc119887
119881CYP11A1max 119870CYP11A1
119898 119881CYP17HmaxA 119881HSD3B2
maxA and 119881CYP21A2maxA whichhad higher than 30 fitting sensitivity
33 The Validation Using the Adrenal Toxicants
331 Mechanistic Analysis of Adrenal Toxicants Effects ofadrenal toxic compounds on steroidogenic enzymes werequantitatively predicted from the change in the ratio ofthe measured medium concentrations of the 12 steroid
8 Journal of Toxicology
0001
001
01
1
10
PREG
HPR
EG
DH
EA
PRO
G
HPR
OG
AN
DRO
DCO
RTIC
O
DCO
RT
CORT
ICO
CORT
ALD
O
TEST
O E1 E2C
once
ntra
tion
in m
ediu
m (120583
M)
lowast
lowast
lowast
lowast
lowast
lowast
lowast lowast lowastlowast
lowast
lowast
lowast
lowast
(a)
0 8 24 48 72Time (hours)
0
40
80
120
160
200
240
Num
ber o
f mol
ecul
es (n
mol
)
(b)
0 8 24 48 72Time (hours)
0
40
80
120
160
200
240
Num
ber o
f mol
ecul
es (n
mol
)
(c)
Figure 2 Experimental data of metabolic profiling of adrenal steroid hormones and the mass balance Concentrations of steroid hormonessecreted from NCI-H295R cells in the culture medium at 72 h after stimulation were compared with the untreated condition and thestimulated condition by 50 nM ACTH 20 120583M forskolin and 100 nM angiotensin II (a) Net molecular amounts including cholesterol andsteroid hormones in the culture medium and intracellular space are plotted at five time points (0 8 24 48 and 72 h after stimulation)under the untreated condition (b) and the stimulated condition (c) In the bar graphs dark and light bars indicate the amount of cholesteroland adrenal steroids respectively All data are shown as the mean plusmn SD (119873 = 4) lowast119901 values corrected by the familywise error ratelt001 PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-hydroxyprogesteroneDIONE androstenedioneDCORTICO 11-deoxycorticosteroneDCORT 11-deoxycortisol CORTICO corticosteroneCORT cortisol ALDO aldosterone and TESTO testosterone
hormones at 72 h after drug exposure using themathematicalmodel of adrenal steroidogenesis in NCI-H295R cells Thereproducibility of the estimated results was confirmed byperforming the test twice The estimated effects of 11 adrenaltoxic compounds on eight steroidogenic enzymes are shownin Figure 5 The adrenal toxic compounds without steroido-genic inhibition such as vasculotoxic agents (AN SM TGand FN) were not estimated for the target steroidogenicenzymes under noncytotoxic conditions Every fitness valueswere under 005 of NLSD values used as the fitting objectivefunction (Figures 5(a)ndash5(d)) Other steroidogenic inhibitors(AGT DDD SPMP KCMC andDZ) are described in detailbelow
332 AGT The mechanism of action of AGT in adrenalsteroidogenesis was estimated by inhibition of CYP11A1HSD3B2 CYP21A2 and CYP11B1 at 100 120583M (estimatedinhibitions were 770 780 811 and 598 resp)(Figure 5(e)) AGT has been reported to inhibit CYP11A1
CYP21A2 CYP11B1 and CYP11B2 [6 27ndash30] Our resultswere mostly consistent with the previous reports In particu-lar CYP11A1 appeared to be inhibited strongly byAGT In ourstudy HSD3B2 inhibition of AGT was shown by mechanisticanalysis based on systems biology approaches as a novelmechanism of action of AGT Inhibition of AGT by CYP11B2was not estimated in our study However the concentrationof ALDO in the culture medium decreased to 38 ofthe normal stimulated condition Inhibition of AGT byCYP11B2 has been shown using sheep adrenal homogenatesas well as a human adrenal homogenate from a patientwith Cushingrsquos syndrome [30]The activity of 18-hydroxylaseinduced by CYP11B2 was determined as the conversion ofcorticosterone to 18-hydroxycorticosterone in the previousstudy The cause of the discrepancy regarding the effect ofAGT on CYP11B2 was suggested to be substrate inhibitionbecause the intracellular concentration of CORTICO wasincreased by over 10 times of that in the culture mediumto reach 50 120583M Another possibility was poor quantitative
Journal of Toxicology 9
log2
ratio
4
3
2
1
0
minus1
minus2
minus3
minus4
ALDO
CORTICO
CORT
TESTO
DIONE
HPROG
DCORTICO
DCORT
PROG
PREG
HPREG
DHEA
DZ(
100)
SP(1
0)SP
(1)
DZ(
10)
DD
D(0
1)
AN
(01
)A
N(1
)A
N(1
0)SM
(00
1)TG
(00
1)M
P(0
1)FN
(01
)TG
(1)
AGT(
1)AG
T(0
1)SM
(00
0001
)D
DD
(1)
SM(0
1)
AGT(
10)
MP(
1)FN
(1)
MC(
1)D
Z(1)
KC(0
1)
MC(
01)
DZ(
01)
KC( 1
)D
DD
(10)
MP(
10)
MC(
10)
MP(
100)
AGT(
100)
DD
D(2
5)KC
(10)
Figure 3 Hierarchical cluster analysis of differential metabolic profiles of 12 steroid hormones by exposure to adrenal toxicity compoundsAdrenal toxicants were classified by using the differential metabolic profiling of 12 steroid hormones Concentrations of 12 adrenalsteroids secreted from NCI-H295R cells were drastically changed by the exposure of adrenal steroidogenesis inhibitors The 12 adrenalsteroid hormones were quantitatively measured by LC-MSMS simultaneously Four adrenal vasculotoxic compounds acrylonitrile ANfumaronitrile FN salinomycin SM and thioguanine TG were used as negative control compounds for adrenal steroidogenesis inhibitorsSeven adrenal steroidogenesis inhibitors aminoglutethimide AGT op1015840-DDD DDD spironolactone SP metyrapone MP ketoconazoleKC miconazole MC and daidzein DZ showed a characteristic steroid profile each and were classified as each independent singletonat the maximum exposure condition Exposure concentrations of adrenal toxic compounds were described in brackets of the samplename and the units were prepared as 120583M A blue cluster was classified as a group of nonchange samples including negative controlcompounds and low exposure conditions of adrenal steroidogenesis inhibitors PREG pregnenolone HPREG 17120572-hydroxypregnenoloneDHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-hydroxyprogesterone DIONE androstenedione DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICO corticosterone CORT cortisol ALDO aldosterone and TESTO testosterone
reliability of the experimental data because the ALDOconcentration was under the lower limit of quantification at100 120583MAGT Hecker and colleagues reported that 3 120583MAGTdecreases PREG and PROG concentrations and increases theTESTO concentration [10] However AGT did not increasethe TESTO concentration in our study One possibility isthat the concentration of TESTO was already enhanced byabout 33-fold through stimulation with ACTH forskolinand angiotensin II
333 DDD The mechanism of action of DDD in adrenalsteroidogenesis was estimated by dose-dependent inhibitionof CYP11A1 HSD3B2 CYP21A2 and CYP11B1 (estimatedinhibitions at 25120583M were 870 869 769 and 849resp) (Figure 5(f)) DDD has been reported to inhibitCYP11A1 HSD3B2 CYP21A2 CYP11B1 and CYP11B2 [29 3132] Inhibition of DDD by CYP11B2 was not estimated in ourstudy However the concentration of ALDO in the culturemedium decreased to 3 of that in the normal stimulatedcondition Inhibition of DDD by CYP11B2 has been shownusing mitochondrial and microsomal fractions prepared bystandard centrifugation procedures from a bovine adrenal
cortex homogenate [32]The cause of the discrepancy regard-ing the inhibition ofDDDbyCYP11B2 could not be explainedby same effect in the case of AGT
334 SP Themechanism of action of SP in adrenal steroido-genesis was estimated by inhibition of HSD3B2 CYP21A2and HSD17B3 (estimated inhibitions at 10120583M were 702595 and 593 resp) (Figure 5(g)) SP has been reportedto inhibit CYP17H CYP17L CYP11B1 and CYP11B2 [6 33ndash35] The inhibitory effect of SP on the HSD3B2 enzyme wasa novel mechanism of action The main action of SP is asa mineralocorticoid receptor (MR) antagonist SP has alsobeen reported to exert some off-target effects by binding toandrogen glucocorticoid and progesterone receptors [56ndash58] SP has been shown to inhibit the production of ALDOand CORT from PREG induced by angiotensin II in H295Rcells but the specific MR antagonist eplerenone did not showthe inhibitory effects [59] Therefore HSD3B2 inhibition bySP is not mediated via MR and the action might be directinhibition of HSD3B2 enzymes or a part of known off-targeteffects mediated through other nuclear hormone receptorsRegarding CYP17H and CYP17L our results were consistent
10 Journal of Toxicology
CHOL
0
20
40
60
80
100C
once
ntra
tion
in m
ediu
m (120583
M)
10 20 30 40 50 60 70 800Time (hours)
(a)
PREG
DCORT
DCORTICO
CORTICO
CORT
Con
cent
ratio
n in
med
ium
(120583M
)
10
4
2
0
8
6
10 20 30 40 50 60 70 800Time (hours)
(b)
Con
cent
ratio
n in
med
ium
(120583M
)
0
020406
081
1214
0 10 20 30 40 50 60 70 80Time (hours)
HPREG
DIONE
PROGHPROG
E1
(c)
DHEAE2
ALDO
TESTO
0
005
01
015
02
025
Con
cent
ratio
n in
med
ium
(120583M
)
0 10 20 30Time (hours)
40 50 60 70 80
(d)
CHOL
0
5000
10000
15000
20000
25000
30000
Con
cent
ratio
n in
cell
(120583M
)
10 20 30 40 50 60 70 800Time (hours)
(e)
Con
cent
ratio
n in
cell
(120583M
)
0
50
100
150
200
250
0 10 20 30 40 50 60 70 80Time (hours)
PREG
DCORT
DCORTICO
CORTICOCORT
(f)
0 10 20 30 40 50 60 70 80Time (hours)
HPREG
DIONE
PROGHPROG
E1
Con
cent
ratio
n in
cell
(120583M
)
3025201510
35
05
(g)
005
115
2
253
35
Con
cent
ratio
n in
cell
(120583M
)
0 10 20 30 40 50 60 70 80Time (hours)
DHEA
E2ALDO
TESTO
(h)
Figure 4 Comparison of time-course profiles of cholesterol and adrenal steroids produced by NCI-H295R cells between experimentallymeasured and simulated data To intuitively confirm reconstruction of the measured experimental data in the developed simulation modelof NCI-H295R cells dynamics of cholesterol and adrenal steroids produced by NCI-H295R cells were plotted to overlay experimentaldata with the simulated results Graphs show the dynamics of medium concentrations of cholesterol (a) and adrenal steroids ((b)ndash(d))and intracellular concentrations of cholesterol (e) and adrenal steroids ((f)ndash(h)) Steroid hormones were categorized into three groups byconcentration levels Major steroids were PREG pregnenolone DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICOcorticosterone and CORT cortisol ((b) and (f)) Moderate steroids were HPREG 17120572-hydroxypregnenolone PROG progesterone HPROG17120572-hydroxyprogesterone DIONE androstenedione and E1 estrone ((c) and (g)) Minor steroids were DHEA dehydroepiandrosteroneALDO aldosterone TESTO testosterone and E2 17120573-estradiol ((d) and (h)) Experimental data are shown as symbols with dotted lines Alldata represent the mean plusmn SD (119873 = 4) Simulation data are shown as solid lines
Journal of Toxicology 11
0110
10
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
AN (120583M)
(a)
0110
FN (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(b)
00101
SM (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
100E minus 05
(c)
0110
TG (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B1
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(d)
0110
10100
AGT (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(e)
0110
1025
DDD (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(f)
1010
SP (120583M)
2750005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(g)
0110
10100
MP (120583M)
2110005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(h)
Figure 5 Continued
12 Journal of Toxicology
0110
10
KC (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(i)
0110
10
MC (120583M)
4430005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(j)
0110
10100
DZ (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(k)
Figure 5 Estimated mechanism of action of adrenal toxicants by using the mathematical model of adrenal steroidogenesis in NCI-H295Rcells Mechanisms of action of adrenal toxicity compounds were quantitatively estimated from experimental results of differential steroidprofiling using the mathematical model of adrenal steroidogenesis in NCI-H295R cells The drug action was defined as a scaling factorof enzymatic activity in the simulation model These scaling factors were optimized to fit experimental data by a hybrid optimizationmethod of the RCGA and nonlinear least squares Estimated drug actions by the exposure of vasculotoxic agents acrylonitrile AN (a)fumaronitrile FN (b) salinomycin SM (c) and thioguanine TG (d) and the steroidogenic inhibitors aminoglutethimide AGT (e) op1015840-DDD DDD (f) spironolactone SP (g) metyrapone MP (h) ketoconazole KC (i) miconazole MC (j) and daidzein DZ (k) are shownas a spider radar chart CYP11A1 P450 side chain cleavage enzyme CYP17H 17120572-hydroxylase of CYP17 CYP17L C
17ndash20 lyase of CYP17HSD3B2 3120573-hydroxysteroid dehydrogenase CYP21A2 21-hydroxylase CYP11B1 11120573-hydroxylase CYP11B2 18-hydroxylase and HSD17B317120573-hydroxysteroid dehydrogenase
with previous reports [33 34] 7120572-thiospironolactone whichis synthesized by deacetylation of SP inhibits CYP17H andCYP17L [34] The fact that there were no inhibitions ofCYP17HorCYP17L in our study suggests that SPmight not bedeacetylated to 7120572-thiospironolactone in NCI-H295R cellsRegarding CYP11B1 and CYP11B2 our results were unclearcompared with a previous study It has been shown that30 120583M SP inhibits CYP11B1 and CYP11B2 in human andbovine adrenal mitochondria [35]The cause of CYP11B1 andCYP11B2 inhibition by SP could not be determined in ourstudy which might be due to the lower maximum exposureconcentration of SP than that in the previous reportWe couldnot examine SP concentrations over 10 120583M because theseconcentrations were cytotoxic in NCI-H295R cells
335 MP The mechanism of action of MP in adrenalsteroidogenesis was estimated by dose-dependent inhibitionof CYP11B1 (estimated inhibitions at 1 10 and 100 120583M were571 827 and 982 resp) (Figure 5(h)) MP has beenreported to inhibit CYP11B1 as its major effect and CYP11A1and CYP11B2 as a weak effect [6 29 36ndash39] The results wereable to show that MP is a selective inhibitor of CYP11B1 inthe previous report However the estimated effect of MPat a high concentration 100 120583M as the maximum exposureconcentration was unclear According to the previous reportselectivity of MP for CYP11B1CYP11B2 is about five times[39] In addition 20 120583M MP has a slight inhibition effect onCYP11A1 in H295R cells [29]
336 KC The mechanism of action of KC in adrenalsteroidogenesis was estimated by inhibition of CYP11A1CYP17H CYP17L HSD3B2 CYP21A2 CYP11B1 and
CYP11B2 (estimated inhibitions at 10 120583Mwere 926 943518 830 882 974 and 798 resp) (Figure 5(i))KC has been reported to inhibit CYP11A1 CYP17H CYP17LHSD3B2 CYP21A2 and CYP11B1 [6 29 40ndash43] Our resultswere almost consistent with the previous reports KC inhibitsCYP11A1 CYP17H CYP21A2 and CYP11B1 in NCI-H295Rcells at 10 120583M [29] and CYP17H CYP17L CYP21A2 andCYP11B1 in human adrenal mitochondria and Leydig cellmicrosomes at 2ndash5120583M [60 61] However KC has shownonly weak inhibition of HSD3B2 and HSD17B3 in Leydigcells at the millimolar level [60 61] Regarding CYP11B2 andHSD17B3 we considered that these estimated inhibitions ofKC did have sufficient reliability in terms of quantitative pre-diction precision because ALDO and TESTO concentrationswere less than the lower limit of quantification at 10 120583MKC
337 MC Themechanism of action of MC in adrenal stero-idogenesis was estimated by inhibition of CYP17H CYP17LCYP11B1 and HSD17B3 (estimated inhibitions at 10120583Mwere691 530 764 and 571 resp) (Figure 5(j)) MC hasbeen reported to inhibit not only CYP17H and CYP17L butalsoCYP11A1 CYP21A2 andCYP11B1 [41 44 45]The resultsin the previous reports were able to estimate that MC isa CYP17 inhibitor However CYP11A1 inhibition by MCprobably instead of a reduction in StAR expression wasnot clearly detected in our study using NCI-H295R cellsbecause there were no decreases in the concentrations ofPREG and PROG in the culture medium Indirect inhibitionof CYP11A1 via the peripheral-type benzodiazepine receptorhas been reported in mouse adrenocortical Y-1 cells treatedwith MC in the absence of stimuli by measuring PREGproduction [44] On the other hand reductions of StAR
Journal of Toxicology 13
protein expression andor transport activity without affectingtotal steroid synthesis or CYP11A1 and HSD3B2 enzymeexpression or activities have been reported in (BU)
2cAMP-
stimulated MA-10 Leydig tumor cells treated with MC bymeasuring PROG production [45] Therefore the effect ofMC on the initial reaction in adrenal steroidogenesis fromcholesterol should be different according to the cell type andstimulation condition Inhibition of CYP21A2 and CYP11B1by MC has been reported as decreases in the consumptionof PROG and DCORTICO respectively [41] Inhibition ofCYP11B1 was estimated by the action of MC in this studybut that of CYP21A2 was not detected In the previousexperimental report inhibitory sites by MCmight have beenreflected by inhibition of CYP17H activity because CYP21A2activity was measured as a decrease in labeled PROG
338 DZ The mechanism of action of DZ in adrenalsteroidogenesis was estimated by inhibition of CYP11A1HSD3B2 CYP21A2 CYP11B1 and HSD17B3 (estimated inhi-bitions at 100 120583M were 586 941 965 872 and981 resp) (Figure 5(k)) DZ has been reported to inhibitHSD3B2 and CYP21A2 [46] The results of HSD3B2 andCYP21A2 were consistent with the previous report Howeverinhibitions have not been reported for CYP11A1 CYP11B1and HSD17B3These estimated effects of DZ on CYP11B1 andHSD17B3 were unclear because the concentrations of ALDOand TESTO were less than the lower limit of quantificationat 100 120583M In addition these enzymes act downstream of thestrong action points of DZ such as HSD3B2 and CYP21A2
34 The Simulations and the Systematic Model Analysis
341 Dynamic Sensitivity Analysis of Adrenal SteroidogenesisTo comprehensively understand the dynamics of adrenalsteroidogenesis dynamic sensitivities were calculated forsteroid concentrations secreted by NCI-H295R cells usingour constructed mathematical model of steroidogenesis Theresults of dynamic sensitivity analysis at 72 h of duration and6 h of interval time are presented as a heat-map in Figure 6
The top 10 parameters of the total area under the curveof dynamic sensitivity for cholesterol and the 14 steroids inculture medium were 119881HSD3B2
maxA 119881CYP11A1max 119881CYP21A2maxA 119896loc119891 119896loc119887
119870CYP11A1119898
119881CYP17HmaxA 119896Cholesterol Transport 119896acc119891 and 119881CYP21A2maxB in
order from the top Cholesterol uptake (119896Cholesterol Transport)StAR protein (119896loc
119891) and CYP11A1 (119881CYP11A1max ) which are
determining factors of the capacity for steroidogenesis pro-moted the production of mineralocorticoids (DCORTICOCORTICO and ALDO) and restrained the synthesis of glu-cocorticoids (DCORT and CORT) and sex steroids (DIONETESTO and E1) because of the accumulation of interme-diate molecules in steroidogenesis (PREG HPREG DHEAPROG and HPROG) only by self-activation The dynamicpatterns of the intermediate molecules in steroidogenesiswere mainly dependent on the activity of CYP17H andHSD3B2 with PREG as the substrate of these enzymes inwhich the dynamic sensitivities of 119881CYP17HmaxA for HPREG andHPROG and 119881HSD3B2
maxA for PROG HPROG and DCORTICO
reversed the direction of sensitivity at 49ndash66 h after stimu-lation The dynamic sensitivities of the maximum activitiesof HSD3B2 for PREG (119881HSD3B2
maxA ) and CYP21A2 for PROG(119881CYP21A2maxA ) were related to all steroids at 72 h Almost allmodel parameters had positive sensitivity for downstreamsteroids in the adrenal steroidogenic pathway and negativesensitivity for direct-binding steroids as substrates of thesteroidogenic enzyme The sensitivity of 119881max in all steroido-genic enzymes was relatively higher than 119870
119898for the same
steroid substrate
342 Simulation of the Metabolic Balance of AdrenalSteroidogenesis Pathway To clearly show the property of themetabolic shift between mineralocorticoid and glucocorti-coid biosynthesis we performed two-dimensional parameterscanning of the enzymatic activities of CYP17H and HSD3B2(Figure 7) NCI-H295R cells lost the ability to produce allsteroid hormones when enzymatic activities of CYP17H andHSD3B2 were changed by over 60 and 30 respectivelyActivation ofCYP17HandorHSD3B2 induced themetabolicshift that enhanced the glucocorticoid biosynthesis and devi-ated from the mineralocorticoid biosynthesis On the otherhand inhibition of CYP17H andor HSD3B2 induced themetabolic shift that enhanced the mineralocorticoid biosyn-thesis and deviated from the glucocorticoid biosynthesisMoreover the enzymatic activity of HSD3B2 regulated themetabolic balance of sex steroids and the precursors onadrenal steroidogenesis of NCI-H295R cells E1 TESTO andDIONE were produced by NCI-H295R cells when activatingthe enzymatic activity of HSD3B2 Conversely E2 andDHEAwere produced by NCI-H295R cells when suppressing theenzymatic activity of HSD3B2 The biosynthesis of down-stream steroids in adrenal steroidogenesis pathway such asmineralocorticoids and glucocorticoids was almost com-pletely terminated when the enzymatic activity of HSD3B2was decreased by over 80
4 Discussion
41 Importance of 3120573-HSDActivity in Adrenal SteroidogenesisOur systematic analysis using the mathematical model ofadrenal steroidogenesis in NCI-H295R cells revealed thatthe enzymatic activity of 3120573-HSD controls the dynamics ofadrenal steroidogenesis The activity of the StAR proteincontrols the net capacity of steroidogenesis in steroidogeniccells which is the transport of cholesterol from the outer toinner mitochondrial membranes Both the expression levelsof StAR protein and mRNA are rapidly elevated in responseto stimulation by tropic hormones such as ACTH [62 63]Another important factor in adrenal steroidogenesisis is thecholesterol side chain cleavage enzyme CYP11A1 the firstrate-limiting and hormonally regulated step in the synthesisof all steroid hormones which is conversion of cholesterolto pregnenolone in mitochondria [64] According to ourresults of global sensitivity analysis (Supplementary Table 1and Figure 6(d)) in addition to CYP11A1 and StAR proteins3120573-HSD was one of the key regulators in adrenal steroido-genesis of NCI-H295R cells And also this result suggeststhat a significant regulatory mechanism in steroidogenesis
14 Journal of Toxicology
60544830 4236 66 7218 246 120
Time (hours)
0
1
2
3C
once
ntra
tion
(a)
60544830 4236 66 7218 246 120
Time (hours)
minus05
0
05
minus1
1
Sens
itivi
ty
(b)
0
6
12
18
24
30
36
42
48
54
60
66
72
Sens
itivi
ty fo
r med
ium
conc
entr
atio
ns
of ch
oles
tero
l or s
tero
id h
orm
ones
Dur
atio
n tim
e (ho
urs)
Model parameters
+10
+05
00
minus05
minus10
(c)
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
Parameter number in the mathematical model of steroidogenesis in NCI-H295R cells
CHOL
PREG
HPREG
DHEA
PROG
HPROG
DIONE
DCORTICO
DCORT
CORTICO
CORT
ALDO
TESTO
E1
E2
Chol
este
rol o
r ste
roid
hor
mon
es
0
05
10
15
20
minus05
minus10
minus15
minus20
(d)
Figure 6 Heat-map of the global dynamic sensitivity analysis of adrenal steroid concentrations produced by NCI-H295R cells The globaldynamic sensitivity analysis is a powerful tool to comprehensively understand the dependencies of themodel parameters in themathematicalmodel of the biological complex system Dynamic sensitivities of model parameters in the mathematical model of adrenal steroidogenesisin NCI-H295R cells were calculated for all steroid concentrations in the culture medium every 6 h until 72 h after stimulation To clarify theview of heat-map of global dynamic sensitivity analysis imaginary data of the dynamics of steroid concentrations in the original model (blueline) and perturbed model for sensitivity analysis (red line) were prepared (a) Using this imaginary data the calculated dynamic sensitivities(b) and the visualized dynamic sensitivity as one block of the heat-map (c) were shown respectively By the same method that explainedusing imaginary data the large-scale data of the global dynamic sensitivity analysis on the mathematical model of adrenal steroidogenesisin NCI-H295R cells was comprehensively visualized as a big graph of heat-map (d) Parameter numbers in the horizontal axis are (1)119896Choresterol Transport (2) 119896CEH (3) 119896MTR
119891 (4) 119896MTR
119887 (5) 119896acc
119891 (6) 119896acc
119887 (7) 119896loc
119891 (8) 119896loc
119887 (9) 119896Oxysterol Synthesis (10)119870
CYP11A1119898
(11) 119881CYP11A1max (12)119870CYP17H119898A
(13) 119870CYP17H119898B (14) 119881CYP17HmaxA (15) 119881CYP17HmaxB (16) 119881CYP17HmaxB (17) 119870CYP17L
119898B (18) 119881CYP17LmaxA (19) 119881CYP17LmaxB (20) 119870HSD3B2119898A (21) 119870HSD3B2
119898B (22) 119870HSD3B2119898C (23)
119881HSD3B2maxA (24) 119881HSD3B2
maxB (25) 119881HSD3B2maxB (26) 119870CYP21A2
119898A (27) 119870CYP21A2119898B (28) 119881CYP21A2maxA (29) 119881CYP21A2maxB (30) 119870CYP11B1
119898A (31) 119870CYP11B1119898B (32) 119881CYP11B1maxA
(33) 119881CYP11B1maxB (34) 119896CYP11B2 (35) 119870HSD17B3119898A (36) 119870HSD17B3
119898B (37) 119881HSD17B3maxA (38) 119881HSD17B3
maxB (39) 119870CYP19A1119898A (40) 119870CYP19A1
119898B (41) 119881CYP19A1maxA and (42)119881
CYP19A1maxB PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-
hydroxyprogesteroneDIONE androstenedioneDCORTICO 11-deoxycorticosteroneDCORT 11-deoxycortisol CORTICO corticosteroneCORT cortisol ALDO aldosterone and TESTO testosterone
pathway is very reasonable StAR CYP11A1 and 3120573-HSD(isoforms 1 or 2 in humans) proteins generally respond to thesame hormones that stimulate steroid production throughcommon pathways such as cAMP signaling in adrenal glandsand testes [65 66] Moreover our data also support recentexperimental evidence from clinical and in vivo studiessuggesting that the enzymatic activity of 3120573-HSD plays
an important role in the regulation of mineralocorticoidsynthesis in adrenal steroidogenesis and contributes to hyper-tension caused by abnormal overproduction of aldosterone[67ndash70] Circadian clock-deficient Cry-null mice show salt-sensitive hypertension due to abnormally high synthesis ofaldosterone which is caused by constitutively high expressionof HSD3B6mRNA and protein in the adrenal cortex [67 68]
Journal of Toxicology 15
All steroid producing cells
Mineralocorticoid producing cells Glucocorticoid producing cells
Mineralocorticoid and glucocorticoid producing cells
Terminated steroidogenesis
Sex steroids and precursor sex steroidsDIONE TESTO E1DEHA E2
NCI-H295Rcells at controlcondition
Normalized activities of HSD3B2
Nor
mal
ized
activ
ities
of C
YP17
H
2
18
16
14
12
1
08
06
04
02
00 05 1 15 2
Figure 7 Metabolic categories of the steroidogenic cells determined by the balance of HSD3B2 and CYP17H activitiesThe two-dimensionalparameter scanning analysis by the perturbation of two focused parameters clarifies the interaction and the relationship between the twoparameters in the complex system Functional cellular categories of steroidogenic cells were defined by the levels of mineralocorticoid(ALDO) glucocorticoids (DCORTandCORT) and androgens (DHEAandDIONE) at 72 h after stimulation Enzymatic activities ofHSD3B2andCYP17were normalized by standard values of the simulationmodel inNCI-H295R cells Scanning ranges ofHSD3B2 andCYP17 activitieswere 0ndash200 each 10 Green regions are all 14 steroids produced by NCI-H295R cells Red blue and purple regions are mineralocorticoidproducing cells glucocorticoids-producing cells and both corticoid-producing cells respectively Yellow regions are steroidogenesis of NCI-H295R cells terminated upstream of the adrenal steroidogenesis pathway
Recent clinical observations have revealed predominantexpression of HSD3B2 mRNA and protein in tumor cellsof aldosterone-producing adenoma (APA) and HSD3B1mRNA significantly correlated with CYP11B2 mRNA levelsand plasma aldosterone concentrations in APA patients [6970] However the relationship is unclear and disputed ina small-scale clinical study indicating that genetic variationin HSD3B1 affects blood pressure and hypertension in APApatients [71] The results of our simulation study suggest that3120573-HSD protein (human genes are HSD3B1 and HSD3B2) isone of the determination factors for the dynamic propertyof adrenal steroidogenesis Our results support the clinicalevidence of Doi and colleagues [69] and we believe that theHSD3B1 enzyme has a promising potential as novel drugtarget for endocrine hypertension
42 Metabolic Shift of Adrenal Steroidogenesis and the Contri-butions of HSD3B2 and CYP17H The metabolic propertiesof adrenal steroidogenesis in NCI-H295R cells were revealedby dynamic sensitivity analysis using themathematicalmodel(Figure 6) Mineralocorticoids such as DCORTICO COR-TICO and ALDO and intermediate steroids upstream of
the adrenal steroidogenesis pathway such as PREG HPREGDHEA PROG and HPROG were accelerated by reactionsof cholesterol import (119896Cholesterol Transport) StAR protein (119896MTR
119891
and 119896loc119891) and CYP11A1 (119881CYP11A1max ) On the other hand glu-
cocorticoids such as DCORT and CORT and sex hormonessuch as DIONE TESTO and E1 were suppressed by thesemodel parametersTherefore enhancement of the net adrenalsteroidogenesis capacity which supplies PREG precursor tothe pathway causes a production shift from glucocorticoidsto mineralocorticoids by substrate inhibitions of CYP17HHSD3B2 and CYP21A2 caused by accumulation of initialintermediate steroids such as PREG and PROG Sensitivi-ties of CYP17H (119881CYP17HmaxA ) and HSD3B2 (119881HSD3B2
maxA ) for theproducts were dynamically changed and these parametersdetermined the metabolic balance of downstream steroidsin the adrenal steroidogenesis pathway According to theseresults of dynamic sensitivity analysis of StAR CYP11A1CYP17H and HSD3B2 we suggest that the enhancementof CYP17H and HSD3B2 activity during ACTH stimulationwas important to shift the steroidogenic output away fromALDO biosynthesis towards CORT biosynthesis as wellas adrenal androgen production This suggestion partially
16 Journal of Toxicology
supports a comparative animal study in which molecularand cellular variations in CYP17H activity dramatically affectacute cortisol production resulting in distinct physiologicaland behavioral responses [72]
Results of two-dimensional parameter scanning of theenzymatic activities of CYP17H and HSD3B2 quantitativelyshowed the detail of the metabolic relationship betweenmin-eralocorticoid and glucocorticoid biosynthesis (Figure 7)Particularly the results showed that the balance of theseenzymatic activities was very important for the typicalfunction of NCI-H295R cells namely the ability to produceall steroid hormones NCI-H295R cells lost this functionwhen enzymatic activities of CYP17H and HSD3B2 werechanged by over 60 and 30 respectively In addition theybecame mineralocorticoid (ALDO) secreting cells when theenzymatic activity of CYP17H or HSD3B2 was inhibited byover 50 or glucocorticoid (DCORT and CORT) secretingcells when these enzymes were activated by over 50 Inparticular this analysis also showed that HSD3B2 was a keyplayer in the adrenal steroidogenesis of NCI-H295R cellsbecause HSD3B2 inhibition by over 80 almost completelyinhibited the biosynthesis of downstream steroids The ratioof CYP17A1 to HSD3B2 mRNA expression levels has beenrelated to several endocrine diseases with a low level in APAs[73] and high level in cortisol-producing adenomas [74]Furthermore the expression levels or enzymatic activitiesof CYP17A1 and HSD3B1 have been related to androgenproduction in polycystic ovary syndrome [75 76] Theseclinical studies support our simulation results indicating thatthe balance of enzymatic activity of CYP17H and HSD3B2determines the shift in steroidogenic output to mineralocor-ticoids glucocorticoids or androgens
43 Methodologies of Quantitative Mechanistic Analysis forDrug Discovery According to our results obtained using themathematical model of steroidogenesis in NCI-H295R cellssuch as sensitivity analysis comprehensive analysis based onsystems biology is available to quantitatively estimate themechanism of action of steroidogenic disrupting compoundsfrom differential profiling of adrenal steroid hormonesbecause dynamic patterns of steroid hormones in adrenalsteroidogenesis pathway are highly complex Our proposedmethod of quantitative mechanistic analysis of steroidogenicinhibitors was able to predict known action sites in theadrenal steroidogenesis pathway at only one time point (72 hafter drug exposure) Moreover according to the resultsof sensitivity analysis (Figure 6) 119881max of all steroidogenicenzymes was more sensitive than the 119870
119898 because the
intracellular concentrations of steroid hormones were almostmaintained at sufficiently high levels compared with 119870
119898
values of steroidogenic enzymesThese results suggested thatestimation of the mechanism of action of drugs is moreeffective and detectable when using the influences of 119881maxas the searching parameters such as our proposed methodOur data showed that the proposed method based on asystems biology model is a very powerful tool for exploratoryscreening of steroidogenic disrupting compounds
RCGA as a solver of parameter estimation problemsin systems biology has been applied to biological network
identification of gene regulatory networks and metabolicpathways and optimization of biological processes usingexperimentally observed time-course data [77ndash82] In thisstudy RCGA was useful to estimate the mechanism ofaction of novel pharmaceutical drug candidates for adrenalsteroidogenesis as a new application of RCGA in systemsbiology We had two issues when applying RCGA to thequantitative mechanistic analysis of drug actions Theseissues were the vast calculation cost and multimodality ofquasi-optimum solutions in solving the optimization prob-lem because the mathematical model in systems biologyconsists of many equations and parameters A proposedoptimization strategy using RCGA based on REXJGG wasa highly stable and efficient calculation method for a betterquasi-optimum solution than the unimodal normal distri-bution crossover (UNDX)minimum generation gap (MGG)method that is well applied in the engineering field In addi-tion we expanded the RCGA optimization program basedon REXJGG to a hybrid method of GA and then applieda local search as recommended by Harada and Kobayashi[28 83] A final optimal solution was obtained with a goodconvergence property Because these problems are generalin systems biology studies we suggest that the proposedhybrid method based on REXJGG is a very useful tool forquantitative mechanistic analysis of novel pharmaceuticaldrugs not limited to steroidogenic disrupting compounds
5 Conclusions
Thenovelmathematicalmodel of adrenal steroidogenesis wasconstructed in this study including cholesterol transport anddistribution the C
21-steroid hormone pathway steroid trans-
port and cell proliferation which could reproduce adrenalsteroidogenesis in NCI-H295R cells According to the resultsof dynamic sensitivity analysis using the newmodel HSD3B2plays the most important role in the metabolic balance ofadrenal steroidogenesis in NCI-H295R cells Moreover toquantitatively estimate mechanisms of action of adrenal toxiccompounds we analyzed differential metabolic profiles of12 steroid hormones at 3 days after exposure to 11 adrenaltoxic compounds by using the new mathematical modeland a hybrid optimization method of the RCGA and a localsearch (nonlinear least squares) We could estimate whichsteroidogenic enzymes were affected by these compoundsusing the hybrid optimization method Vasculotoxic agentswere estimated to have no effect according to the resultsobtained by our method In terms of adrenal steroidogenicinhibitors the predicted action sites were approximatelymatched to the target enzymes as reported in the literatureThus our computer-aided method based on a systems biol-ogy approach may be useful to analyze the mechanism ofaction of endocrine-disrupting compounds and to assess thehuman adrenal toxicity of novel pharmaceutical drugs basedon steroid hormone profiling
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Journal of Toxicology 17
Authorsrsquo Contribution
Ryuta Saito Natsuko Terasaki and Naohisa Tsutsui designedthe study Naohisa Tsutsui and Naoya Masutomi supervisedthe experiments Natsuko Terasaki performed all in vitroexperiments and collected dataMakotoYamazaki performedall measurements of steroids Ryuta Saito conceived the insilico strategies analyzed data developed all programs andthe mathematical model and performed all simulation anal-yses with Masahiro Okamoto Ryuta Saito Natsuko TerasakiMakoto Yamazaki NaoyaMasutomi andMasahiroOkamotowrote the paper
Acknowledgments
The authors would like to thank Dr Michael S Breenand Dr Miyuki Breen for helpful discussions concerningprevious mathematical models of adrenal steroidogenesisThis work was partially supported by the Grant-in-Aid forScientific Research on Innovative Areas ldquoSynthetic Biologyrdquo(No 23119001 (MO)) from Ministry of Education CultureSports Science and Technology in Japan
References
[1] P W Harvey D J Everett and C J Springall ldquoAdrenal toxi-cology a strategy for assessment of functional toxicity to theadrenal cortex and steroidogenesisrdquo Journal of Applied Toxi-cology vol 27 no 2 pp 103ndash115 2007
[2] W E Ribelin ldquoThe effects of drugs and chemicals upon thestructure of the adrenal glandrdquo Fundamental and Applied Toxi-cology vol 4 no 1 pp 105ndash119 1984
[3] H D Colby and P A Longhurst ldquoToxicology of the adrenalglandrdquo in Endocrine Toxicology C K Atterwill and J D FlackEds pp 243ndash281 CambridgeUniversity Press Cambridge UK1992
[4] H D Colby ldquoIn vitro assessment of adrenocortical toxicityrdquoJournal of Pharmacological and Toxicological Methods vol 32no 1 pp 1ndash6 1994
[5] T J Rosol J T Yarrington J Latendresse and C C CapenldquoAdrenal gland structure function and mechanisms of toxic-ityrdquo Toxicologic Pathology vol 29 no 1 pp 41ndash48 2001
[6] P W Harvey and D J Everett ldquoThe adrenal cortex and ster-oidogenesis as cellular and molecular targets for toxicity crit-ical omissions from regulatory endocrine disrupter screeningstrategies for human healthrdquo Journal of Applied Toxicology vol23 no 2 pp 81ndash87 2003
[7] A F Gazdar H K Oie C H Shackleton et al ldquoEstablishmentand characterization of a human adrenocortical carcinoma cellline that expresses multiple pathways of steroid biosynthesisrdquoCancer Research vol 50 no 17 pp 5488ndash5496 1990
[8] W E Rainey I M Bird C Sawetawan et al ldquoRegulation ofhuman adrenal carcinoma cell (NCI-H295) production of C19steroidsrdquo The Journal of Clinical Endocrinology amp Metabolismvol 77 no 3 pp 731ndash737 1993
[9] B Staels D W Hum andW L Miller ldquoRegulation of steroido-genesis in NCI-H295 cells a cellular model of the human fetaladrenalrdquo Molecular Endocrinology vol 7 no 3 pp 423ndash4331993
[10] M Hecker J L Newsted M B Murphy et al ldquoHuman adreno-carcinoma (H295R) cells for rapid in vitro determination of
effects on steroidogenesis hormone productionrdquo Toxicologyand Applied Pharmacology vol 217 no 1 pp 114ndash124 2006
[11] A Oskarsson E Ulleras K E Plant J P Hinson and P SGoldfarb ldquoSteroidogenic gene expression in H295R cells andthe human adrenal gland adrenotoxic effects of lindane invitrordquo Journal of Applied Toxicology vol 26 no 6 pp 484ndash4922006
[12] UMuller-Vieira M Angotti and RW Hartman ldquoThe adreno-cortical tumor cell line NCI-H295R as an in vitro screeningsystem for the evaluation of CYP11B2 (aldosterone synthase)and CYP11B1 (steroid-11120573-hydroxylase) inhibitorsrdquoThe Journalof Steroid Biochemistry and Molecular Biology vol 96 no 3-4pp 259ndash270 2005
[13] K Imagawa S Okayama M Takaoka et al ldquoInhibitory effectof efonidipine on aldosterone synthesis and secretion in humanadrenocarcinoma (H295R) cellsrdquo Journal of CardiovascularPharmacology vol 47 no 1 pp 133ndash138 2006
[14] M Voets UMuller-Vieira S Marchais-Oberwinkler and RWHartmann ldquoSynthesis of amidinohydrazones and evaluation oftheir inhibitory effect towards aldosterone synthase (CYP11B2)and the formation of selected steroidsrdquo Archiv der Pharmazievol 337 no 7 pp 411ndash416 2004
[15] F K Nielsen C H Hansen J A Fey et al ldquoH295R cells as amodel for steroidogenic disruption a broader perspective usingsimultaneous chemical analysis of 7 key steroid hormonesrdquoToxicology in Vitro vol 26 no 2 pp 343ndash350 2012
[16] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cortisolproduction and on activities of steroidogenic enzymes inhuman adrenocortical H295R cellsrdquo Journal of Steroid Biochem-istry and Molecular Biology vol 80 no 3 pp 355ndash363 2002
[17] R F Canton J T Sanderson S Nijmeijer A Bergman R JLetcher and M van den Berg ldquoIn vitro effects of brominatedflame retardants and metabolites on CYP17 catalytic activity anovel mechanism of actionrdquo Toxicology and Applied Pharma-cology vol 216 no 2 pp 274ndash281 2006
[18] K Hilscherova P D Jones T Gracia et al ldquoAssessment ofthe effects of chemicals on the expression of ten steroidogenicgenes in the H295R cell line using real-time PCRrdquo ToxicologicalSciences vol 81 no 1 pp 78ndash89 2004
[19] T Gracia K Hilscherova P D Jones et al ldquoThe H295R systemfor evaluation of endocrine-disrupting effectsrdquo Ecotoxicologyand Environmental Safety vol 65 no 3 pp 293ndash305 2006
[20] X Zhang J L NewstedM Hecker E B Higley P D Jones andJ P Giesy ldquoClassification of chemicals based on concentration-dependent toxicological data using ToxClustrdquo EnvironmentalScience and Technology vol 43 no 10 pp 3926ndash3932 2009
[21] M S Breen M Breen N Terasaki M Yamazaki and R BConolly ldquoComputational model of steroidogenesis in humanH295R cells to predict biochemical response to endocrine-active chemicalsmodel development formetyraponerdquoEnviron-mental Health Perspectives vol 118 no 2 pp 265ndash272 2010
[22] M Breen M S Breen N Terasaki M Yamazaki A LLloyd and R B Conolly ldquoMechanistic computational model ofsteroidogenesis in H295R cells role of oxysterols and cell pro-liferation to improve predictability of biochemical response toendocrine active chemical-metyraponerdquo Toxicological Sciencesvol 123 no 1 pp 80ndash93 2011
[23] F B Kraemer ldquoAdrenal cholesterol utilizationrdquo Molecular andCellular Endocrinology vol 265-266 pp 42ndash45 2007
[24] W L Miller ldquoSteroidogenic acute regulatory protein (StAR)a novel mitochondrial cholesterol transporterrdquo Biochimica et
18 Journal of Toxicology
Biophysica Acta (BBA)mdashMolecular and Cell Biology of Lipidsvol 1771 no 6 pp 663ndash676 2007
[25] S Szabo and I T Lippe ldquoAdrenal gland chemically inducedstructural and functional changes in the cortexrdquo ToxicologicPathology vol 17 no 2 pp 317ndash329 1989
[26] C Chen-Pan I-J Pan Y Yamamoto T Sakogawa J Yamadaand YHayashi ldquoPrompt recovery of damaged adrenalmedullaeinduced by salinomycinrdquoToxicologic Pathology vol 27 no 5 pp563ndash572 1999
[27] R J Santen and R I Misbin ldquoAminoglutethimide review ofpharmacology and clinical userdquo Pharmacotherapy vol 1 no 2pp 95ndash120 1981
[28] M J Carella N V Dimitrov V V Gossain L Srivastava andD R Rovner ldquoAdrenal effects of low-dose aminoglutethimidewhen used alone in postmenopausal women with advancedbreast cancerrdquoMetabolism vol 43 no 6 pp 723ndash727 1994
[29] M K Johansson J T Sanderson and B-O Lund ldquoEffects of3-MeSO2-DDE and some CYP inhibitors on glucocorticoidsteroidogenesis in the H295R human adrenocortical carcinomacell linerdquo Toxicology in Vitro vol 16 no 2 pp 113ndash121 2002
[30] Y Touitou A Bogdan J C Legrand and P Desgrez ldquoAmino-glutethimide and glutethimide effects on 18-hydroxycorti-costerone biosynthesis by human and sheep adrenals in vitrordquoActa Endocrinologica vol 80 no 3 pp 517ndash526 1975
[31] M M Hart and J A Straw ldquoStudies on the site of action ofop1015840-DDD in the dog adrenal cortex 1 Inhibition of ACTH-mediated pregnenolone synthesisrdquo Steroids vol 17 no 1ndash5 pp559ndash574 1971
[32] M Ojima M Saitoh N Itoh Y Kusano S Fukuchi and HNaganuma ldquoThe effects of op1015840-DDD on adrenal steroidogen-esis and hepatic steroid metabolismrdquo Nippon Naibunpi Gakkaizasshi vol 61 no 3 pp 168ndash178 1985
[33] L F Canosa and N R Ceballos ldquoEffects of different steroid-biosynthesis inhibitors on the testicular steroidogenesis of thetoad Bufo arenarumrdquo Journal of Comparative Physiology B vol171 no 6 pp 519ndash526 2001
[34] D C Kossor S Kominami S Takemori and H D ColbyldquoRole of the steroid 17 alpha-hydroxylase in spironolactone-mediated destruction of adrenal cytochrome P-450rdquoMolecularPharmacology vol 40 no 2 pp 321ndash325 1991
[35] S C Cheng K Suzuki W Sadee and B W Harding ldquoEffectsof spironolactone canrenone and canrenoate K on cytochromeP450 and 11120573 and 18 hydroxylation in bovine and humanadrenal corticalmitochondriardquo Endocrinology vol 99 no 4 pp1097ndash1106 1976
[36] K Yanagibashi M Haniu J E Shively W H Shen and P HallldquoThe synthesis of aldosterone by the adrenal cortex Two zones(fasciculata and glomerulosa) possess one enzyme for 11120573- 18-hydroxylation and aldehyde synthesisrdquoThe Journal of BiologicalChemistry vol 261 no 8 pp 3556ndash3562 1986
[37] K Morishita H Okumura N Ito and N Takahashi ldquoPrimaryculture system of adrenocortical cells from dogs to evaluatedirect effects of chemicals on steroidogenesisrdquo Toxicology vol165 no 2-3 pp 171ndash178 2001
[38] S W J Lamberts E G Bons H A Bruining and F H de JongldquoDifferential effects of the imidazole derivatives etomidateketoconazole and miconazole and of metyrapone on the secre-tion of cortisol and its precursors by human adrenocorticalcellsrdquo Journal of Pharmacology and Experimental Therapeuticsvol 240 no 1 pp 259ndash264 1987
[39] L Yin S Lucas F Maurer U Kazmaier Q Hu and RW Hartmann ldquoNovel imidazol-1-ylmethyl substituted 1256-tetrahydropyrrolo[321-ij]quinolin-4-ones as potent and selec-tive CYP11B1 inhibitors for the treatment of Cushingrsquos syn-dromerdquo Journal ofMedicinal Chemistry vol 55 no 14 pp 6629ndash6633 2012
[40] M DiMattina N Maronian H Ashby D L Loriaux andB D Albertson ldquoKetoconazole inhibits multiple steroidogenicenzymes involved in androgen biosynthesis in the humanovaryrdquo Fertility and Sterility vol 49 no 1 pp 62ndash65 1988
[41] M Ayub and M J Levell ldquoInhibition of human adrenal steroi-dogenic enzymes in vitro by imidazole drugs including keto-conazolerdquo Journal of Steroid Biochemistry vol 32 no 4 pp 515ndash524 1989
[42] D Engelhardt M M Weber T Miksch F Abedinpour andC Jaspers ldquoThe influence of ketoconazole on human adrenalsteroidogenesis incubation studies with tissue slicesrdquo ClinicalEndocrinology vol 35 no 2 pp 163ndash168 1991
[43] S Bhasin S Sikka T Fielder et al ldquoHormonal effects of keto-conazole in vivo in the male rat mechanism of actionrdquo Endo-crinology vol 118 no 3 pp 1229ndash1232 1986
[44] D M Zisterer and D C Williams ldquoCalmidazolium and otherimidazole compounds affect steroidogenesis in Y1 cells lack ofinvolvement of the peripheral-type benzodiazepine receptorrdquoJournal of Steroid Biochemistry and Molecular Biology vol 60no 3-4 pp 189ndash195 1997
[45] L P Walsh C N Kuratko and D M Stocco ldquoEconazoleand miconazole inhibit steroidogenesis and disrupt steroido-genic acute regulatory (StAR) protein expression post-trans-criptionallyrdquo Journal of Steroid Biochemistry and Molecular Bio-logy vol 75 no 4-5 pp 229ndash236 2000
[46] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cor-tisol production and on activities of steroidogenic enzymesin human adrenocortical H295R cellsrdquo The Journal of SteroidBiochemistry and Molecular Biology vol 80 no 3 pp 355ndash3632002
[47] D P Dempsher D S Gann and R D Phair ldquoA mechanisticmodel of ACTH-stimulated cortisol secretionrdquo American Jour-nal of Physiology vol 246 no 4 part 2 pp R587ndashR596 1984
[48] N Sarai S Matsuoka and A Noma ldquosimBio a Java package forthe development of detailed cell modelsrdquo Progress in Biophysicsand Molecular Biology vol 90 no 1ndash3 pp 360ndash377 2006
[49] K Levenberg ldquoA method for the solution of certain non-linearproblems in least squaresrdquo Quarterly of Applied Mathematicsvol 2 pp 164ndash168 1944
[50] D W Marquardt ldquoAn algorithm for least-squares estimation ofnonlinear parametersrdquo SIAM Journal on Applied Mathematicsvol 11 no 2 pp 431ndash441 1963
[51] P E Gill and W Murray ldquoAlgorithms for the solution of thenonlinear least-squares problemrdquo SIAM Journal on NumericalAnalysis vol 15 no 5 pp 977ndash992 1978
[52] Y Akimoto R Hasada J Sakuma I Ono and S KobayashildquoGeneration alternation model for real-coded GA using multi-parent proposal and evaluation of just generation gap (JGG)rdquoin Proceedings of the 19th SICE Symposium on DecentralizedAutonomous Systems pp 341ndash346 Tokyo Japan January 2007
[53] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoProposal and evaluation of adaptive real-coded crossoverAREXrdquo Transactions of the Japanese Society for Artificial Intel-ligence vol 24 no 6 pp 446ndash458 2009
Journal of Toxicology 19
[54] S Kobayashi ldquoThe frontiers of real-coded genetic algorithmsrdquoJournal of Japanese Society for Artificial Intelligence vol 24 no1 pp 147ndash162 2009
[55] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoAnalysis of the behavior of MGG and JGG as a selection modelfor real-coded genetic algorithmsrdquo Transactions of the JapaneseSociety for Artificial Intelligence vol 25 no 2 pp 281ndash289 2010
[56] A Struthers H Krum and G H Williams ldquoA comparisonof the aldosterone-blocking agents eplerenone and spironolac-tonerdquo Clinical Cardiology vol 31 no 4 pp 153ndash158 2008
[57] K Yamasaki M Sawaki S Noda et al ldquoComparison of theHershberger assay and androgen receptor binding assay oftwelve chemicalsrdquo Toxicology vol 195 no 2-3 pp 177ndash1862004
[58] C Eil and S K Edelson ldquoThe use of human skin fibroblasts toobtain potency estimates of drug binding to androgen recept-orsrdquo The Journal of Clinical Endocrinology amp Metabolism vol59 no 1 pp 51ndash55 1984
[59] P Ye T Yamashita D M Pollock H Sasano and W ERainey ldquoContrasting effects of eplerenone and spironolac-tone on adrenal cell steroidogenesisrdquo Hormone and MetabolicResearch vol 41 no 1 pp 35ndash39 2009
[60] B D Albertson K L Frederick N C Maronian et al ldquoTheeffect of ketoconazole on steroidogenesis I Leydig cell enzymeactivity in vitrordquo Research Communications in Chemical Patho-logy and Pharmacology vol 61 no 1 pp 17ndash26 1988
[61] B D Albertson N C Maronian K L Frederick et al ldquoTheeffect of ketoconazole on steroidogenesis II Adrenocorticalenzyme activity in vitrordquoResearch Communications in ChemicalPathology and Pharmacology vol 61 no 1 pp 27ndash34 1988
[62] DM Stocco andB J Clark ldquoRegulation of the acute productionof steroids in steroidogenic cellsrdquo Endocrine Reviews vol 17 no3 pp 221ndash244 1996
[63] L K Christenson and J F Strauss III ldquoSteroidogenic acute regu-latory protein (StAR) and the intramitochondrial translocationof cholesterolrdquo Biochimica et Biophysica Acta (BBA)mdashMolecularand Cell Biology of Lipids vol 1529 no 1ndash3 pp 175ndash187 2000
[64] E R Simpson ldquoCholesterol side-chain cleavage cytochromeP450 and the control of steroidogenesisrdquo Molecular and Cel-lular Endocrinology vol 13 no 3 pp 213ndash227 1979
[65] H A LaVoie and S R King ldquoTranscriptional regulation of ster-oidogenic genes STARD1 CYP11A1 and HSD3Brdquo ExperimentalBiology and Medicine vol 234 no 8 pp 880ndash907 2009
[66] F Raucci A DrsquoAniello and M M Di Fiore ldquoStimulation ofandrogen production by d-aspartate through the enhancementof StAR P450scc and 3120573-HSD mRNA levels in vivo rat testisand in culture of immature rat Leydig cellsrdquo Steroids vol 84pp 103ndash110 2014
[67] M Doi Y Takahashi R Komatsu et al ldquoSalt-sensitive hyper-tension in circadian clock-deficient Cry-null mice involves dys-regulated adrenal Hsd3b6rdquo Nature Medicine vol 16 no 1 pp67ndash74 2010
[68] H Okamura M Doi Y Yamaguchi and J-M Fustin ldquoHyper-tension due to loss of clock novel insight from the molecu-lar analysis of Cry1Cry2-deleted micerdquo Current HypertensionReports vol 13 no 2 pp 103ndash108 2011
[69] M Doi F Satoh T Maekawa et al ldquoIsoform-specific mono-clonal antibodies against 3120573-hydroxysteroid dehydrogenaseisomerase family provide markers for subclassification ofhuman primary aldosteronismrdquo The Journal of Clinical Endo-crinology amp Metabolism vol 99 no 2 pp E257ndashE262 2014
[70] S Konosu-Fukaya Y Nakamura F Satoh et al ldquo3120573-hydroxy-steroid dehydrogenase isoforms in human aldosterone-produc-ing adenomardquo Molecular and Cellular Endocrinology vol 408pp 205ndash212 2015
[71] G C Verwoert J Hofland N Amin et al ldquoExpression and genevariation studies deny association of humanHSD3B1 gene withaldosterone production or blood pressurerdquoAmerican Journal ofHypertension vol 28 no 1 pp 113ndash120 2015
[72] D Hough K Storbeck SW P Cloete A C Swart and P SwartldquoRelative contribution of P450c17 towards the acute cortisolresponse lessons from sheep and goatsrdquoMolecular and CellularEndocrinology vol 408 pp 107ndash113 2015
[73] I Sakuma S Suematsu Y Matsuzawa et al ldquoCharacterizationof steroidogenic enzyme expression in aldosterone-producingadenoma a comparison with various human adrenal tumorsrdquoEndocrine Journal vol 60 no 3 pp 329ndash336 2013
[74] A Tong A Jia S Yan Y Zhang Y Xie and G Liu ldquoEctopiccortisol-producing adrenocortical adenoma in the renal hilumhistopathological features and steroidogenic enzyme profilerdquoInternational Journal of Clinical and Experimental Pathologyvol 7 no 7 pp 4415ndash4421 2014
[75] A Hirsch D Hahn P Kempna et al ldquoMetformin inhibitshuman androgen production by regulating steroidogenicenzymes HSD3B2 and CYP17A1 and complex I activity of therespiratory chainrdquo Endocrinology vol 153 no 9 pp 4354ndash43662012
[76] B Xu L Gao Y Cui et al ldquoSET protein up-regulated test-osterone production in the cultured preantral folliclesrdquo Repro-ductive Biology and Endocrinology vol 11 article 9 2013
[77] S Kikuchi D Tominaga M Arita M Takahashi and MTomita ldquoDynamic modeling of genetic networks using geneticalgorithm and S-systemrdquo Bioinformatics vol 19 no 5 pp 643ndash650 2003
[78] N Shikata YMaki M Nakatsui et al ldquoDetermining importantregulatory relations of amino acids from dynamic networkanalysis of plasma amino acidsrdquo Amino Acids vol 38 no 1 pp179ndash187 2010
[79] M Nakatsui T Ueda Y Maki I Ono and M OkamotoldquoMethod for inferring and extracting reliable genetic interac-tions from time-series profile of gene expressionrdquoMathematicalBiosciences vol 215 no 1 pp 105ndash114 2008
[80] A Komori Y Maki M Nakatsui I Ono and M OkamotoldquoEfficient numerical optimization algorithm based on new real-coded genetic algorithm AREX + JGG and application to theinverse problem in systems biologyrdquo Applied Mathematics vol3 no 10 pp 1463ndash1470 2012
[81] J A Roubos G van Straten and A J B van Boxtel ldquoAn evolu-tionary strategy for fed-batch bioreactor optimization conceptsand performancerdquo Journal of Biotechnology vol 67 no 2-3 pp173ndash187 1999
[82] D Sarkar and J M Modak ldquoGenetic algorithms with filters foroptimal control problems in fed-batch bioreactorsrdquo Bioprocessand Biosystems Engineering vol 26 no 5 pp 295ndash306 2004
[83] K Harada K Ikeda J Sakuma I Ono and S KobayashildquoHybridization of genetic algorithm with local search in mul-tiobjective function optimization recommendation of GA thenLSrdquoTransactions of the Japanese Society for Artificial Intelligencevol 21 no 6 pp 482ndash492 2006
Submit your manuscripts athttpwwwhindawicom
PainResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
ToxinsJournal of
VaccinesJournal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AntibioticsInternational Journal of
ToxicologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
StrokeResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Drug DeliveryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Pharmacological Sciences
Tropical MedicineJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AddictionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Emergency Medicine InternationalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Autoimmune Diseases
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anesthesiology Research and Practice
ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Pharmaceutics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
MEDIATORSINFLAMMATION
of
2 Journal of Toxicology
bioassays using the H295R human cell line have been able toevaluate the effects of chemicals on steroid hormone produc-tion [10ndash15] steroidogenic enzyme activities [11 16 17] andthe expression of steroidogenic genes [11 18] In transcrip-tome studies the mechanisms of action of many steroido-genic disrupting compounds have been qualitatively assessedin terms of adrenal toxicity However gene expression doesnot always reflect the production of steroid hormones [19]Furthermore measuring a few specific steroid hormonesmay not be a useful approach to study the mechanisms ofsteroidogenic disrupting effects in complex pathways suchas adrenal steroidogenesis To systematically understand howexogenous compounds affect adrenal steroidogenesis simul-taneous determination of all detectable steroid hormones andintegrative analysis of these complex data would be impor-tant As an exploratory approach to analyze complex dataToxClust developed by Zhang and colleagues in 2009 is ableto visualize concentration-dependent response relationshipsin the characteristics of chemically induced toxicologicaleffects [20] However this exploratory approach is unable toprovide a quantitative understanding of the mechanism ofaction of adrenal toxicants or reveal systematic informationabout the effect of each enzymatic reaction interactions andfeedback in the adrenal steroidogenesis pathway
Systems biology based on computational models of bio-logical processes and the comprehensive measurement ofbiological molecules is the most powerful approach to quan-titatively understand the influence of each factor in complexbiological pathways In recent studies by our collaboratorsa computational model of adrenal steroidogenesis has beendeveloped in NCI-H295R cells including the steroidogenicdisrupting effects of metyrapone to inhibit enzymatic reac-tions of CYP11B1 [21 22]Themodel reproduces the dynamicsof adrenal steroidogenesis in NCI-H295R cells and theinfluence of metyrapone A current computational model ofadrenal steroidogenesis was incorporated with a reaction ofoxysterol synthesis as a bypass to consume cellular cholesterol[22] In addition all reactions in this model are describedby a kinetic equation of the first-order reaction [22] Itis difficult to quantitatively evaluate the influence of eachprotein in the complicated system of adrenal steroidoge-nesis using the reported models because it is simple andany biochemical and cellular biological information is notsufficient For example to clearly understand the causeof the change from the differentially dynamic patterns ofsteroid hormones it is necessary to consider the substrateinhibition of steroidogenic enzyme because most of steroido-genic enzymes recognize multiple steroids as the enzymaticsubstrate However the substrate inhibition of steroidogenicenzyme cannot be described by the mathematical modelbased on kinetic equations of first-order reaction that doesnot consider Michaelis constant 119870
119898expressing the affinity
of the substrate To quantitatively estimate the mechanismof steroidogenic disrupting compounds from comprehensiveexperimental data of adrenal steroidogenesis in NCI-H295Rcells the reportedmodel should be improved according to thefollowing two points First the kinetic equation of enzymaticreactions should be exchanged from the first-order equationto a steady-state kinetic equation based on the mechanism of
the enzymatic reaction Because a mathematical model orga-nized by first-order equations operates in a simple structure-dependent manner it does not show complex behavior basedon molecular interactions feedback or regulation Secondintracellular localization processes of cholesterol should beincorporated as a considerable mechanism Because intra-cellular cholesterol molecules are stored as cholesterol estersor widely distributed as membrane components only a fewcholesterol molecules localized on the mitochondrial innermembrane are available for the adrenal steroidogenesis path-way [23 24]Moreover cholesterol-trafficking processes fromthe outer to inner mitochondrial membranes which are reg-ulated by steroidogenic acute regulatory (StAR) protein areone of the rate-limiting steps in adrenal steroidogenesis [24]By overcoming these limitations in the reported steroido-genesis model systems analysis of adrenal steroidogenesisin H295R cells may be able to quantitatively estimate themechanismof action of steroidogenic disrupting compounds
In the present study to quantitatively estimate thetoxicological mechanism of endocrine-active compoundsin adrenal steroidogenesis and to predict human adrenaltoxicity of novel pharmaceutical drugs in the drug discov-ery phase we developed a novel computational model ofsteroidogenesis in NCI-H295R cells It includes cholesteroltransport into intracellular regions from the extracellularspace the cholesterol translocation system in intracellularregions including oxysterol synthesis themetabolic pathwayof adrenal steroidogenesis and transport of steroid hor-mones Global sensitivity analysis of this adrenal steroido-genesis model is able to evaluate the influence of eachsteroidogenic enzyme and related protein for each steroidhormone observed in an in vitro steroidogenesis assay ofNCI-H295R cells Furthermore the mechanisms of actionof steroidogenesis disrupting compounds for steroidogenicenzymes can be estimated by the optimization method tosolve the reverse problem from the concentration changesof 12 steroid hormones measured by liquid chromatogra-phymass spectrometry in the steroidogenesis assay of NCI-H295R cells in vitro Using this developed model of adrenalsteroidogenesis and the analytical approach the in vitrosteroidogenesis assay of NCI-H295R cells can assess thehuman adrenal toxicity of a novel pharmaceutical drug basedon quantitative understanding of its toxicological mechanismin adrenal steroidogenesis
2 Materials and Methods
21 The Experimental Part
211 Cell Culture NCI-H295R human adrenocortical carci-noma cells were purchased from the American Type CultureCollection (Cat CRL-2128 Manassas VA) and cultured at37∘C in a humidified atmosphere with 5 CO
2 The cells
were maintained in a 1 1 mixture of Dulbeccorsquos modifiedEaglersquos medium (DMEM GIBCO Life Technologies Carls-bad CA) and F-12 medium (MP Biomedicals Inc IrvineCA) supplemented with 15mM HEPES (Dojindo Labora-tories Kumamoto Japan) 000625mgmL insulin (Sigma-Aldrich Inc St Louis MO) 000625mgmL transferrin
Journal of Toxicology 3
(Sigma-Aldrich Inc St Louis MO) 30 nM sodium selen-ite (Wako Pure Chemical Industries Ltd Osaka Japan)125mgmLbovine serumalbumin (BSA Sigma-Aldrich IncSt Louis MO) 000535mgmL linoleic acid (Sigma-AldrichInc St Louis MO) 25 Nu Serum (Becton Dickinson andCompany Franklin Lakes NJ) 100UmL penicillin (MeijiSeika Pharma Tokyo Japan) and 100mgL streptomycin(Meiji Seika Pharma Tokyo Japan)
212 Adrenal Steroidogenesis in Human Adrenal Corticocar-cinoma NCI-H295R Cells NCI-H295R cells were stimulatedwith adrenocorticotrophic hormone (ACTH) forskolin andangiotensin II to initiate steroidogenesis Changes in steroidconcentrations over time were measured after stimulationin both cells and culture medium to construct a simulationmodel
The cells were seeded at 6 times 105 cellswell in 6-well platesAfter 3 days of culture the culture medium was changed tostimulationmediumconsisting ofDMEMF-12 (1 1)mediumsupplemented with 000625mgmL insulin 000625mgmLtransferrin 30 nM sodium selenite 125mgmL BSA000535mgmL linoleic acid 10 fetal bovine serum(GIBCO Life Technologies Carlsbad CA) 100UmL peni-cillin 100mgL streptomycin 50 nM ACTH (Sigma-AldrichInc St Louis MO) 20120583M forskolin (Sigma-Aldrich Inc StLouis MO) and 100 nM angiotensin II (Calbiochem MerckMillipore Darmstadt Germany) Culture media and cellswere collected at 0 8 24 48 and 72 h after stimulation Thecells were collected in 100120583L distilled water and sonicatedto produce a cell lysate The cultures were conducted in fourwellstime point (119873 = 4)
The concentrations of 12 steroids pregnenolone (PREG)17120572-hydroxypregnenolone (HPREG) dehydroepiandroster-one (DHEA) progesterone (PROG) 17120572-hydroxyprogester-one (HPROG) androstenedione (DIONE) testosterone(TESTO) 11-deoxycorticosterone (DCORTICO) 11-deoxy-cortisol (DCORT) corticosterone (CORTICO) cortisol(CORT) and aldosterone (ALDO) in the medium and celllysate were measured by LCMS Concentrations of estrone(E1) and 17120573-estradiol (E2) were measured by enzyme-linkedimmunosorbent assays (Wako Pure Chemical Industries LtdOsaka Japan) In addition the concentration of cholesterolwas measured using a commercial kit (Wako Pure ChemicalIndustries Ltd Osaka Japan) based on the cholesterol oxi-dase method
213 Liquid Chromatography A LC-VP series (ShimadzuKyoto Japan) consisting of an SIL-HTc autosamplerLC-10ADvp Pump CTO-10ACvp column oven and DGU-14AM degasser was used to set the reverse-phase liquid chro-matographic conditions The column was a Cadenza CD-C18 column (100 times 2mm id 3 120583m Imtakt Corp KyotoJapan) used at 45∘C The mobile phase included waterace-tonitrileformic acid 955005 (vvv Solvent A) andwateracetonitrileformic acid 3565005 (vvv Solvent B)The gradient elution programs were 0 B (0-1min with anisocratic gradient) 0ndash40 B (1-2min with a linear gradient)40 B (2ndash7min with an isocratic gradient) 40ndash100 B(7ndash12min with a linear gradient) 100 B (12ndash14min with
an isocratic gradient) 100ndash0 B (14-15min with a lineargradient) and 0 B (15-16min with an isocratic gradient) ata flow rate of 03mLmin The autosampler tray was cooledto 45∘C and the injection volume was 5120583L HPLC gradeacetonitrile and formic acid were purchased fromWAKO
214 Mass Spectrometry A triple quadrupole mass spec-trometer API4000 (Applied BiosystemsMDS Sciex Con-cord Canada) coupled with an electrospray ionization sourcewas operated in the positive ion mode The optimized ionsource conditions were as follows collision gas 6 psi curtaingas 40 psi ion source gas 1 50 psi ion source gas 2 80 psiion source voltage 5500V ion source temperature 600∘CNitrogenwas used as the collision gas in themultiple reactionmonitoring (MRM) mode The conditions of declusteringpotential collision energy and collision cell exit potentialwere optimized by every steroid The transitions in MRMwere as follows PREG 119898119911 317 rarr 299 HPREG mz 315 rarr297 DHEA mz 289rarr 271 PROG mz 315rarr 109 HPROGmz 331 rarr 109 DIONE mz 287 rarr 97 DCORT mz 331 rarr123 DCORTICOmz 347rarr 161 CORTICOmz 347rarr 100CORT mz 363 rarr 309 ALDO mz 361 rarr 343 and TESTOmz 289 rarr 109 Mass spectroscopic data were acquired andquantified using the Analyst 142 software package (AppliedBiosystemsMDS Sciex Concord Canada)
215 Estimation of the Cell Volume Cell volume was esti-mated from the number of cells in the well and the averagediameter of the cells Cells were detached from the well using0025 trypsin (MP Biomedicals Inc Irvine CA) in a 002EDTA solution (Dojindo Laboratories Kumamoto Japan)at the start of preculture start of stimulation and at 2448 and 72 h after stimulation The numbers and diametersof the cells were measured by a cell counter Vi-cell XR201 (Beckman Coulter Krefeld Germany) after trypan bluestaining Parameters of the cell volume and number of cellswere estimated to fit experimental time-course data usingexponential curves
216 Test Compounds in Validation Study NCI-H295Rcells were exposed to seven well-characterized inhibitors ofsteroidogenesis and then the concentrations of the steroidsin the culture medium were measured to estimate theenzyme inhibition to evaluate the performance of the simula-tion model The adrenal steroidogenic inhibitors includedaminoglutethimide (AGT Bachem AG Bubendorf Switzer-land) op1015840-DDD (DDD Sigma-Aldrich Inc St Louis MO)spironolactone (SP Sigma-Aldrich Inc St Louis MO)metyrapone (MP Sigma-Aldrich Inc St Louis MO) keto-conazole (KC Wako Pure Chemical Industries Ltd OsakaJapan) miconazole (MC Wako Pure Chemical IndustriesLtd Osaka Japan) and daidzein (DZ Sigma-Aldrich IncSt Louis MO) The cells were also exposed to four adrenaltoxicants whose adrenal toxicity is not mediated throughsteroidogenesis inhibition The toxicants were acryloni-trile (AN Wako Pure Chemical Industries Ltd OsakaJapan) salinomycin (SM Sigma-Aldrich Inc St LouisMO) thioguanine (TG Tokyo Chemical Industry Co LtdTokyo Japan) and fumaronitrile (FN Wako Pure Chemical
4 Journal of Toxicology
Industries Ltd Osaka Japan) All chemicals were dissolvedin DMSO (Wako Pure Chemical Industries Ltd OsakaJapan) and added to the culture medium at 1 1000 dilutions
217 Validation Study Using Adrenal Toxicants NCI-H295Rcells were cultured for 3 days in 6-well plates and thenstimulated with the above-mentioned compounds Upon thestart of stimulation various concentrations of test chemicalswere added to the cultures After a further 3 days of culturewith the chemicals the concentrations of 12 steroids (PREGHPREG DHEA PROG HPROG DIONE DCORTICODCORT CORTICO CORT ALDO and TESTO) in theculture medium were measured by LCMSMSThe test con-centrations of the chemicalswere determined by dose-findingcytotoxicity assays The cytotoxicity assay was conducted in96-well plates using ATP content in cells as an endpoint(CellTiter-Glotrade Luminescent Cell Viability Assay Promega)Concentrations that caused more than 20 cytotoxicity werenot used in the steroidogenesis assay The test concentrationsof adrenal steroidogenesis inhibitors and other compoundsare shown in Table 1
218 Statistical Analysis Comparisons were performed bythe two-sample Welchrsquos 119905-test with Bonferroni multiple test-ing correction for each steroid hormone species Statisticallysignificant steroid hormones were considered at adjusted119901 values of less than 001 Differential metabolic steroidprofiles were classified by hierarchical cluster analysis Pair-wise distances between all compounds and all steroids werecalculated by standardized Euclidean metric This distancematrix was analyzed with Wardrsquos method for hierarchicalclustering Statistical analysis was performed usingMATLABsoftware (MathWorks Inc Natick MA)
22 The Computational Part
221 Mathematical Modeling of Adrenal Steroidogenesis inNCI-H295R Cells Steroid hormones secreted from humanadrenal corticocarcinoma NCI-H295R cells are synthesizedfrom cholesterol through the C
21-steroid hormone biosyn-
thesis pathway A mathematical model of adrenal steroido-genesis in NCI-H295R cells was constructed with cholesteroltransport and the intracellular localization pathway theoxysterol synthesis pathway as a bypass of steroidogenesisthe C
21-steroid hormone biosynthesis pathway as the main
steroidogenesis pathway passive transport of steroid hor-mones and cell proliferation (Figure 1) In this model twocompartments the intracellular space and culture mediumwere incorporated as the available region Equations andparameters of the cell proliferation and diffusional trans-port of steroid hormones have been proposed by previousstudies [21 22] Cholesterol transport and the intracellularlocalization pathway including the oxysterol bypass wereintegrated using a part of the ACTH-stimulated cortisolsecretion model described by Dempsher and colleagues [47]The C
21-steroid hormone biosynthesis pathway includes 14
steroid hormones PREG HPREG DHEA PROG HPROGDIONE TESTO DCORTICO DCORT CORTICO CORTALDO E1 and E2 and 17 enzymatic reactions catalyzed by
nine steroidogenic enzymes cholesterol side chain cleavageenzyme (CYP11A1) 17120572-hydroxylase (CYP17H) C
1720-lyase
(CYP17L) 3120573-hydroxysteroid dehydrogenase (HSD3B2) 21-hydroxylase (CYP21A2) 11120573-hydroxylase (CYP11B1) 18-hydroxylase (CYP11B2) 17120573-hydroxysteroid dehydrogenase(HSD17B3) and aromatase (CYP19A1) In this mathematicalmodel of adrenal steroidogenesis inNCI-H295R cells the fluxvelocities ofmolecular transportation and enzymatic reactionrates of steroidogenic enzymes were defined based on thefirst-order reaction and rapid-equilibrium enzyme kineticsrespectively All equations in the mathematical model ofadrenal steroidogenesis of NCI-H295R cells were describedin a supplementary document (see Supplementary Materialavailable online at httpdxdoiorg10115520164041827)The rate constants and the maximum activities were esti-mated by fitting to experimental time-course data of theconcentrations of cholesterol and all steroids Initial valuesof cholesterol and the 14 steroid concentrations were usedin each experimentally measured value and every steroidconcentration was assumed to rapidly reach the equilibriumstate between the culture medium and intracellular space Allfixed values of static parameters and initial values of variableparameters in this model were described in Tables S1 and S2in a supplementary document respectively
222 Modeling and Simulation Environment This compu-tational model of adrenal steroidogenesis in NCI-H295Rcells was developed on the simBio platform which is ageneral environment of biological dynamic simulation andcomputational model development [48] ODEs were solvedby the fourth-order Runge-Kuttamethodwith a variable timestepThe time step (119889119905) was adjusted to refer to themaximumabsolute value of flux velocities or enzymatic reaction ratesat each time point and the range of the time step was from1 times 10minus5 to 10minus2 To confirm whether the range of the timestep was suitable the numerical error ratio was calculated bycertain fixed time steps in the range of the time step whichwas under 1 times 10minus8 in every time step The duration time ofcomputational simulation of adrenal steroidogenesis in NCI-H295R cells was set at 72 h
223 Parameter Optimization To reconstruct experimentaltime-course patterns of the concentrations of cholesterol andthe 14 steroids in the culture medium and intracellular spacewe optimized every rate constant and maximum velocityof the steroidogenic enzymes This parameter optimizationproblem was solved by the Levenberg-Marquardt methodwhich is one of the nonlinear least squares methods [49ndash51] The objective function of optimization was used as thefollowing normalized least squares distance (NLSD)
NLSD = sumℎ
sum
119894
sum
119895
(119883expℎ119894119895minus 119883
simℎ119894119895)
2
119883maxℎ119894
2 (1)
where ℎ is the compartment (culture medium or intracellularspace) 119894 is the molecular species (cholesterol and the 14steroids) 119895 is the time point (0 8 24 48 and 72 h) 119883exp
ℎ119894119895
is the experimentally measured concentration of molecule
Journal of Toxicology 5
Table1Ad
renaltoxicities
andactio
nsof
teste
dcompo
unds
Testchem
ical
Testconcentrations
(120583M)
Pathologicalfeatures
ofadrenaltoxicity
Repo
rted
enzymeinh
ibition
sRe
ferences
Acrylonitrile
(AN)
01110and
100
Hem
orrhagicadrenaln
ecrosis
Not
repo
rted
[25]
Salin
omycin
(SM)
000
00100101
1and10
Dam
agetoadrenalm
edullae
Not
repo
rted
[26]
Thioguanine(TG
)001110and100
Hem
orrhagicadrenaln
ecrosis
Not
repo
rted
[25]
Fumaron
itrile
(FN)
01110and
100
Hem
orrhagicadrenaln
ecrosis
Not
repo
rted
[25]
Aminoglutethim
ide(AG
T)01110and
100
Hypertro
phyvascular
degeneratio
nCY
P11A
1CY
P21A
2CY
P11B1andCY
P11B2
[627ndash30]
op1015840-D
DD(D
DD)
0111025and100
Atroph
yCY
P11A
1HSD
3B2CY
P21A
2CY
P11B1andCY
P11B2
[293132]
Spiro
nolacton
e(SP
)11050and100
Hypertro
phy
CYP17H
CYP
17LCY
P11B1andCY
P11B2
[633ndash35]
Metyrapon
e(MP)
01110and
100
Hypertro
phyvascular
degeneratio
nCY
P11A
1CY
P11B1andCY
P11B2
[62936ndash39]
Ketoconazole(KC)
01110and
100
Hypertro
phy
CYP11A
1CY
P17H
CYP
17LHSD
3B2CY
P21A
2and
CYP11B1
[62940
ndash43]
Micon
azole(MC)
011102550and
100
Hypertro
phy
CYP11A
1CY
P17H
CYP
17LCY
P21A
2andCY
P11B1
[4144
45]
Daidzein(D
Z)01110and
100
Unk
nown
HSD
3B2andCY
P21A
2[46]
6 Journal of Toxicology
CHOC
PREG
PROG
DCORTICO
CORTICO
ALDO
CORT
DCORT
HPROG
HPREG DHEA
DIONE
TESTO
CYP11A1
E1
E2
HSD3B2
CYP1
7L
CYP21A2 HSD17B3
CYP1
9A1
CYP11B2
OXYOxysterol synthesis
PREGHPREG DHEAPROG
HPROG DIONE E1DCORTICODCORT TESTO E2CORTICO
CORTALDO
NCI-H295R cell
Medium culture
CYP11B1
CHOL
StAR
CHOM
CHOSCEH
CHOR
CHOLCholesterol transport
Total intracellular cholesterol
CYP1
7H
Vloc Vacc
VMTR
VCEH
Figure 1 Schematic diagram of the mathematical model of adrenal steroidogenesis in NCI-H295R cells Overview of the mathematicalmodel of adrenal steroidogenesis in NCI-H295R cells including cholesterol transport and intracellular localization oxysterol synthesisthe C
21-steroid hormone biosynthesis pathway passive diffusional transport of steroid hormones and cell proliferation CHOL total
cholesterol in medium culture CHOS stored cholesterol esters in the endoplasmic reticulum CHOC intracellular free cholesterolCHOM mitochondrial free cholesterol CHON mitochondrial free cholesterol close to CYP11A1 enzymes CHOR mitochondrial freecholesterol remote from CYP11A1 enzymes PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosteronePROG progesterone HPROG 17120572-hydroxyprogesterone DIONE androstenedione DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICO corticosterone CORT cortisol ALDO aldosterone TESTO testosterone E1 estrone E2 17120573-estradiol OXYoxysterol CEH cholesterol ester hydrolase StAR steroidogenic acute regulatory protein CYP11A1 P450 side chain cleavage enzymeCYP17H 17120572-hydroxylase of CYP17 CYP17L C
17ndash20 lyase of CYP17 HSD3B2 3120573-hydroxysteroid dehydrogenase CYP21A2 21-hydroxylaseCYP11B1 11120573-hydroxylase CYP11B2 18-hydroxylase HSD17B3 17120573-hydroxysteroid dehydrogenase and CYP19A1 aromatase
119894 in compartment ℎ at time point 119895 119883simℎ119894119895
is the simulatedconcentration of molecule 119894 in compartment ℎ at time point119895 and 119883max
ℎ119894is the maximum concentration of molecule 119894 in
compartment ℎ over all time points Data points under thelower quantitation limit were excluded from the evaluationby the objective function
Effects of every static model parameter for parameteroptimization were calculated from differences of fitting theobjective function using sensitivity analysis
224 Quantitative Estimation of the Mechanism of Action ofAdrenal Toxicants Metabolic steroid profiling and differen-tial patterns of the adrenal steroid hormones by chemicalperturbation were reconstructed to optimize the relativeactivities of the steroidogenic enzymes The input datafor the quantitative mechanistic analysis of adrenal toxiccompounds was a fold change (ratio) of the measured 12
steroid concentrations induced by drug exposure for 72 hThe two-step optimization method of the real-coded geneticalgorithm (RCGA) was adopted as a global optimizationmethod in the quantitative mechanistic analysis of adrenaltoxic compoundsThe operations of the crossover and gener-ation alterationmodel in RCGAwere used for the real-codedensemble crossover (REX) and just generation gap (JGG)[52ndash55] As the initial parameters of RCGA maximum gen-eration population size selection size of parent individualspopulation size of child individuals and termination criteriawere 1000 100 6 25 and under 01 of NLSD respectivelyThe search space for the relative activities of the steroidogenicenzymes was from 1100 to 100 To evaluate the fitness ofeach individual the sum of squared residuals for fold changesof measured 12 steroid concentrations was used as theobjective function Nonlinear least squares optimization bythe Levenberg-Marquardt method was used as a local
Journal of Toxicology 7
search [49ndash51] As the estimated mechanisms of actionsof the adrenal toxic compounds the relative activities ofeight steroidogenic enzymes (CYP11A1 CYP17H CYP17LHSD3B2 CYP21A2 CYP11B1 CYP11B2 and HSD17B3)were optimized by the above-mentioned 2-step optimizationmethod Every optimization calculation was duplicated tocheck the numerical stability of the optimal parameters
225 Global Dynamic Sensitivity Analysis The propertyof every kinetic parameter in this computational modelof steroidogenesis in NCI-H295R cells was evaluated bydynamic sensitivity analysis The sensitivity (119878
119909119910) of kinetic
parameter 119909 for variable parameter 119910 was defined by the fol-lowing equation
119878119909119910(119905)=
Δ119910(119905)119910(119905)
Δ119909119909 (2)
where variable parameter119910was the concentration of a steroidhormone in the cytosolic space of NCI-H295R cells Theperturbation for kinetic parameters was +10 (Δ119909119909 = 01)
3 Results
31 Experimental Data on Adrenal Steroidogenesis
311 Adrenal Steroidogenesis of NCI-H295R Cells and theMass Balance All steroid hormones in the culture mediumwere significantly increased after 72 h of stimulation with50 nM ACTH 20120583M forskolin and 100 nM angiotensinII (Figure 2(a)) Mass balances in steroidogenesis of NCI-H295R cells under nontreatment and control (stimulated)conditions are shown in Figures 2(b) and 2(c) respectivelyUnder stimulation the dynamics of net mass in these exper-iments were unchanged and accumulated cholesterol wasconverted to adrenal steroids
312 Cytotoxicity of Adrenal Toxicants Viabilities of cellstreated with each compound were expressed as a relativevalue to the ATP level of the control Effects of AN SM TGFN AGT DDD SP MP KC MC and DZ on cell viabilitywere determined to be valid under 80 of the relative ATPlevel at 7 days after treatment AN SM TG and FN showedcytotoxicity at over 100 1 10 and 10120583M respectively AGTMP andDZ did not affect cell viability at up to 100 120583MDDDSP KC andMC induced less than 80 of cell viability at over100 50 100 and 25 120583M respectively
313 Differentially Steroid Profiling of Adrenal ToxicantsAfter NCI-H295R cells were exposed to each test compoundduring three days the concentrations of 12 steroid hor-mones in the culture mediumwere simultaneously measuredby LCMSMS All effects of the compounds on adrenalsteroidogenesis were evaluated at the concentration withoutany overt cytotoxicity The differential metabolic steroidprofiles of 11 adrenal toxic compounds were classified andvisualized by using hierarchical clustering analysis (Figure 3)
Four adrenal toxicants without steroidogenic inhibitionAN SM TG and FN did not change the medium concen-trations of all steroid hormones by more than 2-fold Above-mentioned 4 compounds at every condition and 7 adrenalsteroidogenic inhibitors at the low exposure concentrationwere gathered into a big cluster as nonchange group The7 steroidogenic inhibitors at the maximum exposure con-centration showed the characteristic steroid profiles eachbut 100 120583M DZ and 10 120583M SP were classified as a clusterAGT drastically decreased the medium concentrations ofPREG HPREG DHEA PROG DCORTICO CORTICOand ALDO at 100 120583M DDD dose-dependently decreased themedium concentrations of PROG DCORTICO CORTICOCORT and ALDO at gt10 120583M and decreased PREG HPREGDHEA PROG HPROG DIONE and DCORT at the maxi-mum exposure concentration of 25120583M SP increased PREGHPREG and DHEA and decreased PROG DIONE DCOR-TICODCORTCORTICOALDO andTESTOat 10120583MMPdose-dependently decreased CORTICO CORT and ALDOand decreased DHEA HPROG DIONE and TESTO at themaximum exposure concentration of 100 120583M KC drasticallydecreased the medium concentrations of PREG HPREGDHEAHPROGDIONEDCORTICODCORTCORTICOCORTO ALDO and TESTO at 10 120583M MC increased themedium concentrations of PROG and decreased DIONEDCORT CORT and TESTO at 10 120583M DZ increased PREGHPREG and DHEA and decreased DIONE DCORTICODCORT CORTICO CORT ALDO and TESTO at 100 120583M
32 The Mathematical Modeling
321 Optimization of the Mathematical Model of AdrenalSteroidogenesis in NCI-H295R Cells The mathematicalmodel of adrenal steroidogenesis in NCI-H295R cells wasoptimized for several kinetic parameters of cholesteroltransport intracellular localization the oxysterol pathwayand maximum velocity of steroidogenic enzymes to fit theexperimental time-course data All optimized kinetic para-meters are shown in Table S1 in a supplementary documentThe optimized mathematical model was reconstructedwith the experimental dynamic patterns of cholesteroland the 14 steroid hormones in the intracellular space andculture medium The fitness was 0621761 of NLSD valuesas the fitting objective function The simulation results andexperimental data are shown in Figure 4
Optimized kinetic parameters were calculated sensitivi-ties for the NLSD value as the fitting score and are shown inTable S1 in a supplementary document The highly sensitiveparameters for fitting the NLSD score were the extractednine kinetic parameters 119896Cholesterol Transport 119896
acc119891 119896loc119891 119896loc119887
119881CYP11A1max 119870CYP11A1
119898 119881CYP17HmaxA 119881HSD3B2
maxA and 119881CYP21A2maxA whichhad higher than 30 fitting sensitivity
33 The Validation Using the Adrenal Toxicants
331 Mechanistic Analysis of Adrenal Toxicants Effects ofadrenal toxic compounds on steroidogenic enzymes werequantitatively predicted from the change in the ratio ofthe measured medium concentrations of the 12 steroid
8 Journal of Toxicology
0001
001
01
1
10
PREG
HPR
EG
DH
EA
PRO
G
HPR
OG
AN
DRO
DCO
RTIC
O
DCO
RT
CORT
ICO
CORT
ALD
O
TEST
O E1 E2C
once
ntra
tion
in m
ediu
m (120583
M)
lowast
lowast
lowast
lowast
lowast
lowast
lowast lowast lowastlowast
lowast
lowast
lowast
lowast
(a)
0 8 24 48 72Time (hours)
0
40
80
120
160
200
240
Num
ber o
f mol
ecul
es (n
mol
)
(b)
0 8 24 48 72Time (hours)
0
40
80
120
160
200
240
Num
ber o
f mol
ecul
es (n
mol
)
(c)
Figure 2 Experimental data of metabolic profiling of adrenal steroid hormones and the mass balance Concentrations of steroid hormonessecreted from NCI-H295R cells in the culture medium at 72 h after stimulation were compared with the untreated condition and thestimulated condition by 50 nM ACTH 20 120583M forskolin and 100 nM angiotensin II (a) Net molecular amounts including cholesterol andsteroid hormones in the culture medium and intracellular space are plotted at five time points (0 8 24 48 and 72 h after stimulation)under the untreated condition (b) and the stimulated condition (c) In the bar graphs dark and light bars indicate the amount of cholesteroland adrenal steroids respectively All data are shown as the mean plusmn SD (119873 = 4) lowast119901 values corrected by the familywise error ratelt001 PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-hydroxyprogesteroneDIONE androstenedioneDCORTICO 11-deoxycorticosteroneDCORT 11-deoxycortisol CORTICO corticosteroneCORT cortisol ALDO aldosterone and TESTO testosterone
hormones at 72 h after drug exposure using themathematicalmodel of adrenal steroidogenesis in NCI-H295R cells Thereproducibility of the estimated results was confirmed byperforming the test twice The estimated effects of 11 adrenaltoxic compounds on eight steroidogenic enzymes are shownin Figure 5 The adrenal toxic compounds without steroido-genic inhibition such as vasculotoxic agents (AN SM TGand FN) were not estimated for the target steroidogenicenzymes under noncytotoxic conditions Every fitness valueswere under 005 of NLSD values used as the fitting objectivefunction (Figures 5(a)ndash5(d)) Other steroidogenic inhibitors(AGT DDD SPMP KCMC andDZ) are described in detailbelow
332 AGT The mechanism of action of AGT in adrenalsteroidogenesis was estimated by inhibition of CYP11A1HSD3B2 CYP21A2 and CYP11B1 at 100 120583M (estimatedinhibitions were 770 780 811 and 598 resp)(Figure 5(e)) AGT has been reported to inhibit CYP11A1
CYP21A2 CYP11B1 and CYP11B2 [6 27ndash30] Our resultswere mostly consistent with the previous reports In particu-lar CYP11A1 appeared to be inhibited strongly byAGT In ourstudy HSD3B2 inhibition of AGT was shown by mechanisticanalysis based on systems biology approaches as a novelmechanism of action of AGT Inhibition of AGT by CYP11B2was not estimated in our study However the concentrationof ALDO in the culture medium decreased to 38 ofthe normal stimulated condition Inhibition of AGT byCYP11B2 has been shown using sheep adrenal homogenatesas well as a human adrenal homogenate from a patientwith Cushingrsquos syndrome [30]The activity of 18-hydroxylaseinduced by CYP11B2 was determined as the conversion ofcorticosterone to 18-hydroxycorticosterone in the previousstudy The cause of the discrepancy regarding the effect ofAGT on CYP11B2 was suggested to be substrate inhibitionbecause the intracellular concentration of CORTICO wasincreased by over 10 times of that in the culture mediumto reach 50 120583M Another possibility was poor quantitative
Journal of Toxicology 9
log2
ratio
4
3
2
1
0
minus1
minus2
minus3
minus4
ALDO
CORTICO
CORT
TESTO
DIONE
HPROG
DCORTICO
DCORT
PROG
PREG
HPREG
DHEA
DZ(
100)
SP(1
0)SP
(1)
DZ(
10)
DD
D(0
1)
AN
(01
)A
N(1
)A
N(1
0)SM
(00
1)TG
(00
1)M
P(0
1)FN
(01
)TG
(1)
AGT(
1)AG
T(0
1)SM
(00
0001
)D
DD
(1)
SM(0
1)
AGT(
10)
MP(
1)FN
(1)
MC(
1)D
Z(1)
KC(0
1)
MC(
01)
DZ(
01)
KC( 1
)D
DD
(10)
MP(
10)
MC(
10)
MP(
100)
AGT(
100)
DD
D(2
5)KC
(10)
Figure 3 Hierarchical cluster analysis of differential metabolic profiles of 12 steroid hormones by exposure to adrenal toxicity compoundsAdrenal toxicants were classified by using the differential metabolic profiling of 12 steroid hormones Concentrations of 12 adrenalsteroids secreted from NCI-H295R cells were drastically changed by the exposure of adrenal steroidogenesis inhibitors The 12 adrenalsteroid hormones were quantitatively measured by LC-MSMS simultaneously Four adrenal vasculotoxic compounds acrylonitrile ANfumaronitrile FN salinomycin SM and thioguanine TG were used as negative control compounds for adrenal steroidogenesis inhibitorsSeven adrenal steroidogenesis inhibitors aminoglutethimide AGT op1015840-DDD DDD spironolactone SP metyrapone MP ketoconazoleKC miconazole MC and daidzein DZ showed a characteristic steroid profile each and were classified as each independent singletonat the maximum exposure condition Exposure concentrations of adrenal toxic compounds were described in brackets of the samplename and the units were prepared as 120583M A blue cluster was classified as a group of nonchange samples including negative controlcompounds and low exposure conditions of adrenal steroidogenesis inhibitors PREG pregnenolone HPREG 17120572-hydroxypregnenoloneDHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-hydroxyprogesterone DIONE androstenedione DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICO corticosterone CORT cortisol ALDO aldosterone and TESTO testosterone
reliability of the experimental data because the ALDOconcentration was under the lower limit of quantification at100 120583MAGT Hecker and colleagues reported that 3 120583MAGTdecreases PREG and PROG concentrations and increases theTESTO concentration [10] However AGT did not increasethe TESTO concentration in our study One possibility isthat the concentration of TESTO was already enhanced byabout 33-fold through stimulation with ACTH forskolinand angiotensin II
333 DDD The mechanism of action of DDD in adrenalsteroidogenesis was estimated by dose-dependent inhibitionof CYP11A1 HSD3B2 CYP21A2 and CYP11B1 (estimatedinhibitions at 25120583M were 870 869 769 and 849resp) (Figure 5(f)) DDD has been reported to inhibitCYP11A1 HSD3B2 CYP21A2 CYP11B1 and CYP11B2 [29 3132] Inhibition of DDD by CYP11B2 was not estimated in ourstudy However the concentration of ALDO in the culturemedium decreased to 3 of that in the normal stimulatedcondition Inhibition of DDD by CYP11B2 has been shownusing mitochondrial and microsomal fractions prepared bystandard centrifugation procedures from a bovine adrenal
cortex homogenate [32]The cause of the discrepancy regard-ing the inhibition ofDDDbyCYP11B2 could not be explainedby same effect in the case of AGT
334 SP Themechanism of action of SP in adrenal steroido-genesis was estimated by inhibition of HSD3B2 CYP21A2and HSD17B3 (estimated inhibitions at 10120583M were 702595 and 593 resp) (Figure 5(g)) SP has been reportedto inhibit CYP17H CYP17L CYP11B1 and CYP11B2 [6 33ndash35] The inhibitory effect of SP on the HSD3B2 enzyme wasa novel mechanism of action The main action of SP is asa mineralocorticoid receptor (MR) antagonist SP has alsobeen reported to exert some off-target effects by binding toandrogen glucocorticoid and progesterone receptors [56ndash58] SP has been shown to inhibit the production of ALDOand CORT from PREG induced by angiotensin II in H295Rcells but the specific MR antagonist eplerenone did not showthe inhibitory effects [59] Therefore HSD3B2 inhibition bySP is not mediated via MR and the action might be directinhibition of HSD3B2 enzymes or a part of known off-targeteffects mediated through other nuclear hormone receptorsRegarding CYP17H and CYP17L our results were consistent
10 Journal of Toxicology
CHOL
0
20
40
60
80
100C
once
ntra
tion
in m
ediu
m (120583
M)
10 20 30 40 50 60 70 800Time (hours)
(a)
PREG
DCORT
DCORTICO
CORTICO
CORT
Con
cent
ratio
n in
med
ium
(120583M
)
10
4
2
0
8
6
10 20 30 40 50 60 70 800Time (hours)
(b)
Con
cent
ratio
n in
med
ium
(120583M
)
0
020406
081
1214
0 10 20 30 40 50 60 70 80Time (hours)
HPREG
DIONE
PROGHPROG
E1
(c)
DHEAE2
ALDO
TESTO
0
005
01
015
02
025
Con
cent
ratio
n in
med
ium
(120583M
)
0 10 20 30Time (hours)
40 50 60 70 80
(d)
CHOL
0
5000
10000
15000
20000
25000
30000
Con
cent
ratio
n in
cell
(120583M
)
10 20 30 40 50 60 70 800Time (hours)
(e)
Con
cent
ratio
n in
cell
(120583M
)
0
50
100
150
200
250
0 10 20 30 40 50 60 70 80Time (hours)
PREG
DCORT
DCORTICO
CORTICOCORT
(f)
0 10 20 30 40 50 60 70 80Time (hours)
HPREG
DIONE
PROGHPROG
E1
Con
cent
ratio
n in
cell
(120583M
)
3025201510
35
05
(g)
005
115
2
253
35
Con
cent
ratio
n in
cell
(120583M
)
0 10 20 30 40 50 60 70 80Time (hours)
DHEA
E2ALDO
TESTO
(h)
Figure 4 Comparison of time-course profiles of cholesterol and adrenal steroids produced by NCI-H295R cells between experimentallymeasured and simulated data To intuitively confirm reconstruction of the measured experimental data in the developed simulation modelof NCI-H295R cells dynamics of cholesterol and adrenal steroids produced by NCI-H295R cells were plotted to overlay experimentaldata with the simulated results Graphs show the dynamics of medium concentrations of cholesterol (a) and adrenal steroids ((b)ndash(d))and intracellular concentrations of cholesterol (e) and adrenal steroids ((f)ndash(h)) Steroid hormones were categorized into three groups byconcentration levels Major steroids were PREG pregnenolone DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICOcorticosterone and CORT cortisol ((b) and (f)) Moderate steroids were HPREG 17120572-hydroxypregnenolone PROG progesterone HPROG17120572-hydroxyprogesterone DIONE androstenedione and E1 estrone ((c) and (g)) Minor steroids were DHEA dehydroepiandrosteroneALDO aldosterone TESTO testosterone and E2 17120573-estradiol ((d) and (h)) Experimental data are shown as symbols with dotted lines Alldata represent the mean plusmn SD (119873 = 4) Simulation data are shown as solid lines
Journal of Toxicology 11
0110
10
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
AN (120583M)
(a)
0110
FN (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(b)
00101
SM (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
100E minus 05
(c)
0110
TG (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B1
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(d)
0110
10100
AGT (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(e)
0110
1025
DDD (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(f)
1010
SP (120583M)
2750005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(g)
0110
10100
MP (120583M)
2110005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(h)
Figure 5 Continued
12 Journal of Toxicology
0110
10
KC (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(i)
0110
10
MC (120583M)
4430005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(j)
0110
10100
DZ (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(k)
Figure 5 Estimated mechanism of action of adrenal toxicants by using the mathematical model of adrenal steroidogenesis in NCI-H295Rcells Mechanisms of action of adrenal toxicity compounds were quantitatively estimated from experimental results of differential steroidprofiling using the mathematical model of adrenal steroidogenesis in NCI-H295R cells The drug action was defined as a scaling factorof enzymatic activity in the simulation model These scaling factors were optimized to fit experimental data by a hybrid optimizationmethod of the RCGA and nonlinear least squares Estimated drug actions by the exposure of vasculotoxic agents acrylonitrile AN (a)fumaronitrile FN (b) salinomycin SM (c) and thioguanine TG (d) and the steroidogenic inhibitors aminoglutethimide AGT (e) op1015840-DDD DDD (f) spironolactone SP (g) metyrapone MP (h) ketoconazole KC (i) miconazole MC (j) and daidzein DZ (k) are shownas a spider radar chart CYP11A1 P450 side chain cleavage enzyme CYP17H 17120572-hydroxylase of CYP17 CYP17L C
17ndash20 lyase of CYP17HSD3B2 3120573-hydroxysteroid dehydrogenase CYP21A2 21-hydroxylase CYP11B1 11120573-hydroxylase CYP11B2 18-hydroxylase and HSD17B317120573-hydroxysteroid dehydrogenase
with previous reports [33 34] 7120572-thiospironolactone whichis synthesized by deacetylation of SP inhibits CYP17H andCYP17L [34] The fact that there were no inhibitions ofCYP17HorCYP17L in our study suggests that SPmight not bedeacetylated to 7120572-thiospironolactone in NCI-H295R cellsRegarding CYP11B1 and CYP11B2 our results were unclearcompared with a previous study It has been shown that30 120583M SP inhibits CYP11B1 and CYP11B2 in human andbovine adrenal mitochondria [35]The cause of CYP11B1 andCYP11B2 inhibition by SP could not be determined in ourstudy which might be due to the lower maximum exposureconcentration of SP than that in the previous reportWe couldnot examine SP concentrations over 10 120583M because theseconcentrations were cytotoxic in NCI-H295R cells
335 MP The mechanism of action of MP in adrenalsteroidogenesis was estimated by dose-dependent inhibitionof CYP11B1 (estimated inhibitions at 1 10 and 100 120583M were571 827 and 982 resp) (Figure 5(h)) MP has beenreported to inhibit CYP11B1 as its major effect and CYP11A1and CYP11B2 as a weak effect [6 29 36ndash39] The results wereable to show that MP is a selective inhibitor of CYP11B1 inthe previous report However the estimated effect of MPat a high concentration 100 120583M as the maximum exposureconcentration was unclear According to the previous reportselectivity of MP for CYP11B1CYP11B2 is about five times[39] In addition 20 120583M MP has a slight inhibition effect onCYP11A1 in H295R cells [29]
336 KC The mechanism of action of KC in adrenalsteroidogenesis was estimated by inhibition of CYP11A1CYP17H CYP17L HSD3B2 CYP21A2 CYP11B1 and
CYP11B2 (estimated inhibitions at 10 120583Mwere 926 943518 830 882 974 and 798 resp) (Figure 5(i))KC has been reported to inhibit CYP11A1 CYP17H CYP17LHSD3B2 CYP21A2 and CYP11B1 [6 29 40ndash43] Our resultswere almost consistent with the previous reports KC inhibitsCYP11A1 CYP17H CYP21A2 and CYP11B1 in NCI-H295Rcells at 10 120583M [29] and CYP17H CYP17L CYP21A2 andCYP11B1 in human adrenal mitochondria and Leydig cellmicrosomes at 2ndash5120583M [60 61] However KC has shownonly weak inhibition of HSD3B2 and HSD17B3 in Leydigcells at the millimolar level [60 61] Regarding CYP11B2 andHSD17B3 we considered that these estimated inhibitions ofKC did have sufficient reliability in terms of quantitative pre-diction precision because ALDO and TESTO concentrationswere less than the lower limit of quantification at 10 120583MKC
337 MC Themechanism of action of MC in adrenal stero-idogenesis was estimated by inhibition of CYP17H CYP17LCYP11B1 and HSD17B3 (estimated inhibitions at 10120583Mwere691 530 764 and 571 resp) (Figure 5(j)) MC hasbeen reported to inhibit not only CYP17H and CYP17L butalsoCYP11A1 CYP21A2 andCYP11B1 [41 44 45]The resultsin the previous reports were able to estimate that MC isa CYP17 inhibitor However CYP11A1 inhibition by MCprobably instead of a reduction in StAR expression wasnot clearly detected in our study using NCI-H295R cellsbecause there were no decreases in the concentrations ofPREG and PROG in the culture medium Indirect inhibitionof CYP11A1 via the peripheral-type benzodiazepine receptorhas been reported in mouse adrenocortical Y-1 cells treatedwith MC in the absence of stimuli by measuring PREGproduction [44] On the other hand reductions of StAR
Journal of Toxicology 13
protein expression andor transport activity without affectingtotal steroid synthesis or CYP11A1 and HSD3B2 enzymeexpression or activities have been reported in (BU)
2cAMP-
stimulated MA-10 Leydig tumor cells treated with MC bymeasuring PROG production [45] Therefore the effect ofMC on the initial reaction in adrenal steroidogenesis fromcholesterol should be different according to the cell type andstimulation condition Inhibition of CYP21A2 and CYP11B1by MC has been reported as decreases in the consumptionof PROG and DCORTICO respectively [41] Inhibition ofCYP11B1 was estimated by the action of MC in this studybut that of CYP21A2 was not detected In the previousexperimental report inhibitory sites by MCmight have beenreflected by inhibition of CYP17H activity because CYP21A2activity was measured as a decrease in labeled PROG
338 DZ The mechanism of action of DZ in adrenalsteroidogenesis was estimated by inhibition of CYP11A1HSD3B2 CYP21A2 CYP11B1 and HSD17B3 (estimated inhi-bitions at 100 120583M were 586 941 965 872 and981 resp) (Figure 5(k)) DZ has been reported to inhibitHSD3B2 and CYP21A2 [46] The results of HSD3B2 andCYP21A2 were consistent with the previous report Howeverinhibitions have not been reported for CYP11A1 CYP11B1and HSD17B3These estimated effects of DZ on CYP11B1 andHSD17B3 were unclear because the concentrations of ALDOand TESTO were less than the lower limit of quantificationat 100 120583M In addition these enzymes act downstream of thestrong action points of DZ such as HSD3B2 and CYP21A2
34 The Simulations and the Systematic Model Analysis
341 Dynamic Sensitivity Analysis of Adrenal SteroidogenesisTo comprehensively understand the dynamics of adrenalsteroidogenesis dynamic sensitivities were calculated forsteroid concentrations secreted by NCI-H295R cells usingour constructed mathematical model of steroidogenesis Theresults of dynamic sensitivity analysis at 72 h of duration and6 h of interval time are presented as a heat-map in Figure 6
The top 10 parameters of the total area under the curveof dynamic sensitivity for cholesterol and the 14 steroids inculture medium were 119881HSD3B2
maxA 119881CYP11A1max 119881CYP21A2maxA 119896loc119891 119896loc119887
119870CYP11A1119898
119881CYP17HmaxA 119896Cholesterol Transport 119896acc119891 and 119881CYP21A2maxB in
order from the top Cholesterol uptake (119896Cholesterol Transport)StAR protein (119896loc
119891) and CYP11A1 (119881CYP11A1max ) which are
determining factors of the capacity for steroidogenesis pro-moted the production of mineralocorticoids (DCORTICOCORTICO and ALDO) and restrained the synthesis of glu-cocorticoids (DCORT and CORT) and sex steroids (DIONETESTO and E1) because of the accumulation of interme-diate molecules in steroidogenesis (PREG HPREG DHEAPROG and HPROG) only by self-activation The dynamicpatterns of the intermediate molecules in steroidogenesiswere mainly dependent on the activity of CYP17H andHSD3B2 with PREG as the substrate of these enzymes inwhich the dynamic sensitivities of 119881CYP17HmaxA for HPREG andHPROG and 119881HSD3B2
maxA for PROG HPROG and DCORTICO
reversed the direction of sensitivity at 49ndash66 h after stimu-lation The dynamic sensitivities of the maximum activitiesof HSD3B2 for PREG (119881HSD3B2
maxA ) and CYP21A2 for PROG(119881CYP21A2maxA ) were related to all steroids at 72 h Almost allmodel parameters had positive sensitivity for downstreamsteroids in the adrenal steroidogenic pathway and negativesensitivity for direct-binding steroids as substrates of thesteroidogenic enzyme The sensitivity of 119881max in all steroido-genic enzymes was relatively higher than 119870
119898for the same
steroid substrate
342 Simulation of the Metabolic Balance of AdrenalSteroidogenesis Pathway To clearly show the property of themetabolic shift between mineralocorticoid and glucocorti-coid biosynthesis we performed two-dimensional parameterscanning of the enzymatic activities of CYP17H and HSD3B2(Figure 7) NCI-H295R cells lost the ability to produce allsteroid hormones when enzymatic activities of CYP17H andHSD3B2 were changed by over 60 and 30 respectivelyActivation ofCYP17HandorHSD3B2 induced themetabolicshift that enhanced the glucocorticoid biosynthesis and devi-ated from the mineralocorticoid biosynthesis On the otherhand inhibition of CYP17H andor HSD3B2 induced themetabolic shift that enhanced the mineralocorticoid biosyn-thesis and deviated from the glucocorticoid biosynthesisMoreover the enzymatic activity of HSD3B2 regulated themetabolic balance of sex steroids and the precursors onadrenal steroidogenesis of NCI-H295R cells E1 TESTO andDIONE were produced by NCI-H295R cells when activatingthe enzymatic activity of HSD3B2 Conversely E2 andDHEAwere produced by NCI-H295R cells when suppressing theenzymatic activity of HSD3B2 The biosynthesis of down-stream steroids in adrenal steroidogenesis pathway such asmineralocorticoids and glucocorticoids was almost com-pletely terminated when the enzymatic activity of HSD3B2was decreased by over 80
4 Discussion
41 Importance of 3120573-HSDActivity in Adrenal SteroidogenesisOur systematic analysis using the mathematical model ofadrenal steroidogenesis in NCI-H295R cells revealed thatthe enzymatic activity of 3120573-HSD controls the dynamics ofadrenal steroidogenesis The activity of the StAR proteincontrols the net capacity of steroidogenesis in steroidogeniccells which is the transport of cholesterol from the outer toinner mitochondrial membranes Both the expression levelsof StAR protein and mRNA are rapidly elevated in responseto stimulation by tropic hormones such as ACTH [62 63]Another important factor in adrenal steroidogenesisis is thecholesterol side chain cleavage enzyme CYP11A1 the firstrate-limiting and hormonally regulated step in the synthesisof all steroid hormones which is conversion of cholesterolto pregnenolone in mitochondria [64] According to ourresults of global sensitivity analysis (Supplementary Table 1and Figure 6(d)) in addition to CYP11A1 and StAR proteins3120573-HSD was one of the key regulators in adrenal steroido-genesis of NCI-H295R cells And also this result suggeststhat a significant regulatory mechanism in steroidogenesis
14 Journal of Toxicology
60544830 4236 66 7218 246 120
Time (hours)
0
1
2
3C
once
ntra
tion
(a)
60544830 4236 66 7218 246 120
Time (hours)
minus05
0
05
minus1
1
Sens
itivi
ty
(b)
0
6
12
18
24
30
36
42
48
54
60
66
72
Sens
itivi
ty fo
r med
ium
conc
entr
atio
ns
of ch
oles
tero
l or s
tero
id h
orm
ones
Dur
atio
n tim
e (ho
urs)
Model parameters
+10
+05
00
minus05
minus10
(c)
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
Parameter number in the mathematical model of steroidogenesis in NCI-H295R cells
CHOL
PREG
HPREG
DHEA
PROG
HPROG
DIONE
DCORTICO
DCORT
CORTICO
CORT
ALDO
TESTO
E1
E2
Chol
este
rol o
r ste
roid
hor
mon
es
0
05
10
15
20
minus05
minus10
minus15
minus20
(d)
Figure 6 Heat-map of the global dynamic sensitivity analysis of adrenal steroid concentrations produced by NCI-H295R cells The globaldynamic sensitivity analysis is a powerful tool to comprehensively understand the dependencies of themodel parameters in themathematicalmodel of the biological complex system Dynamic sensitivities of model parameters in the mathematical model of adrenal steroidogenesisin NCI-H295R cells were calculated for all steroid concentrations in the culture medium every 6 h until 72 h after stimulation To clarify theview of heat-map of global dynamic sensitivity analysis imaginary data of the dynamics of steroid concentrations in the original model (blueline) and perturbed model for sensitivity analysis (red line) were prepared (a) Using this imaginary data the calculated dynamic sensitivities(b) and the visualized dynamic sensitivity as one block of the heat-map (c) were shown respectively By the same method that explainedusing imaginary data the large-scale data of the global dynamic sensitivity analysis on the mathematical model of adrenal steroidogenesisin NCI-H295R cells was comprehensively visualized as a big graph of heat-map (d) Parameter numbers in the horizontal axis are (1)119896Choresterol Transport (2) 119896CEH (3) 119896MTR
119891 (4) 119896MTR
119887 (5) 119896acc
119891 (6) 119896acc
119887 (7) 119896loc
119891 (8) 119896loc
119887 (9) 119896Oxysterol Synthesis (10)119870
CYP11A1119898
(11) 119881CYP11A1max (12)119870CYP17H119898A
(13) 119870CYP17H119898B (14) 119881CYP17HmaxA (15) 119881CYP17HmaxB (16) 119881CYP17HmaxB (17) 119870CYP17L
119898B (18) 119881CYP17LmaxA (19) 119881CYP17LmaxB (20) 119870HSD3B2119898A (21) 119870HSD3B2
119898B (22) 119870HSD3B2119898C (23)
119881HSD3B2maxA (24) 119881HSD3B2
maxB (25) 119881HSD3B2maxB (26) 119870CYP21A2
119898A (27) 119870CYP21A2119898B (28) 119881CYP21A2maxA (29) 119881CYP21A2maxB (30) 119870CYP11B1
119898A (31) 119870CYP11B1119898B (32) 119881CYP11B1maxA
(33) 119881CYP11B1maxB (34) 119896CYP11B2 (35) 119870HSD17B3119898A (36) 119870HSD17B3
119898B (37) 119881HSD17B3maxA (38) 119881HSD17B3
maxB (39) 119870CYP19A1119898A (40) 119870CYP19A1
119898B (41) 119881CYP19A1maxA and (42)119881
CYP19A1maxB PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-
hydroxyprogesteroneDIONE androstenedioneDCORTICO 11-deoxycorticosteroneDCORT 11-deoxycortisol CORTICO corticosteroneCORT cortisol ALDO aldosterone and TESTO testosterone
pathway is very reasonable StAR CYP11A1 and 3120573-HSD(isoforms 1 or 2 in humans) proteins generally respond to thesame hormones that stimulate steroid production throughcommon pathways such as cAMP signaling in adrenal glandsand testes [65 66] Moreover our data also support recentexperimental evidence from clinical and in vivo studiessuggesting that the enzymatic activity of 3120573-HSD plays
an important role in the regulation of mineralocorticoidsynthesis in adrenal steroidogenesis and contributes to hyper-tension caused by abnormal overproduction of aldosterone[67ndash70] Circadian clock-deficient Cry-null mice show salt-sensitive hypertension due to abnormally high synthesis ofaldosterone which is caused by constitutively high expressionof HSD3B6mRNA and protein in the adrenal cortex [67 68]
Journal of Toxicology 15
All steroid producing cells
Mineralocorticoid producing cells Glucocorticoid producing cells
Mineralocorticoid and glucocorticoid producing cells
Terminated steroidogenesis
Sex steroids and precursor sex steroidsDIONE TESTO E1DEHA E2
NCI-H295Rcells at controlcondition
Normalized activities of HSD3B2
Nor
mal
ized
activ
ities
of C
YP17
H
2
18
16
14
12
1
08
06
04
02
00 05 1 15 2
Figure 7 Metabolic categories of the steroidogenic cells determined by the balance of HSD3B2 and CYP17H activitiesThe two-dimensionalparameter scanning analysis by the perturbation of two focused parameters clarifies the interaction and the relationship between the twoparameters in the complex system Functional cellular categories of steroidogenic cells were defined by the levels of mineralocorticoid(ALDO) glucocorticoids (DCORTandCORT) and androgens (DHEAandDIONE) at 72 h after stimulation Enzymatic activities ofHSD3B2andCYP17were normalized by standard values of the simulationmodel inNCI-H295R cells Scanning ranges ofHSD3B2 andCYP17 activitieswere 0ndash200 each 10 Green regions are all 14 steroids produced by NCI-H295R cells Red blue and purple regions are mineralocorticoidproducing cells glucocorticoids-producing cells and both corticoid-producing cells respectively Yellow regions are steroidogenesis of NCI-H295R cells terminated upstream of the adrenal steroidogenesis pathway
Recent clinical observations have revealed predominantexpression of HSD3B2 mRNA and protein in tumor cellsof aldosterone-producing adenoma (APA) and HSD3B1mRNA significantly correlated with CYP11B2 mRNA levelsand plasma aldosterone concentrations in APA patients [6970] However the relationship is unclear and disputed ina small-scale clinical study indicating that genetic variationin HSD3B1 affects blood pressure and hypertension in APApatients [71] The results of our simulation study suggest that3120573-HSD protein (human genes are HSD3B1 and HSD3B2) isone of the determination factors for the dynamic propertyof adrenal steroidogenesis Our results support the clinicalevidence of Doi and colleagues [69] and we believe that theHSD3B1 enzyme has a promising potential as novel drugtarget for endocrine hypertension
42 Metabolic Shift of Adrenal Steroidogenesis and the Contri-butions of HSD3B2 and CYP17H The metabolic propertiesof adrenal steroidogenesis in NCI-H295R cells were revealedby dynamic sensitivity analysis using themathematicalmodel(Figure 6) Mineralocorticoids such as DCORTICO COR-TICO and ALDO and intermediate steroids upstream of
the adrenal steroidogenesis pathway such as PREG HPREGDHEA PROG and HPROG were accelerated by reactionsof cholesterol import (119896Cholesterol Transport) StAR protein (119896MTR
119891
and 119896loc119891) and CYP11A1 (119881CYP11A1max ) On the other hand glu-
cocorticoids such as DCORT and CORT and sex hormonessuch as DIONE TESTO and E1 were suppressed by thesemodel parametersTherefore enhancement of the net adrenalsteroidogenesis capacity which supplies PREG precursor tothe pathway causes a production shift from glucocorticoidsto mineralocorticoids by substrate inhibitions of CYP17HHSD3B2 and CYP21A2 caused by accumulation of initialintermediate steroids such as PREG and PROG Sensitivi-ties of CYP17H (119881CYP17HmaxA ) and HSD3B2 (119881HSD3B2
maxA ) for theproducts were dynamically changed and these parametersdetermined the metabolic balance of downstream steroidsin the adrenal steroidogenesis pathway According to theseresults of dynamic sensitivity analysis of StAR CYP11A1CYP17H and HSD3B2 we suggest that the enhancementof CYP17H and HSD3B2 activity during ACTH stimulationwas important to shift the steroidogenic output away fromALDO biosynthesis towards CORT biosynthesis as wellas adrenal androgen production This suggestion partially
16 Journal of Toxicology
supports a comparative animal study in which molecularand cellular variations in CYP17H activity dramatically affectacute cortisol production resulting in distinct physiologicaland behavioral responses [72]
Results of two-dimensional parameter scanning of theenzymatic activities of CYP17H and HSD3B2 quantitativelyshowed the detail of the metabolic relationship betweenmin-eralocorticoid and glucocorticoid biosynthesis (Figure 7)Particularly the results showed that the balance of theseenzymatic activities was very important for the typicalfunction of NCI-H295R cells namely the ability to produceall steroid hormones NCI-H295R cells lost this functionwhen enzymatic activities of CYP17H and HSD3B2 werechanged by over 60 and 30 respectively In addition theybecame mineralocorticoid (ALDO) secreting cells when theenzymatic activity of CYP17H or HSD3B2 was inhibited byover 50 or glucocorticoid (DCORT and CORT) secretingcells when these enzymes were activated by over 50 Inparticular this analysis also showed that HSD3B2 was a keyplayer in the adrenal steroidogenesis of NCI-H295R cellsbecause HSD3B2 inhibition by over 80 almost completelyinhibited the biosynthesis of downstream steroids The ratioof CYP17A1 to HSD3B2 mRNA expression levels has beenrelated to several endocrine diseases with a low level in APAs[73] and high level in cortisol-producing adenomas [74]Furthermore the expression levels or enzymatic activitiesof CYP17A1 and HSD3B1 have been related to androgenproduction in polycystic ovary syndrome [75 76] Theseclinical studies support our simulation results indicating thatthe balance of enzymatic activity of CYP17H and HSD3B2determines the shift in steroidogenic output to mineralocor-ticoids glucocorticoids or androgens
43 Methodologies of Quantitative Mechanistic Analysis forDrug Discovery According to our results obtained using themathematical model of steroidogenesis in NCI-H295R cellssuch as sensitivity analysis comprehensive analysis based onsystems biology is available to quantitatively estimate themechanism of action of steroidogenic disrupting compoundsfrom differential profiling of adrenal steroid hormonesbecause dynamic patterns of steroid hormones in adrenalsteroidogenesis pathway are highly complex Our proposedmethod of quantitative mechanistic analysis of steroidogenicinhibitors was able to predict known action sites in theadrenal steroidogenesis pathway at only one time point (72 hafter drug exposure) Moreover according to the resultsof sensitivity analysis (Figure 6) 119881max of all steroidogenicenzymes was more sensitive than the 119870
119898 because the
intracellular concentrations of steroid hormones were almostmaintained at sufficiently high levels compared with 119870
119898
values of steroidogenic enzymesThese results suggested thatestimation of the mechanism of action of drugs is moreeffective and detectable when using the influences of 119881maxas the searching parameters such as our proposed methodOur data showed that the proposed method based on asystems biology model is a very powerful tool for exploratoryscreening of steroidogenic disrupting compounds
RCGA as a solver of parameter estimation problemsin systems biology has been applied to biological network
identification of gene regulatory networks and metabolicpathways and optimization of biological processes usingexperimentally observed time-course data [77ndash82] In thisstudy RCGA was useful to estimate the mechanism ofaction of novel pharmaceutical drug candidates for adrenalsteroidogenesis as a new application of RCGA in systemsbiology We had two issues when applying RCGA to thequantitative mechanistic analysis of drug actions Theseissues were the vast calculation cost and multimodality ofquasi-optimum solutions in solving the optimization prob-lem because the mathematical model in systems biologyconsists of many equations and parameters A proposedoptimization strategy using RCGA based on REXJGG wasa highly stable and efficient calculation method for a betterquasi-optimum solution than the unimodal normal distri-bution crossover (UNDX)minimum generation gap (MGG)method that is well applied in the engineering field In addi-tion we expanded the RCGA optimization program basedon REXJGG to a hybrid method of GA and then applieda local search as recommended by Harada and Kobayashi[28 83] A final optimal solution was obtained with a goodconvergence property Because these problems are generalin systems biology studies we suggest that the proposedhybrid method based on REXJGG is a very useful tool forquantitative mechanistic analysis of novel pharmaceuticaldrugs not limited to steroidogenic disrupting compounds
5 Conclusions
Thenovelmathematicalmodel of adrenal steroidogenesis wasconstructed in this study including cholesterol transport anddistribution the C
21-steroid hormone pathway steroid trans-
port and cell proliferation which could reproduce adrenalsteroidogenesis in NCI-H295R cells According to the resultsof dynamic sensitivity analysis using the newmodel HSD3B2plays the most important role in the metabolic balance ofadrenal steroidogenesis in NCI-H295R cells Moreover toquantitatively estimate mechanisms of action of adrenal toxiccompounds we analyzed differential metabolic profiles of12 steroid hormones at 3 days after exposure to 11 adrenaltoxic compounds by using the new mathematical modeland a hybrid optimization method of the RCGA and a localsearch (nonlinear least squares) We could estimate whichsteroidogenic enzymes were affected by these compoundsusing the hybrid optimization method Vasculotoxic agentswere estimated to have no effect according to the resultsobtained by our method In terms of adrenal steroidogenicinhibitors the predicted action sites were approximatelymatched to the target enzymes as reported in the literatureThus our computer-aided method based on a systems biol-ogy approach may be useful to analyze the mechanism ofaction of endocrine-disrupting compounds and to assess thehuman adrenal toxicity of novel pharmaceutical drugs basedon steroid hormone profiling
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Journal of Toxicology 17
Authorsrsquo Contribution
Ryuta Saito Natsuko Terasaki and Naohisa Tsutsui designedthe study Naohisa Tsutsui and Naoya Masutomi supervisedthe experiments Natsuko Terasaki performed all in vitroexperiments and collected dataMakotoYamazaki performedall measurements of steroids Ryuta Saito conceived the insilico strategies analyzed data developed all programs andthe mathematical model and performed all simulation anal-yses with Masahiro Okamoto Ryuta Saito Natsuko TerasakiMakoto Yamazaki NaoyaMasutomi andMasahiroOkamotowrote the paper
Acknowledgments
The authors would like to thank Dr Michael S Breenand Dr Miyuki Breen for helpful discussions concerningprevious mathematical models of adrenal steroidogenesisThis work was partially supported by the Grant-in-Aid forScientific Research on Innovative Areas ldquoSynthetic Biologyrdquo(No 23119001 (MO)) from Ministry of Education CultureSports Science and Technology in Japan
References
[1] P W Harvey D J Everett and C J Springall ldquoAdrenal toxi-cology a strategy for assessment of functional toxicity to theadrenal cortex and steroidogenesisrdquo Journal of Applied Toxi-cology vol 27 no 2 pp 103ndash115 2007
[2] W E Ribelin ldquoThe effects of drugs and chemicals upon thestructure of the adrenal glandrdquo Fundamental and Applied Toxi-cology vol 4 no 1 pp 105ndash119 1984
[3] H D Colby and P A Longhurst ldquoToxicology of the adrenalglandrdquo in Endocrine Toxicology C K Atterwill and J D FlackEds pp 243ndash281 CambridgeUniversity Press Cambridge UK1992
[4] H D Colby ldquoIn vitro assessment of adrenocortical toxicityrdquoJournal of Pharmacological and Toxicological Methods vol 32no 1 pp 1ndash6 1994
[5] T J Rosol J T Yarrington J Latendresse and C C CapenldquoAdrenal gland structure function and mechanisms of toxic-ityrdquo Toxicologic Pathology vol 29 no 1 pp 41ndash48 2001
[6] P W Harvey and D J Everett ldquoThe adrenal cortex and ster-oidogenesis as cellular and molecular targets for toxicity crit-ical omissions from regulatory endocrine disrupter screeningstrategies for human healthrdquo Journal of Applied Toxicology vol23 no 2 pp 81ndash87 2003
[7] A F Gazdar H K Oie C H Shackleton et al ldquoEstablishmentand characterization of a human adrenocortical carcinoma cellline that expresses multiple pathways of steroid biosynthesisrdquoCancer Research vol 50 no 17 pp 5488ndash5496 1990
[8] W E Rainey I M Bird C Sawetawan et al ldquoRegulation ofhuman adrenal carcinoma cell (NCI-H295) production of C19steroidsrdquo The Journal of Clinical Endocrinology amp Metabolismvol 77 no 3 pp 731ndash737 1993
[9] B Staels D W Hum andW L Miller ldquoRegulation of steroido-genesis in NCI-H295 cells a cellular model of the human fetaladrenalrdquo Molecular Endocrinology vol 7 no 3 pp 423ndash4331993
[10] M Hecker J L Newsted M B Murphy et al ldquoHuman adreno-carcinoma (H295R) cells for rapid in vitro determination of
effects on steroidogenesis hormone productionrdquo Toxicologyand Applied Pharmacology vol 217 no 1 pp 114ndash124 2006
[11] A Oskarsson E Ulleras K E Plant J P Hinson and P SGoldfarb ldquoSteroidogenic gene expression in H295R cells andthe human adrenal gland adrenotoxic effects of lindane invitrordquo Journal of Applied Toxicology vol 26 no 6 pp 484ndash4922006
[12] UMuller-Vieira M Angotti and RW Hartman ldquoThe adreno-cortical tumor cell line NCI-H295R as an in vitro screeningsystem for the evaluation of CYP11B2 (aldosterone synthase)and CYP11B1 (steroid-11120573-hydroxylase) inhibitorsrdquoThe Journalof Steroid Biochemistry and Molecular Biology vol 96 no 3-4pp 259ndash270 2005
[13] K Imagawa S Okayama M Takaoka et al ldquoInhibitory effectof efonidipine on aldosterone synthesis and secretion in humanadrenocarcinoma (H295R) cellsrdquo Journal of CardiovascularPharmacology vol 47 no 1 pp 133ndash138 2006
[14] M Voets UMuller-Vieira S Marchais-Oberwinkler and RWHartmann ldquoSynthesis of amidinohydrazones and evaluation oftheir inhibitory effect towards aldosterone synthase (CYP11B2)and the formation of selected steroidsrdquo Archiv der Pharmazievol 337 no 7 pp 411ndash416 2004
[15] F K Nielsen C H Hansen J A Fey et al ldquoH295R cells as amodel for steroidogenic disruption a broader perspective usingsimultaneous chemical analysis of 7 key steroid hormonesrdquoToxicology in Vitro vol 26 no 2 pp 343ndash350 2012
[16] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cortisolproduction and on activities of steroidogenic enzymes inhuman adrenocortical H295R cellsrdquo Journal of Steroid Biochem-istry and Molecular Biology vol 80 no 3 pp 355ndash363 2002
[17] R F Canton J T Sanderson S Nijmeijer A Bergman R JLetcher and M van den Berg ldquoIn vitro effects of brominatedflame retardants and metabolites on CYP17 catalytic activity anovel mechanism of actionrdquo Toxicology and Applied Pharma-cology vol 216 no 2 pp 274ndash281 2006
[18] K Hilscherova P D Jones T Gracia et al ldquoAssessment ofthe effects of chemicals on the expression of ten steroidogenicgenes in the H295R cell line using real-time PCRrdquo ToxicologicalSciences vol 81 no 1 pp 78ndash89 2004
[19] T Gracia K Hilscherova P D Jones et al ldquoThe H295R systemfor evaluation of endocrine-disrupting effectsrdquo Ecotoxicologyand Environmental Safety vol 65 no 3 pp 293ndash305 2006
[20] X Zhang J L NewstedM Hecker E B Higley P D Jones andJ P Giesy ldquoClassification of chemicals based on concentration-dependent toxicological data using ToxClustrdquo EnvironmentalScience and Technology vol 43 no 10 pp 3926ndash3932 2009
[21] M S Breen M Breen N Terasaki M Yamazaki and R BConolly ldquoComputational model of steroidogenesis in humanH295R cells to predict biochemical response to endocrine-active chemicalsmodel development formetyraponerdquoEnviron-mental Health Perspectives vol 118 no 2 pp 265ndash272 2010
[22] M Breen M S Breen N Terasaki M Yamazaki A LLloyd and R B Conolly ldquoMechanistic computational model ofsteroidogenesis in H295R cells role of oxysterols and cell pro-liferation to improve predictability of biochemical response toendocrine active chemical-metyraponerdquo Toxicological Sciencesvol 123 no 1 pp 80ndash93 2011
[23] F B Kraemer ldquoAdrenal cholesterol utilizationrdquo Molecular andCellular Endocrinology vol 265-266 pp 42ndash45 2007
[24] W L Miller ldquoSteroidogenic acute regulatory protein (StAR)a novel mitochondrial cholesterol transporterrdquo Biochimica et
18 Journal of Toxicology
Biophysica Acta (BBA)mdashMolecular and Cell Biology of Lipidsvol 1771 no 6 pp 663ndash676 2007
[25] S Szabo and I T Lippe ldquoAdrenal gland chemically inducedstructural and functional changes in the cortexrdquo ToxicologicPathology vol 17 no 2 pp 317ndash329 1989
[26] C Chen-Pan I-J Pan Y Yamamoto T Sakogawa J Yamadaand YHayashi ldquoPrompt recovery of damaged adrenalmedullaeinduced by salinomycinrdquoToxicologic Pathology vol 27 no 5 pp563ndash572 1999
[27] R J Santen and R I Misbin ldquoAminoglutethimide review ofpharmacology and clinical userdquo Pharmacotherapy vol 1 no 2pp 95ndash120 1981
[28] M J Carella N V Dimitrov V V Gossain L Srivastava andD R Rovner ldquoAdrenal effects of low-dose aminoglutethimidewhen used alone in postmenopausal women with advancedbreast cancerrdquoMetabolism vol 43 no 6 pp 723ndash727 1994
[29] M K Johansson J T Sanderson and B-O Lund ldquoEffects of3-MeSO2-DDE and some CYP inhibitors on glucocorticoidsteroidogenesis in the H295R human adrenocortical carcinomacell linerdquo Toxicology in Vitro vol 16 no 2 pp 113ndash121 2002
[30] Y Touitou A Bogdan J C Legrand and P Desgrez ldquoAmino-glutethimide and glutethimide effects on 18-hydroxycorti-costerone biosynthesis by human and sheep adrenals in vitrordquoActa Endocrinologica vol 80 no 3 pp 517ndash526 1975
[31] M M Hart and J A Straw ldquoStudies on the site of action ofop1015840-DDD in the dog adrenal cortex 1 Inhibition of ACTH-mediated pregnenolone synthesisrdquo Steroids vol 17 no 1ndash5 pp559ndash574 1971
[32] M Ojima M Saitoh N Itoh Y Kusano S Fukuchi and HNaganuma ldquoThe effects of op1015840-DDD on adrenal steroidogen-esis and hepatic steroid metabolismrdquo Nippon Naibunpi Gakkaizasshi vol 61 no 3 pp 168ndash178 1985
[33] L F Canosa and N R Ceballos ldquoEffects of different steroid-biosynthesis inhibitors on the testicular steroidogenesis of thetoad Bufo arenarumrdquo Journal of Comparative Physiology B vol171 no 6 pp 519ndash526 2001
[34] D C Kossor S Kominami S Takemori and H D ColbyldquoRole of the steroid 17 alpha-hydroxylase in spironolactone-mediated destruction of adrenal cytochrome P-450rdquoMolecularPharmacology vol 40 no 2 pp 321ndash325 1991
[35] S C Cheng K Suzuki W Sadee and B W Harding ldquoEffectsof spironolactone canrenone and canrenoate K on cytochromeP450 and 11120573 and 18 hydroxylation in bovine and humanadrenal corticalmitochondriardquo Endocrinology vol 99 no 4 pp1097ndash1106 1976
[36] K Yanagibashi M Haniu J E Shively W H Shen and P HallldquoThe synthesis of aldosterone by the adrenal cortex Two zones(fasciculata and glomerulosa) possess one enzyme for 11120573- 18-hydroxylation and aldehyde synthesisrdquoThe Journal of BiologicalChemistry vol 261 no 8 pp 3556ndash3562 1986
[37] K Morishita H Okumura N Ito and N Takahashi ldquoPrimaryculture system of adrenocortical cells from dogs to evaluatedirect effects of chemicals on steroidogenesisrdquo Toxicology vol165 no 2-3 pp 171ndash178 2001
[38] S W J Lamberts E G Bons H A Bruining and F H de JongldquoDifferential effects of the imidazole derivatives etomidateketoconazole and miconazole and of metyrapone on the secre-tion of cortisol and its precursors by human adrenocorticalcellsrdquo Journal of Pharmacology and Experimental Therapeuticsvol 240 no 1 pp 259ndash264 1987
[39] L Yin S Lucas F Maurer U Kazmaier Q Hu and RW Hartmann ldquoNovel imidazol-1-ylmethyl substituted 1256-tetrahydropyrrolo[321-ij]quinolin-4-ones as potent and selec-tive CYP11B1 inhibitors for the treatment of Cushingrsquos syn-dromerdquo Journal ofMedicinal Chemistry vol 55 no 14 pp 6629ndash6633 2012
[40] M DiMattina N Maronian H Ashby D L Loriaux andB D Albertson ldquoKetoconazole inhibits multiple steroidogenicenzymes involved in androgen biosynthesis in the humanovaryrdquo Fertility and Sterility vol 49 no 1 pp 62ndash65 1988
[41] M Ayub and M J Levell ldquoInhibition of human adrenal steroi-dogenic enzymes in vitro by imidazole drugs including keto-conazolerdquo Journal of Steroid Biochemistry vol 32 no 4 pp 515ndash524 1989
[42] D Engelhardt M M Weber T Miksch F Abedinpour andC Jaspers ldquoThe influence of ketoconazole on human adrenalsteroidogenesis incubation studies with tissue slicesrdquo ClinicalEndocrinology vol 35 no 2 pp 163ndash168 1991
[43] S Bhasin S Sikka T Fielder et al ldquoHormonal effects of keto-conazole in vivo in the male rat mechanism of actionrdquo Endo-crinology vol 118 no 3 pp 1229ndash1232 1986
[44] D M Zisterer and D C Williams ldquoCalmidazolium and otherimidazole compounds affect steroidogenesis in Y1 cells lack ofinvolvement of the peripheral-type benzodiazepine receptorrdquoJournal of Steroid Biochemistry and Molecular Biology vol 60no 3-4 pp 189ndash195 1997
[45] L P Walsh C N Kuratko and D M Stocco ldquoEconazoleand miconazole inhibit steroidogenesis and disrupt steroido-genic acute regulatory (StAR) protein expression post-trans-criptionallyrdquo Journal of Steroid Biochemistry and Molecular Bio-logy vol 75 no 4-5 pp 229ndash236 2000
[46] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cor-tisol production and on activities of steroidogenic enzymesin human adrenocortical H295R cellsrdquo The Journal of SteroidBiochemistry and Molecular Biology vol 80 no 3 pp 355ndash3632002
[47] D P Dempsher D S Gann and R D Phair ldquoA mechanisticmodel of ACTH-stimulated cortisol secretionrdquo American Jour-nal of Physiology vol 246 no 4 part 2 pp R587ndashR596 1984
[48] N Sarai S Matsuoka and A Noma ldquosimBio a Java package forthe development of detailed cell modelsrdquo Progress in Biophysicsand Molecular Biology vol 90 no 1ndash3 pp 360ndash377 2006
[49] K Levenberg ldquoA method for the solution of certain non-linearproblems in least squaresrdquo Quarterly of Applied Mathematicsvol 2 pp 164ndash168 1944
[50] D W Marquardt ldquoAn algorithm for least-squares estimation ofnonlinear parametersrdquo SIAM Journal on Applied Mathematicsvol 11 no 2 pp 431ndash441 1963
[51] P E Gill and W Murray ldquoAlgorithms for the solution of thenonlinear least-squares problemrdquo SIAM Journal on NumericalAnalysis vol 15 no 5 pp 977ndash992 1978
[52] Y Akimoto R Hasada J Sakuma I Ono and S KobayashildquoGeneration alternation model for real-coded GA using multi-parent proposal and evaluation of just generation gap (JGG)rdquoin Proceedings of the 19th SICE Symposium on DecentralizedAutonomous Systems pp 341ndash346 Tokyo Japan January 2007
[53] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoProposal and evaluation of adaptive real-coded crossoverAREXrdquo Transactions of the Japanese Society for Artificial Intel-ligence vol 24 no 6 pp 446ndash458 2009
Journal of Toxicology 19
[54] S Kobayashi ldquoThe frontiers of real-coded genetic algorithmsrdquoJournal of Japanese Society for Artificial Intelligence vol 24 no1 pp 147ndash162 2009
[55] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoAnalysis of the behavior of MGG and JGG as a selection modelfor real-coded genetic algorithmsrdquo Transactions of the JapaneseSociety for Artificial Intelligence vol 25 no 2 pp 281ndash289 2010
[56] A Struthers H Krum and G H Williams ldquoA comparisonof the aldosterone-blocking agents eplerenone and spironolac-tonerdquo Clinical Cardiology vol 31 no 4 pp 153ndash158 2008
[57] K Yamasaki M Sawaki S Noda et al ldquoComparison of theHershberger assay and androgen receptor binding assay oftwelve chemicalsrdquo Toxicology vol 195 no 2-3 pp 177ndash1862004
[58] C Eil and S K Edelson ldquoThe use of human skin fibroblasts toobtain potency estimates of drug binding to androgen recept-orsrdquo The Journal of Clinical Endocrinology amp Metabolism vol59 no 1 pp 51ndash55 1984
[59] P Ye T Yamashita D M Pollock H Sasano and W ERainey ldquoContrasting effects of eplerenone and spironolac-tone on adrenal cell steroidogenesisrdquo Hormone and MetabolicResearch vol 41 no 1 pp 35ndash39 2009
[60] B D Albertson K L Frederick N C Maronian et al ldquoTheeffect of ketoconazole on steroidogenesis I Leydig cell enzymeactivity in vitrordquo Research Communications in Chemical Patho-logy and Pharmacology vol 61 no 1 pp 17ndash26 1988
[61] B D Albertson N C Maronian K L Frederick et al ldquoTheeffect of ketoconazole on steroidogenesis II Adrenocorticalenzyme activity in vitrordquoResearch Communications in ChemicalPathology and Pharmacology vol 61 no 1 pp 27ndash34 1988
[62] DM Stocco andB J Clark ldquoRegulation of the acute productionof steroids in steroidogenic cellsrdquo Endocrine Reviews vol 17 no3 pp 221ndash244 1996
[63] L K Christenson and J F Strauss III ldquoSteroidogenic acute regu-latory protein (StAR) and the intramitochondrial translocationof cholesterolrdquo Biochimica et Biophysica Acta (BBA)mdashMolecularand Cell Biology of Lipids vol 1529 no 1ndash3 pp 175ndash187 2000
[64] E R Simpson ldquoCholesterol side-chain cleavage cytochromeP450 and the control of steroidogenesisrdquo Molecular and Cel-lular Endocrinology vol 13 no 3 pp 213ndash227 1979
[65] H A LaVoie and S R King ldquoTranscriptional regulation of ster-oidogenic genes STARD1 CYP11A1 and HSD3Brdquo ExperimentalBiology and Medicine vol 234 no 8 pp 880ndash907 2009
[66] F Raucci A DrsquoAniello and M M Di Fiore ldquoStimulation ofandrogen production by d-aspartate through the enhancementof StAR P450scc and 3120573-HSD mRNA levels in vivo rat testisand in culture of immature rat Leydig cellsrdquo Steroids vol 84pp 103ndash110 2014
[67] M Doi Y Takahashi R Komatsu et al ldquoSalt-sensitive hyper-tension in circadian clock-deficient Cry-null mice involves dys-regulated adrenal Hsd3b6rdquo Nature Medicine vol 16 no 1 pp67ndash74 2010
[68] H Okamura M Doi Y Yamaguchi and J-M Fustin ldquoHyper-tension due to loss of clock novel insight from the molecu-lar analysis of Cry1Cry2-deleted micerdquo Current HypertensionReports vol 13 no 2 pp 103ndash108 2011
[69] M Doi F Satoh T Maekawa et al ldquoIsoform-specific mono-clonal antibodies against 3120573-hydroxysteroid dehydrogenaseisomerase family provide markers for subclassification ofhuman primary aldosteronismrdquo The Journal of Clinical Endo-crinology amp Metabolism vol 99 no 2 pp E257ndashE262 2014
[70] S Konosu-Fukaya Y Nakamura F Satoh et al ldquo3120573-hydroxy-steroid dehydrogenase isoforms in human aldosterone-produc-ing adenomardquo Molecular and Cellular Endocrinology vol 408pp 205ndash212 2015
[71] G C Verwoert J Hofland N Amin et al ldquoExpression and genevariation studies deny association of humanHSD3B1 gene withaldosterone production or blood pressurerdquoAmerican Journal ofHypertension vol 28 no 1 pp 113ndash120 2015
[72] D Hough K Storbeck SW P Cloete A C Swart and P SwartldquoRelative contribution of P450c17 towards the acute cortisolresponse lessons from sheep and goatsrdquoMolecular and CellularEndocrinology vol 408 pp 107ndash113 2015
[73] I Sakuma S Suematsu Y Matsuzawa et al ldquoCharacterizationof steroidogenic enzyme expression in aldosterone-producingadenoma a comparison with various human adrenal tumorsrdquoEndocrine Journal vol 60 no 3 pp 329ndash336 2013
[74] A Tong A Jia S Yan Y Zhang Y Xie and G Liu ldquoEctopiccortisol-producing adrenocortical adenoma in the renal hilumhistopathological features and steroidogenic enzyme profilerdquoInternational Journal of Clinical and Experimental Pathologyvol 7 no 7 pp 4415ndash4421 2014
[75] A Hirsch D Hahn P Kempna et al ldquoMetformin inhibitshuman androgen production by regulating steroidogenicenzymes HSD3B2 and CYP17A1 and complex I activity of therespiratory chainrdquo Endocrinology vol 153 no 9 pp 4354ndash43662012
[76] B Xu L Gao Y Cui et al ldquoSET protein up-regulated test-osterone production in the cultured preantral folliclesrdquo Repro-ductive Biology and Endocrinology vol 11 article 9 2013
[77] S Kikuchi D Tominaga M Arita M Takahashi and MTomita ldquoDynamic modeling of genetic networks using geneticalgorithm and S-systemrdquo Bioinformatics vol 19 no 5 pp 643ndash650 2003
[78] N Shikata YMaki M Nakatsui et al ldquoDetermining importantregulatory relations of amino acids from dynamic networkanalysis of plasma amino acidsrdquo Amino Acids vol 38 no 1 pp179ndash187 2010
[79] M Nakatsui T Ueda Y Maki I Ono and M OkamotoldquoMethod for inferring and extracting reliable genetic interac-tions from time-series profile of gene expressionrdquoMathematicalBiosciences vol 215 no 1 pp 105ndash114 2008
[80] A Komori Y Maki M Nakatsui I Ono and M OkamotoldquoEfficient numerical optimization algorithm based on new real-coded genetic algorithm AREX + JGG and application to theinverse problem in systems biologyrdquo Applied Mathematics vol3 no 10 pp 1463ndash1470 2012
[81] J A Roubos G van Straten and A J B van Boxtel ldquoAn evolu-tionary strategy for fed-batch bioreactor optimization conceptsand performancerdquo Journal of Biotechnology vol 67 no 2-3 pp173ndash187 1999
[82] D Sarkar and J M Modak ldquoGenetic algorithms with filters foroptimal control problems in fed-batch bioreactorsrdquo Bioprocessand Biosystems Engineering vol 26 no 5 pp 295ndash306 2004
[83] K Harada K Ikeda J Sakuma I Ono and S KobayashildquoHybridization of genetic algorithm with local search in mul-tiobjective function optimization recommendation of GA thenLSrdquoTransactions of the Japanese Society for Artificial Intelligencevol 21 no 6 pp 482ndash492 2006
Submit your manuscripts athttpwwwhindawicom
PainResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
ToxinsJournal of
VaccinesJournal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AntibioticsInternational Journal of
ToxicologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
StrokeResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Drug DeliveryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Pharmacological Sciences
Tropical MedicineJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AddictionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Emergency Medicine InternationalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Autoimmune Diseases
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anesthesiology Research and Practice
ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Pharmaceutics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
MEDIATORSINFLAMMATION
of
Journal of Toxicology 3
(Sigma-Aldrich Inc St Louis MO) 30 nM sodium selen-ite (Wako Pure Chemical Industries Ltd Osaka Japan)125mgmLbovine serumalbumin (BSA Sigma-Aldrich IncSt Louis MO) 000535mgmL linoleic acid (Sigma-AldrichInc St Louis MO) 25 Nu Serum (Becton Dickinson andCompany Franklin Lakes NJ) 100UmL penicillin (MeijiSeika Pharma Tokyo Japan) and 100mgL streptomycin(Meiji Seika Pharma Tokyo Japan)
212 Adrenal Steroidogenesis in Human Adrenal Corticocar-cinoma NCI-H295R Cells NCI-H295R cells were stimulatedwith adrenocorticotrophic hormone (ACTH) forskolin andangiotensin II to initiate steroidogenesis Changes in steroidconcentrations over time were measured after stimulationin both cells and culture medium to construct a simulationmodel
The cells were seeded at 6 times 105 cellswell in 6-well platesAfter 3 days of culture the culture medium was changed tostimulationmediumconsisting ofDMEMF-12 (1 1)mediumsupplemented with 000625mgmL insulin 000625mgmLtransferrin 30 nM sodium selenite 125mgmL BSA000535mgmL linoleic acid 10 fetal bovine serum(GIBCO Life Technologies Carlsbad CA) 100UmL peni-cillin 100mgL streptomycin 50 nM ACTH (Sigma-AldrichInc St Louis MO) 20120583M forskolin (Sigma-Aldrich Inc StLouis MO) and 100 nM angiotensin II (Calbiochem MerckMillipore Darmstadt Germany) Culture media and cellswere collected at 0 8 24 48 and 72 h after stimulation Thecells were collected in 100120583L distilled water and sonicatedto produce a cell lysate The cultures were conducted in fourwellstime point (119873 = 4)
The concentrations of 12 steroids pregnenolone (PREG)17120572-hydroxypregnenolone (HPREG) dehydroepiandroster-one (DHEA) progesterone (PROG) 17120572-hydroxyprogester-one (HPROG) androstenedione (DIONE) testosterone(TESTO) 11-deoxycorticosterone (DCORTICO) 11-deoxy-cortisol (DCORT) corticosterone (CORTICO) cortisol(CORT) and aldosterone (ALDO) in the medium and celllysate were measured by LCMS Concentrations of estrone(E1) and 17120573-estradiol (E2) were measured by enzyme-linkedimmunosorbent assays (Wako Pure Chemical Industries LtdOsaka Japan) In addition the concentration of cholesterolwas measured using a commercial kit (Wako Pure ChemicalIndustries Ltd Osaka Japan) based on the cholesterol oxi-dase method
213 Liquid Chromatography A LC-VP series (ShimadzuKyoto Japan) consisting of an SIL-HTc autosamplerLC-10ADvp Pump CTO-10ACvp column oven and DGU-14AM degasser was used to set the reverse-phase liquid chro-matographic conditions The column was a Cadenza CD-C18 column (100 times 2mm id 3 120583m Imtakt Corp KyotoJapan) used at 45∘C The mobile phase included waterace-tonitrileformic acid 955005 (vvv Solvent A) andwateracetonitrileformic acid 3565005 (vvv Solvent B)The gradient elution programs were 0 B (0-1min with anisocratic gradient) 0ndash40 B (1-2min with a linear gradient)40 B (2ndash7min with an isocratic gradient) 40ndash100 B(7ndash12min with a linear gradient) 100 B (12ndash14min with
an isocratic gradient) 100ndash0 B (14-15min with a lineargradient) and 0 B (15-16min with an isocratic gradient) ata flow rate of 03mLmin The autosampler tray was cooledto 45∘C and the injection volume was 5120583L HPLC gradeacetonitrile and formic acid were purchased fromWAKO
214 Mass Spectrometry A triple quadrupole mass spec-trometer API4000 (Applied BiosystemsMDS Sciex Con-cord Canada) coupled with an electrospray ionization sourcewas operated in the positive ion mode The optimized ionsource conditions were as follows collision gas 6 psi curtaingas 40 psi ion source gas 1 50 psi ion source gas 2 80 psiion source voltage 5500V ion source temperature 600∘CNitrogenwas used as the collision gas in themultiple reactionmonitoring (MRM) mode The conditions of declusteringpotential collision energy and collision cell exit potentialwere optimized by every steroid The transitions in MRMwere as follows PREG 119898119911 317 rarr 299 HPREG mz 315 rarr297 DHEA mz 289rarr 271 PROG mz 315rarr 109 HPROGmz 331 rarr 109 DIONE mz 287 rarr 97 DCORT mz 331 rarr123 DCORTICOmz 347rarr 161 CORTICOmz 347rarr 100CORT mz 363 rarr 309 ALDO mz 361 rarr 343 and TESTOmz 289 rarr 109 Mass spectroscopic data were acquired andquantified using the Analyst 142 software package (AppliedBiosystemsMDS Sciex Concord Canada)
215 Estimation of the Cell Volume Cell volume was esti-mated from the number of cells in the well and the averagediameter of the cells Cells were detached from the well using0025 trypsin (MP Biomedicals Inc Irvine CA) in a 002EDTA solution (Dojindo Laboratories Kumamoto Japan)at the start of preculture start of stimulation and at 2448 and 72 h after stimulation The numbers and diametersof the cells were measured by a cell counter Vi-cell XR201 (Beckman Coulter Krefeld Germany) after trypan bluestaining Parameters of the cell volume and number of cellswere estimated to fit experimental time-course data usingexponential curves
216 Test Compounds in Validation Study NCI-H295Rcells were exposed to seven well-characterized inhibitors ofsteroidogenesis and then the concentrations of the steroidsin the culture medium were measured to estimate theenzyme inhibition to evaluate the performance of the simula-tion model The adrenal steroidogenic inhibitors includedaminoglutethimide (AGT Bachem AG Bubendorf Switzer-land) op1015840-DDD (DDD Sigma-Aldrich Inc St Louis MO)spironolactone (SP Sigma-Aldrich Inc St Louis MO)metyrapone (MP Sigma-Aldrich Inc St Louis MO) keto-conazole (KC Wako Pure Chemical Industries Ltd OsakaJapan) miconazole (MC Wako Pure Chemical IndustriesLtd Osaka Japan) and daidzein (DZ Sigma-Aldrich IncSt Louis MO) The cells were also exposed to four adrenaltoxicants whose adrenal toxicity is not mediated throughsteroidogenesis inhibition The toxicants were acryloni-trile (AN Wako Pure Chemical Industries Ltd OsakaJapan) salinomycin (SM Sigma-Aldrich Inc St LouisMO) thioguanine (TG Tokyo Chemical Industry Co LtdTokyo Japan) and fumaronitrile (FN Wako Pure Chemical
4 Journal of Toxicology
Industries Ltd Osaka Japan) All chemicals were dissolvedin DMSO (Wako Pure Chemical Industries Ltd OsakaJapan) and added to the culture medium at 1 1000 dilutions
217 Validation Study Using Adrenal Toxicants NCI-H295Rcells were cultured for 3 days in 6-well plates and thenstimulated with the above-mentioned compounds Upon thestart of stimulation various concentrations of test chemicalswere added to the cultures After a further 3 days of culturewith the chemicals the concentrations of 12 steroids (PREGHPREG DHEA PROG HPROG DIONE DCORTICODCORT CORTICO CORT ALDO and TESTO) in theculture medium were measured by LCMSMSThe test con-centrations of the chemicalswere determined by dose-findingcytotoxicity assays The cytotoxicity assay was conducted in96-well plates using ATP content in cells as an endpoint(CellTiter-Glotrade Luminescent Cell Viability Assay Promega)Concentrations that caused more than 20 cytotoxicity werenot used in the steroidogenesis assay The test concentrationsof adrenal steroidogenesis inhibitors and other compoundsare shown in Table 1
218 Statistical Analysis Comparisons were performed bythe two-sample Welchrsquos 119905-test with Bonferroni multiple test-ing correction for each steroid hormone species Statisticallysignificant steroid hormones were considered at adjusted119901 values of less than 001 Differential metabolic steroidprofiles were classified by hierarchical cluster analysis Pair-wise distances between all compounds and all steroids werecalculated by standardized Euclidean metric This distancematrix was analyzed with Wardrsquos method for hierarchicalclustering Statistical analysis was performed usingMATLABsoftware (MathWorks Inc Natick MA)
22 The Computational Part
221 Mathematical Modeling of Adrenal Steroidogenesis inNCI-H295R Cells Steroid hormones secreted from humanadrenal corticocarcinoma NCI-H295R cells are synthesizedfrom cholesterol through the C
21-steroid hormone biosyn-
thesis pathway A mathematical model of adrenal steroido-genesis in NCI-H295R cells was constructed with cholesteroltransport and the intracellular localization pathway theoxysterol synthesis pathway as a bypass of steroidogenesisthe C
21-steroid hormone biosynthesis pathway as the main
steroidogenesis pathway passive transport of steroid hor-mones and cell proliferation (Figure 1) In this model twocompartments the intracellular space and culture mediumwere incorporated as the available region Equations andparameters of the cell proliferation and diffusional trans-port of steroid hormones have been proposed by previousstudies [21 22] Cholesterol transport and the intracellularlocalization pathway including the oxysterol bypass wereintegrated using a part of the ACTH-stimulated cortisolsecretion model described by Dempsher and colleagues [47]The C
21-steroid hormone biosynthesis pathway includes 14
steroid hormones PREG HPREG DHEA PROG HPROGDIONE TESTO DCORTICO DCORT CORTICO CORTALDO E1 and E2 and 17 enzymatic reactions catalyzed by
nine steroidogenic enzymes cholesterol side chain cleavageenzyme (CYP11A1) 17120572-hydroxylase (CYP17H) C
1720-lyase
(CYP17L) 3120573-hydroxysteroid dehydrogenase (HSD3B2) 21-hydroxylase (CYP21A2) 11120573-hydroxylase (CYP11B1) 18-hydroxylase (CYP11B2) 17120573-hydroxysteroid dehydrogenase(HSD17B3) and aromatase (CYP19A1) In this mathematicalmodel of adrenal steroidogenesis inNCI-H295R cells the fluxvelocities ofmolecular transportation and enzymatic reactionrates of steroidogenic enzymes were defined based on thefirst-order reaction and rapid-equilibrium enzyme kineticsrespectively All equations in the mathematical model ofadrenal steroidogenesis of NCI-H295R cells were describedin a supplementary document (see Supplementary Materialavailable online at httpdxdoiorg10115520164041827)The rate constants and the maximum activities were esti-mated by fitting to experimental time-course data of theconcentrations of cholesterol and all steroids Initial valuesof cholesterol and the 14 steroid concentrations were usedin each experimentally measured value and every steroidconcentration was assumed to rapidly reach the equilibriumstate between the culture medium and intracellular space Allfixed values of static parameters and initial values of variableparameters in this model were described in Tables S1 and S2in a supplementary document respectively
222 Modeling and Simulation Environment This compu-tational model of adrenal steroidogenesis in NCI-H295Rcells was developed on the simBio platform which is ageneral environment of biological dynamic simulation andcomputational model development [48] ODEs were solvedby the fourth-order Runge-Kuttamethodwith a variable timestepThe time step (119889119905) was adjusted to refer to themaximumabsolute value of flux velocities or enzymatic reaction ratesat each time point and the range of the time step was from1 times 10minus5 to 10minus2 To confirm whether the range of the timestep was suitable the numerical error ratio was calculated bycertain fixed time steps in the range of the time step whichwas under 1 times 10minus8 in every time step The duration time ofcomputational simulation of adrenal steroidogenesis in NCI-H295R cells was set at 72 h
223 Parameter Optimization To reconstruct experimentaltime-course patterns of the concentrations of cholesterol andthe 14 steroids in the culture medium and intracellular spacewe optimized every rate constant and maximum velocityof the steroidogenic enzymes This parameter optimizationproblem was solved by the Levenberg-Marquardt methodwhich is one of the nonlinear least squares methods [49ndash51] The objective function of optimization was used as thefollowing normalized least squares distance (NLSD)
NLSD = sumℎ
sum
119894
sum
119895
(119883expℎ119894119895minus 119883
simℎ119894119895)
2
119883maxℎ119894
2 (1)
where ℎ is the compartment (culture medium or intracellularspace) 119894 is the molecular species (cholesterol and the 14steroids) 119895 is the time point (0 8 24 48 and 72 h) 119883exp
ℎ119894119895
is the experimentally measured concentration of molecule
Journal of Toxicology 5
Table1Ad
renaltoxicities
andactio
nsof
teste
dcompo
unds
Testchem
ical
Testconcentrations
(120583M)
Pathologicalfeatures
ofadrenaltoxicity
Repo
rted
enzymeinh
ibition
sRe
ferences
Acrylonitrile
(AN)
01110and
100
Hem
orrhagicadrenaln
ecrosis
Not
repo
rted
[25]
Salin
omycin
(SM)
000
00100101
1and10
Dam
agetoadrenalm
edullae
Not
repo
rted
[26]
Thioguanine(TG
)001110and100
Hem
orrhagicadrenaln
ecrosis
Not
repo
rted
[25]
Fumaron
itrile
(FN)
01110and
100
Hem
orrhagicadrenaln
ecrosis
Not
repo
rted
[25]
Aminoglutethim
ide(AG
T)01110and
100
Hypertro
phyvascular
degeneratio
nCY
P11A
1CY
P21A
2CY
P11B1andCY
P11B2
[627ndash30]
op1015840-D
DD(D
DD)
0111025and100
Atroph
yCY
P11A
1HSD
3B2CY
P21A
2CY
P11B1andCY
P11B2
[293132]
Spiro
nolacton
e(SP
)11050and100
Hypertro
phy
CYP17H
CYP
17LCY
P11B1andCY
P11B2
[633ndash35]
Metyrapon
e(MP)
01110and
100
Hypertro
phyvascular
degeneratio
nCY
P11A
1CY
P11B1andCY
P11B2
[62936ndash39]
Ketoconazole(KC)
01110and
100
Hypertro
phy
CYP11A
1CY
P17H
CYP
17LHSD
3B2CY
P21A
2and
CYP11B1
[62940
ndash43]
Micon
azole(MC)
011102550and
100
Hypertro
phy
CYP11A
1CY
P17H
CYP
17LCY
P21A
2andCY
P11B1
[4144
45]
Daidzein(D
Z)01110and
100
Unk
nown
HSD
3B2andCY
P21A
2[46]
6 Journal of Toxicology
CHOC
PREG
PROG
DCORTICO
CORTICO
ALDO
CORT
DCORT
HPROG
HPREG DHEA
DIONE
TESTO
CYP11A1
E1
E2
HSD3B2
CYP1
7L
CYP21A2 HSD17B3
CYP1
9A1
CYP11B2
OXYOxysterol synthesis
PREGHPREG DHEAPROG
HPROG DIONE E1DCORTICODCORT TESTO E2CORTICO
CORTALDO
NCI-H295R cell
Medium culture
CYP11B1
CHOL
StAR
CHOM
CHOSCEH
CHOR
CHOLCholesterol transport
Total intracellular cholesterol
CYP1
7H
Vloc Vacc
VMTR
VCEH
Figure 1 Schematic diagram of the mathematical model of adrenal steroidogenesis in NCI-H295R cells Overview of the mathematicalmodel of adrenal steroidogenesis in NCI-H295R cells including cholesterol transport and intracellular localization oxysterol synthesisthe C
21-steroid hormone biosynthesis pathway passive diffusional transport of steroid hormones and cell proliferation CHOL total
cholesterol in medium culture CHOS stored cholesterol esters in the endoplasmic reticulum CHOC intracellular free cholesterolCHOM mitochondrial free cholesterol CHON mitochondrial free cholesterol close to CYP11A1 enzymes CHOR mitochondrial freecholesterol remote from CYP11A1 enzymes PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosteronePROG progesterone HPROG 17120572-hydroxyprogesterone DIONE androstenedione DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICO corticosterone CORT cortisol ALDO aldosterone TESTO testosterone E1 estrone E2 17120573-estradiol OXYoxysterol CEH cholesterol ester hydrolase StAR steroidogenic acute regulatory protein CYP11A1 P450 side chain cleavage enzymeCYP17H 17120572-hydroxylase of CYP17 CYP17L C
17ndash20 lyase of CYP17 HSD3B2 3120573-hydroxysteroid dehydrogenase CYP21A2 21-hydroxylaseCYP11B1 11120573-hydroxylase CYP11B2 18-hydroxylase HSD17B3 17120573-hydroxysteroid dehydrogenase and CYP19A1 aromatase
119894 in compartment ℎ at time point 119895 119883simℎ119894119895
is the simulatedconcentration of molecule 119894 in compartment ℎ at time point119895 and 119883max
ℎ119894is the maximum concentration of molecule 119894 in
compartment ℎ over all time points Data points under thelower quantitation limit were excluded from the evaluationby the objective function
Effects of every static model parameter for parameteroptimization were calculated from differences of fitting theobjective function using sensitivity analysis
224 Quantitative Estimation of the Mechanism of Action ofAdrenal Toxicants Metabolic steroid profiling and differen-tial patterns of the adrenal steroid hormones by chemicalperturbation were reconstructed to optimize the relativeactivities of the steroidogenic enzymes The input datafor the quantitative mechanistic analysis of adrenal toxiccompounds was a fold change (ratio) of the measured 12
steroid concentrations induced by drug exposure for 72 hThe two-step optimization method of the real-coded geneticalgorithm (RCGA) was adopted as a global optimizationmethod in the quantitative mechanistic analysis of adrenaltoxic compoundsThe operations of the crossover and gener-ation alterationmodel in RCGAwere used for the real-codedensemble crossover (REX) and just generation gap (JGG)[52ndash55] As the initial parameters of RCGA maximum gen-eration population size selection size of parent individualspopulation size of child individuals and termination criteriawere 1000 100 6 25 and under 01 of NLSD respectivelyThe search space for the relative activities of the steroidogenicenzymes was from 1100 to 100 To evaluate the fitness ofeach individual the sum of squared residuals for fold changesof measured 12 steroid concentrations was used as theobjective function Nonlinear least squares optimization bythe Levenberg-Marquardt method was used as a local
Journal of Toxicology 7
search [49ndash51] As the estimated mechanisms of actionsof the adrenal toxic compounds the relative activities ofeight steroidogenic enzymes (CYP11A1 CYP17H CYP17LHSD3B2 CYP21A2 CYP11B1 CYP11B2 and HSD17B3)were optimized by the above-mentioned 2-step optimizationmethod Every optimization calculation was duplicated tocheck the numerical stability of the optimal parameters
225 Global Dynamic Sensitivity Analysis The propertyof every kinetic parameter in this computational modelof steroidogenesis in NCI-H295R cells was evaluated bydynamic sensitivity analysis The sensitivity (119878
119909119910) of kinetic
parameter 119909 for variable parameter 119910 was defined by the fol-lowing equation
119878119909119910(119905)=
Δ119910(119905)119910(119905)
Δ119909119909 (2)
where variable parameter119910was the concentration of a steroidhormone in the cytosolic space of NCI-H295R cells Theperturbation for kinetic parameters was +10 (Δ119909119909 = 01)
3 Results
31 Experimental Data on Adrenal Steroidogenesis
311 Adrenal Steroidogenesis of NCI-H295R Cells and theMass Balance All steroid hormones in the culture mediumwere significantly increased after 72 h of stimulation with50 nM ACTH 20120583M forskolin and 100 nM angiotensinII (Figure 2(a)) Mass balances in steroidogenesis of NCI-H295R cells under nontreatment and control (stimulated)conditions are shown in Figures 2(b) and 2(c) respectivelyUnder stimulation the dynamics of net mass in these exper-iments were unchanged and accumulated cholesterol wasconverted to adrenal steroids
312 Cytotoxicity of Adrenal Toxicants Viabilities of cellstreated with each compound were expressed as a relativevalue to the ATP level of the control Effects of AN SM TGFN AGT DDD SP MP KC MC and DZ on cell viabilitywere determined to be valid under 80 of the relative ATPlevel at 7 days after treatment AN SM TG and FN showedcytotoxicity at over 100 1 10 and 10120583M respectively AGTMP andDZ did not affect cell viability at up to 100 120583MDDDSP KC andMC induced less than 80 of cell viability at over100 50 100 and 25 120583M respectively
313 Differentially Steroid Profiling of Adrenal ToxicantsAfter NCI-H295R cells were exposed to each test compoundduring three days the concentrations of 12 steroid hor-mones in the culture mediumwere simultaneously measuredby LCMSMS All effects of the compounds on adrenalsteroidogenesis were evaluated at the concentration withoutany overt cytotoxicity The differential metabolic steroidprofiles of 11 adrenal toxic compounds were classified andvisualized by using hierarchical clustering analysis (Figure 3)
Four adrenal toxicants without steroidogenic inhibitionAN SM TG and FN did not change the medium concen-trations of all steroid hormones by more than 2-fold Above-mentioned 4 compounds at every condition and 7 adrenalsteroidogenic inhibitors at the low exposure concentrationwere gathered into a big cluster as nonchange group The7 steroidogenic inhibitors at the maximum exposure con-centration showed the characteristic steroid profiles eachbut 100 120583M DZ and 10 120583M SP were classified as a clusterAGT drastically decreased the medium concentrations ofPREG HPREG DHEA PROG DCORTICO CORTICOand ALDO at 100 120583M DDD dose-dependently decreased themedium concentrations of PROG DCORTICO CORTICOCORT and ALDO at gt10 120583M and decreased PREG HPREGDHEA PROG HPROG DIONE and DCORT at the maxi-mum exposure concentration of 25120583M SP increased PREGHPREG and DHEA and decreased PROG DIONE DCOR-TICODCORTCORTICOALDO andTESTOat 10120583MMPdose-dependently decreased CORTICO CORT and ALDOand decreased DHEA HPROG DIONE and TESTO at themaximum exposure concentration of 100 120583M KC drasticallydecreased the medium concentrations of PREG HPREGDHEAHPROGDIONEDCORTICODCORTCORTICOCORTO ALDO and TESTO at 10 120583M MC increased themedium concentrations of PROG and decreased DIONEDCORT CORT and TESTO at 10 120583M DZ increased PREGHPREG and DHEA and decreased DIONE DCORTICODCORT CORTICO CORT ALDO and TESTO at 100 120583M
32 The Mathematical Modeling
321 Optimization of the Mathematical Model of AdrenalSteroidogenesis in NCI-H295R Cells The mathematicalmodel of adrenal steroidogenesis in NCI-H295R cells wasoptimized for several kinetic parameters of cholesteroltransport intracellular localization the oxysterol pathwayand maximum velocity of steroidogenic enzymes to fit theexperimental time-course data All optimized kinetic para-meters are shown in Table S1 in a supplementary documentThe optimized mathematical model was reconstructedwith the experimental dynamic patterns of cholesteroland the 14 steroid hormones in the intracellular space andculture medium The fitness was 0621761 of NLSD valuesas the fitting objective function The simulation results andexperimental data are shown in Figure 4
Optimized kinetic parameters were calculated sensitivi-ties for the NLSD value as the fitting score and are shown inTable S1 in a supplementary document The highly sensitiveparameters for fitting the NLSD score were the extractednine kinetic parameters 119896Cholesterol Transport 119896
acc119891 119896loc119891 119896loc119887
119881CYP11A1max 119870CYP11A1
119898 119881CYP17HmaxA 119881HSD3B2
maxA and 119881CYP21A2maxA whichhad higher than 30 fitting sensitivity
33 The Validation Using the Adrenal Toxicants
331 Mechanistic Analysis of Adrenal Toxicants Effects ofadrenal toxic compounds on steroidogenic enzymes werequantitatively predicted from the change in the ratio ofthe measured medium concentrations of the 12 steroid
8 Journal of Toxicology
0001
001
01
1
10
PREG
HPR
EG
DH
EA
PRO
G
HPR
OG
AN
DRO
DCO
RTIC
O
DCO
RT
CORT
ICO
CORT
ALD
O
TEST
O E1 E2C
once
ntra
tion
in m
ediu
m (120583
M)
lowast
lowast
lowast
lowast
lowast
lowast
lowast lowast lowastlowast
lowast
lowast
lowast
lowast
(a)
0 8 24 48 72Time (hours)
0
40
80
120
160
200
240
Num
ber o
f mol
ecul
es (n
mol
)
(b)
0 8 24 48 72Time (hours)
0
40
80
120
160
200
240
Num
ber o
f mol
ecul
es (n
mol
)
(c)
Figure 2 Experimental data of metabolic profiling of adrenal steroid hormones and the mass balance Concentrations of steroid hormonessecreted from NCI-H295R cells in the culture medium at 72 h after stimulation were compared with the untreated condition and thestimulated condition by 50 nM ACTH 20 120583M forskolin and 100 nM angiotensin II (a) Net molecular amounts including cholesterol andsteroid hormones in the culture medium and intracellular space are plotted at five time points (0 8 24 48 and 72 h after stimulation)under the untreated condition (b) and the stimulated condition (c) In the bar graphs dark and light bars indicate the amount of cholesteroland adrenal steroids respectively All data are shown as the mean plusmn SD (119873 = 4) lowast119901 values corrected by the familywise error ratelt001 PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-hydroxyprogesteroneDIONE androstenedioneDCORTICO 11-deoxycorticosteroneDCORT 11-deoxycortisol CORTICO corticosteroneCORT cortisol ALDO aldosterone and TESTO testosterone
hormones at 72 h after drug exposure using themathematicalmodel of adrenal steroidogenesis in NCI-H295R cells Thereproducibility of the estimated results was confirmed byperforming the test twice The estimated effects of 11 adrenaltoxic compounds on eight steroidogenic enzymes are shownin Figure 5 The adrenal toxic compounds without steroido-genic inhibition such as vasculotoxic agents (AN SM TGand FN) were not estimated for the target steroidogenicenzymes under noncytotoxic conditions Every fitness valueswere under 005 of NLSD values used as the fitting objectivefunction (Figures 5(a)ndash5(d)) Other steroidogenic inhibitors(AGT DDD SPMP KCMC andDZ) are described in detailbelow
332 AGT The mechanism of action of AGT in adrenalsteroidogenesis was estimated by inhibition of CYP11A1HSD3B2 CYP21A2 and CYP11B1 at 100 120583M (estimatedinhibitions were 770 780 811 and 598 resp)(Figure 5(e)) AGT has been reported to inhibit CYP11A1
CYP21A2 CYP11B1 and CYP11B2 [6 27ndash30] Our resultswere mostly consistent with the previous reports In particu-lar CYP11A1 appeared to be inhibited strongly byAGT In ourstudy HSD3B2 inhibition of AGT was shown by mechanisticanalysis based on systems biology approaches as a novelmechanism of action of AGT Inhibition of AGT by CYP11B2was not estimated in our study However the concentrationof ALDO in the culture medium decreased to 38 ofthe normal stimulated condition Inhibition of AGT byCYP11B2 has been shown using sheep adrenal homogenatesas well as a human adrenal homogenate from a patientwith Cushingrsquos syndrome [30]The activity of 18-hydroxylaseinduced by CYP11B2 was determined as the conversion ofcorticosterone to 18-hydroxycorticosterone in the previousstudy The cause of the discrepancy regarding the effect ofAGT on CYP11B2 was suggested to be substrate inhibitionbecause the intracellular concentration of CORTICO wasincreased by over 10 times of that in the culture mediumto reach 50 120583M Another possibility was poor quantitative
Journal of Toxicology 9
log2
ratio
4
3
2
1
0
minus1
minus2
minus3
minus4
ALDO
CORTICO
CORT
TESTO
DIONE
HPROG
DCORTICO
DCORT
PROG
PREG
HPREG
DHEA
DZ(
100)
SP(1
0)SP
(1)
DZ(
10)
DD
D(0
1)
AN
(01
)A
N(1
)A
N(1
0)SM
(00
1)TG
(00
1)M
P(0
1)FN
(01
)TG
(1)
AGT(
1)AG
T(0
1)SM
(00
0001
)D
DD
(1)
SM(0
1)
AGT(
10)
MP(
1)FN
(1)
MC(
1)D
Z(1)
KC(0
1)
MC(
01)
DZ(
01)
KC( 1
)D
DD
(10)
MP(
10)
MC(
10)
MP(
100)
AGT(
100)
DD
D(2
5)KC
(10)
Figure 3 Hierarchical cluster analysis of differential metabolic profiles of 12 steroid hormones by exposure to adrenal toxicity compoundsAdrenal toxicants were classified by using the differential metabolic profiling of 12 steroid hormones Concentrations of 12 adrenalsteroids secreted from NCI-H295R cells were drastically changed by the exposure of adrenal steroidogenesis inhibitors The 12 adrenalsteroid hormones were quantitatively measured by LC-MSMS simultaneously Four adrenal vasculotoxic compounds acrylonitrile ANfumaronitrile FN salinomycin SM and thioguanine TG were used as negative control compounds for adrenal steroidogenesis inhibitorsSeven adrenal steroidogenesis inhibitors aminoglutethimide AGT op1015840-DDD DDD spironolactone SP metyrapone MP ketoconazoleKC miconazole MC and daidzein DZ showed a characteristic steroid profile each and were classified as each independent singletonat the maximum exposure condition Exposure concentrations of adrenal toxic compounds were described in brackets of the samplename and the units were prepared as 120583M A blue cluster was classified as a group of nonchange samples including negative controlcompounds and low exposure conditions of adrenal steroidogenesis inhibitors PREG pregnenolone HPREG 17120572-hydroxypregnenoloneDHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-hydroxyprogesterone DIONE androstenedione DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICO corticosterone CORT cortisol ALDO aldosterone and TESTO testosterone
reliability of the experimental data because the ALDOconcentration was under the lower limit of quantification at100 120583MAGT Hecker and colleagues reported that 3 120583MAGTdecreases PREG and PROG concentrations and increases theTESTO concentration [10] However AGT did not increasethe TESTO concentration in our study One possibility isthat the concentration of TESTO was already enhanced byabout 33-fold through stimulation with ACTH forskolinand angiotensin II
333 DDD The mechanism of action of DDD in adrenalsteroidogenesis was estimated by dose-dependent inhibitionof CYP11A1 HSD3B2 CYP21A2 and CYP11B1 (estimatedinhibitions at 25120583M were 870 869 769 and 849resp) (Figure 5(f)) DDD has been reported to inhibitCYP11A1 HSD3B2 CYP21A2 CYP11B1 and CYP11B2 [29 3132] Inhibition of DDD by CYP11B2 was not estimated in ourstudy However the concentration of ALDO in the culturemedium decreased to 3 of that in the normal stimulatedcondition Inhibition of DDD by CYP11B2 has been shownusing mitochondrial and microsomal fractions prepared bystandard centrifugation procedures from a bovine adrenal
cortex homogenate [32]The cause of the discrepancy regard-ing the inhibition ofDDDbyCYP11B2 could not be explainedby same effect in the case of AGT
334 SP Themechanism of action of SP in adrenal steroido-genesis was estimated by inhibition of HSD3B2 CYP21A2and HSD17B3 (estimated inhibitions at 10120583M were 702595 and 593 resp) (Figure 5(g)) SP has been reportedto inhibit CYP17H CYP17L CYP11B1 and CYP11B2 [6 33ndash35] The inhibitory effect of SP on the HSD3B2 enzyme wasa novel mechanism of action The main action of SP is asa mineralocorticoid receptor (MR) antagonist SP has alsobeen reported to exert some off-target effects by binding toandrogen glucocorticoid and progesterone receptors [56ndash58] SP has been shown to inhibit the production of ALDOand CORT from PREG induced by angiotensin II in H295Rcells but the specific MR antagonist eplerenone did not showthe inhibitory effects [59] Therefore HSD3B2 inhibition bySP is not mediated via MR and the action might be directinhibition of HSD3B2 enzymes or a part of known off-targeteffects mediated through other nuclear hormone receptorsRegarding CYP17H and CYP17L our results were consistent
10 Journal of Toxicology
CHOL
0
20
40
60
80
100C
once
ntra
tion
in m
ediu
m (120583
M)
10 20 30 40 50 60 70 800Time (hours)
(a)
PREG
DCORT
DCORTICO
CORTICO
CORT
Con
cent
ratio
n in
med
ium
(120583M
)
10
4
2
0
8
6
10 20 30 40 50 60 70 800Time (hours)
(b)
Con
cent
ratio
n in
med
ium
(120583M
)
0
020406
081
1214
0 10 20 30 40 50 60 70 80Time (hours)
HPREG
DIONE
PROGHPROG
E1
(c)
DHEAE2
ALDO
TESTO
0
005
01
015
02
025
Con
cent
ratio
n in
med
ium
(120583M
)
0 10 20 30Time (hours)
40 50 60 70 80
(d)
CHOL
0
5000
10000
15000
20000
25000
30000
Con
cent
ratio
n in
cell
(120583M
)
10 20 30 40 50 60 70 800Time (hours)
(e)
Con
cent
ratio
n in
cell
(120583M
)
0
50
100
150
200
250
0 10 20 30 40 50 60 70 80Time (hours)
PREG
DCORT
DCORTICO
CORTICOCORT
(f)
0 10 20 30 40 50 60 70 80Time (hours)
HPREG
DIONE
PROGHPROG
E1
Con
cent
ratio
n in
cell
(120583M
)
3025201510
35
05
(g)
005
115
2
253
35
Con
cent
ratio
n in
cell
(120583M
)
0 10 20 30 40 50 60 70 80Time (hours)
DHEA
E2ALDO
TESTO
(h)
Figure 4 Comparison of time-course profiles of cholesterol and adrenal steroids produced by NCI-H295R cells between experimentallymeasured and simulated data To intuitively confirm reconstruction of the measured experimental data in the developed simulation modelof NCI-H295R cells dynamics of cholesterol and adrenal steroids produced by NCI-H295R cells were plotted to overlay experimentaldata with the simulated results Graphs show the dynamics of medium concentrations of cholesterol (a) and adrenal steroids ((b)ndash(d))and intracellular concentrations of cholesterol (e) and adrenal steroids ((f)ndash(h)) Steroid hormones were categorized into three groups byconcentration levels Major steroids were PREG pregnenolone DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICOcorticosterone and CORT cortisol ((b) and (f)) Moderate steroids were HPREG 17120572-hydroxypregnenolone PROG progesterone HPROG17120572-hydroxyprogesterone DIONE androstenedione and E1 estrone ((c) and (g)) Minor steroids were DHEA dehydroepiandrosteroneALDO aldosterone TESTO testosterone and E2 17120573-estradiol ((d) and (h)) Experimental data are shown as symbols with dotted lines Alldata represent the mean plusmn SD (119873 = 4) Simulation data are shown as solid lines
Journal of Toxicology 11
0110
10
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
AN (120583M)
(a)
0110
FN (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(b)
00101
SM (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
100E minus 05
(c)
0110
TG (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B1
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(d)
0110
10100
AGT (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(e)
0110
1025
DDD (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(f)
1010
SP (120583M)
2750005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(g)
0110
10100
MP (120583M)
2110005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(h)
Figure 5 Continued
12 Journal of Toxicology
0110
10
KC (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(i)
0110
10
MC (120583M)
4430005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(j)
0110
10100
DZ (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(k)
Figure 5 Estimated mechanism of action of adrenal toxicants by using the mathematical model of adrenal steroidogenesis in NCI-H295Rcells Mechanisms of action of adrenal toxicity compounds were quantitatively estimated from experimental results of differential steroidprofiling using the mathematical model of adrenal steroidogenesis in NCI-H295R cells The drug action was defined as a scaling factorof enzymatic activity in the simulation model These scaling factors were optimized to fit experimental data by a hybrid optimizationmethod of the RCGA and nonlinear least squares Estimated drug actions by the exposure of vasculotoxic agents acrylonitrile AN (a)fumaronitrile FN (b) salinomycin SM (c) and thioguanine TG (d) and the steroidogenic inhibitors aminoglutethimide AGT (e) op1015840-DDD DDD (f) spironolactone SP (g) metyrapone MP (h) ketoconazole KC (i) miconazole MC (j) and daidzein DZ (k) are shownas a spider radar chart CYP11A1 P450 side chain cleavage enzyme CYP17H 17120572-hydroxylase of CYP17 CYP17L C
17ndash20 lyase of CYP17HSD3B2 3120573-hydroxysteroid dehydrogenase CYP21A2 21-hydroxylase CYP11B1 11120573-hydroxylase CYP11B2 18-hydroxylase and HSD17B317120573-hydroxysteroid dehydrogenase
with previous reports [33 34] 7120572-thiospironolactone whichis synthesized by deacetylation of SP inhibits CYP17H andCYP17L [34] The fact that there were no inhibitions ofCYP17HorCYP17L in our study suggests that SPmight not bedeacetylated to 7120572-thiospironolactone in NCI-H295R cellsRegarding CYP11B1 and CYP11B2 our results were unclearcompared with a previous study It has been shown that30 120583M SP inhibits CYP11B1 and CYP11B2 in human andbovine adrenal mitochondria [35]The cause of CYP11B1 andCYP11B2 inhibition by SP could not be determined in ourstudy which might be due to the lower maximum exposureconcentration of SP than that in the previous reportWe couldnot examine SP concentrations over 10 120583M because theseconcentrations were cytotoxic in NCI-H295R cells
335 MP The mechanism of action of MP in adrenalsteroidogenesis was estimated by dose-dependent inhibitionof CYP11B1 (estimated inhibitions at 1 10 and 100 120583M were571 827 and 982 resp) (Figure 5(h)) MP has beenreported to inhibit CYP11B1 as its major effect and CYP11A1and CYP11B2 as a weak effect [6 29 36ndash39] The results wereable to show that MP is a selective inhibitor of CYP11B1 inthe previous report However the estimated effect of MPat a high concentration 100 120583M as the maximum exposureconcentration was unclear According to the previous reportselectivity of MP for CYP11B1CYP11B2 is about five times[39] In addition 20 120583M MP has a slight inhibition effect onCYP11A1 in H295R cells [29]
336 KC The mechanism of action of KC in adrenalsteroidogenesis was estimated by inhibition of CYP11A1CYP17H CYP17L HSD3B2 CYP21A2 CYP11B1 and
CYP11B2 (estimated inhibitions at 10 120583Mwere 926 943518 830 882 974 and 798 resp) (Figure 5(i))KC has been reported to inhibit CYP11A1 CYP17H CYP17LHSD3B2 CYP21A2 and CYP11B1 [6 29 40ndash43] Our resultswere almost consistent with the previous reports KC inhibitsCYP11A1 CYP17H CYP21A2 and CYP11B1 in NCI-H295Rcells at 10 120583M [29] and CYP17H CYP17L CYP21A2 andCYP11B1 in human adrenal mitochondria and Leydig cellmicrosomes at 2ndash5120583M [60 61] However KC has shownonly weak inhibition of HSD3B2 and HSD17B3 in Leydigcells at the millimolar level [60 61] Regarding CYP11B2 andHSD17B3 we considered that these estimated inhibitions ofKC did have sufficient reliability in terms of quantitative pre-diction precision because ALDO and TESTO concentrationswere less than the lower limit of quantification at 10 120583MKC
337 MC Themechanism of action of MC in adrenal stero-idogenesis was estimated by inhibition of CYP17H CYP17LCYP11B1 and HSD17B3 (estimated inhibitions at 10120583Mwere691 530 764 and 571 resp) (Figure 5(j)) MC hasbeen reported to inhibit not only CYP17H and CYP17L butalsoCYP11A1 CYP21A2 andCYP11B1 [41 44 45]The resultsin the previous reports were able to estimate that MC isa CYP17 inhibitor However CYP11A1 inhibition by MCprobably instead of a reduction in StAR expression wasnot clearly detected in our study using NCI-H295R cellsbecause there were no decreases in the concentrations ofPREG and PROG in the culture medium Indirect inhibitionof CYP11A1 via the peripheral-type benzodiazepine receptorhas been reported in mouse adrenocortical Y-1 cells treatedwith MC in the absence of stimuli by measuring PREGproduction [44] On the other hand reductions of StAR
Journal of Toxicology 13
protein expression andor transport activity without affectingtotal steroid synthesis or CYP11A1 and HSD3B2 enzymeexpression or activities have been reported in (BU)
2cAMP-
stimulated MA-10 Leydig tumor cells treated with MC bymeasuring PROG production [45] Therefore the effect ofMC on the initial reaction in adrenal steroidogenesis fromcholesterol should be different according to the cell type andstimulation condition Inhibition of CYP21A2 and CYP11B1by MC has been reported as decreases in the consumptionof PROG and DCORTICO respectively [41] Inhibition ofCYP11B1 was estimated by the action of MC in this studybut that of CYP21A2 was not detected In the previousexperimental report inhibitory sites by MCmight have beenreflected by inhibition of CYP17H activity because CYP21A2activity was measured as a decrease in labeled PROG
338 DZ The mechanism of action of DZ in adrenalsteroidogenesis was estimated by inhibition of CYP11A1HSD3B2 CYP21A2 CYP11B1 and HSD17B3 (estimated inhi-bitions at 100 120583M were 586 941 965 872 and981 resp) (Figure 5(k)) DZ has been reported to inhibitHSD3B2 and CYP21A2 [46] The results of HSD3B2 andCYP21A2 were consistent with the previous report Howeverinhibitions have not been reported for CYP11A1 CYP11B1and HSD17B3These estimated effects of DZ on CYP11B1 andHSD17B3 were unclear because the concentrations of ALDOand TESTO were less than the lower limit of quantificationat 100 120583M In addition these enzymes act downstream of thestrong action points of DZ such as HSD3B2 and CYP21A2
34 The Simulations and the Systematic Model Analysis
341 Dynamic Sensitivity Analysis of Adrenal SteroidogenesisTo comprehensively understand the dynamics of adrenalsteroidogenesis dynamic sensitivities were calculated forsteroid concentrations secreted by NCI-H295R cells usingour constructed mathematical model of steroidogenesis Theresults of dynamic sensitivity analysis at 72 h of duration and6 h of interval time are presented as a heat-map in Figure 6
The top 10 parameters of the total area under the curveof dynamic sensitivity for cholesterol and the 14 steroids inculture medium were 119881HSD3B2
maxA 119881CYP11A1max 119881CYP21A2maxA 119896loc119891 119896loc119887
119870CYP11A1119898
119881CYP17HmaxA 119896Cholesterol Transport 119896acc119891 and 119881CYP21A2maxB in
order from the top Cholesterol uptake (119896Cholesterol Transport)StAR protein (119896loc
119891) and CYP11A1 (119881CYP11A1max ) which are
determining factors of the capacity for steroidogenesis pro-moted the production of mineralocorticoids (DCORTICOCORTICO and ALDO) and restrained the synthesis of glu-cocorticoids (DCORT and CORT) and sex steroids (DIONETESTO and E1) because of the accumulation of interme-diate molecules in steroidogenesis (PREG HPREG DHEAPROG and HPROG) only by self-activation The dynamicpatterns of the intermediate molecules in steroidogenesiswere mainly dependent on the activity of CYP17H andHSD3B2 with PREG as the substrate of these enzymes inwhich the dynamic sensitivities of 119881CYP17HmaxA for HPREG andHPROG and 119881HSD3B2
maxA for PROG HPROG and DCORTICO
reversed the direction of sensitivity at 49ndash66 h after stimu-lation The dynamic sensitivities of the maximum activitiesof HSD3B2 for PREG (119881HSD3B2
maxA ) and CYP21A2 for PROG(119881CYP21A2maxA ) were related to all steroids at 72 h Almost allmodel parameters had positive sensitivity for downstreamsteroids in the adrenal steroidogenic pathway and negativesensitivity for direct-binding steroids as substrates of thesteroidogenic enzyme The sensitivity of 119881max in all steroido-genic enzymes was relatively higher than 119870
119898for the same
steroid substrate
342 Simulation of the Metabolic Balance of AdrenalSteroidogenesis Pathway To clearly show the property of themetabolic shift between mineralocorticoid and glucocorti-coid biosynthesis we performed two-dimensional parameterscanning of the enzymatic activities of CYP17H and HSD3B2(Figure 7) NCI-H295R cells lost the ability to produce allsteroid hormones when enzymatic activities of CYP17H andHSD3B2 were changed by over 60 and 30 respectivelyActivation ofCYP17HandorHSD3B2 induced themetabolicshift that enhanced the glucocorticoid biosynthesis and devi-ated from the mineralocorticoid biosynthesis On the otherhand inhibition of CYP17H andor HSD3B2 induced themetabolic shift that enhanced the mineralocorticoid biosyn-thesis and deviated from the glucocorticoid biosynthesisMoreover the enzymatic activity of HSD3B2 regulated themetabolic balance of sex steroids and the precursors onadrenal steroidogenesis of NCI-H295R cells E1 TESTO andDIONE were produced by NCI-H295R cells when activatingthe enzymatic activity of HSD3B2 Conversely E2 andDHEAwere produced by NCI-H295R cells when suppressing theenzymatic activity of HSD3B2 The biosynthesis of down-stream steroids in adrenal steroidogenesis pathway such asmineralocorticoids and glucocorticoids was almost com-pletely terminated when the enzymatic activity of HSD3B2was decreased by over 80
4 Discussion
41 Importance of 3120573-HSDActivity in Adrenal SteroidogenesisOur systematic analysis using the mathematical model ofadrenal steroidogenesis in NCI-H295R cells revealed thatthe enzymatic activity of 3120573-HSD controls the dynamics ofadrenal steroidogenesis The activity of the StAR proteincontrols the net capacity of steroidogenesis in steroidogeniccells which is the transport of cholesterol from the outer toinner mitochondrial membranes Both the expression levelsof StAR protein and mRNA are rapidly elevated in responseto stimulation by tropic hormones such as ACTH [62 63]Another important factor in adrenal steroidogenesisis is thecholesterol side chain cleavage enzyme CYP11A1 the firstrate-limiting and hormonally regulated step in the synthesisof all steroid hormones which is conversion of cholesterolto pregnenolone in mitochondria [64] According to ourresults of global sensitivity analysis (Supplementary Table 1and Figure 6(d)) in addition to CYP11A1 and StAR proteins3120573-HSD was one of the key regulators in adrenal steroido-genesis of NCI-H295R cells And also this result suggeststhat a significant regulatory mechanism in steroidogenesis
14 Journal of Toxicology
60544830 4236 66 7218 246 120
Time (hours)
0
1
2
3C
once
ntra
tion
(a)
60544830 4236 66 7218 246 120
Time (hours)
minus05
0
05
minus1
1
Sens
itivi
ty
(b)
0
6
12
18
24
30
36
42
48
54
60
66
72
Sens
itivi
ty fo
r med
ium
conc
entr
atio
ns
of ch
oles
tero
l or s
tero
id h
orm
ones
Dur
atio
n tim
e (ho
urs)
Model parameters
+10
+05
00
minus05
minus10
(c)
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
Parameter number in the mathematical model of steroidogenesis in NCI-H295R cells
CHOL
PREG
HPREG
DHEA
PROG
HPROG
DIONE
DCORTICO
DCORT
CORTICO
CORT
ALDO
TESTO
E1
E2
Chol
este
rol o
r ste
roid
hor
mon
es
0
05
10
15
20
minus05
minus10
minus15
minus20
(d)
Figure 6 Heat-map of the global dynamic sensitivity analysis of adrenal steroid concentrations produced by NCI-H295R cells The globaldynamic sensitivity analysis is a powerful tool to comprehensively understand the dependencies of themodel parameters in themathematicalmodel of the biological complex system Dynamic sensitivities of model parameters in the mathematical model of adrenal steroidogenesisin NCI-H295R cells were calculated for all steroid concentrations in the culture medium every 6 h until 72 h after stimulation To clarify theview of heat-map of global dynamic sensitivity analysis imaginary data of the dynamics of steroid concentrations in the original model (blueline) and perturbed model for sensitivity analysis (red line) were prepared (a) Using this imaginary data the calculated dynamic sensitivities(b) and the visualized dynamic sensitivity as one block of the heat-map (c) were shown respectively By the same method that explainedusing imaginary data the large-scale data of the global dynamic sensitivity analysis on the mathematical model of adrenal steroidogenesisin NCI-H295R cells was comprehensively visualized as a big graph of heat-map (d) Parameter numbers in the horizontal axis are (1)119896Choresterol Transport (2) 119896CEH (3) 119896MTR
119891 (4) 119896MTR
119887 (5) 119896acc
119891 (6) 119896acc
119887 (7) 119896loc
119891 (8) 119896loc
119887 (9) 119896Oxysterol Synthesis (10)119870
CYP11A1119898
(11) 119881CYP11A1max (12)119870CYP17H119898A
(13) 119870CYP17H119898B (14) 119881CYP17HmaxA (15) 119881CYP17HmaxB (16) 119881CYP17HmaxB (17) 119870CYP17L
119898B (18) 119881CYP17LmaxA (19) 119881CYP17LmaxB (20) 119870HSD3B2119898A (21) 119870HSD3B2
119898B (22) 119870HSD3B2119898C (23)
119881HSD3B2maxA (24) 119881HSD3B2
maxB (25) 119881HSD3B2maxB (26) 119870CYP21A2
119898A (27) 119870CYP21A2119898B (28) 119881CYP21A2maxA (29) 119881CYP21A2maxB (30) 119870CYP11B1
119898A (31) 119870CYP11B1119898B (32) 119881CYP11B1maxA
(33) 119881CYP11B1maxB (34) 119896CYP11B2 (35) 119870HSD17B3119898A (36) 119870HSD17B3
119898B (37) 119881HSD17B3maxA (38) 119881HSD17B3
maxB (39) 119870CYP19A1119898A (40) 119870CYP19A1
119898B (41) 119881CYP19A1maxA and (42)119881
CYP19A1maxB PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-
hydroxyprogesteroneDIONE androstenedioneDCORTICO 11-deoxycorticosteroneDCORT 11-deoxycortisol CORTICO corticosteroneCORT cortisol ALDO aldosterone and TESTO testosterone
pathway is very reasonable StAR CYP11A1 and 3120573-HSD(isoforms 1 or 2 in humans) proteins generally respond to thesame hormones that stimulate steroid production throughcommon pathways such as cAMP signaling in adrenal glandsand testes [65 66] Moreover our data also support recentexperimental evidence from clinical and in vivo studiessuggesting that the enzymatic activity of 3120573-HSD plays
an important role in the regulation of mineralocorticoidsynthesis in adrenal steroidogenesis and contributes to hyper-tension caused by abnormal overproduction of aldosterone[67ndash70] Circadian clock-deficient Cry-null mice show salt-sensitive hypertension due to abnormally high synthesis ofaldosterone which is caused by constitutively high expressionof HSD3B6mRNA and protein in the adrenal cortex [67 68]
Journal of Toxicology 15
All steroid producing cells
Mineralocorticoid producing cells Glucocorticoid producing cells
Mineralocorticoid and glucocorticoid producing cells
Terminated steroidogenesis
Sex steroids and precursor sex steroidsDIONE TESTO E1DEHA E2
NCI-H295Rcells at controlcondition
Normalized activities of HSD3B2
Nor
mal
ized
activ
ities
of C
YP17
H
2
18
16
14
12
1
08
06
04
02
00 05 1 15 2
Figure 7 Metabolic categories of the steroidogenic cells determined by the balance of HSD3B2 and CYP17H activitiesThe two-dimensionalparameter scanning analysis by the perturbation of two focused parameters clarifies the interaction and the relationship between the twoparameters in the complex system Functional cellular categories of steroidogenic cells were defined by the levels of mineralocorticoid(ALDO) glucocorticoids (DCORTandCORT) and androgens (DHEAandDIONE) at 72 h after stimulation Enzymatic activities ofHSD3B2andCYP17were normalized by standard values of the simulationmodel inNCI-H295R cells Scanning ranges ofHSD3B2 andCYP17 activitieswere 0ndash200 each 10 Green regions are all 14 steroids produced by NCI-H295R cells Red blue and purple regions are mineralocorticoidproducing cells glucocorticoids-producing cells and both corticoid-producing cells respectively Yellow regions are steroidogenesis of NCI-H295R cells terminated upstream of the adrenal steroidogenesis pathway
Recent clinical observations have revealed predominantexpression of HSD3B2 mRNA and protein in tumor cellsof aldosterone-producing adenoma (APA) and HSD3B1mRNA significantly correlated with CYP11B2 mRNA levelsand plasma aldosterone concentrations in APA patients [6970] However the relationship is unclear and disputed ina small-scale clinical study indicating that genetic variationin HSD3B1 affects blood pressure and hypertension in APApatients [71] The results of our simulation study suggest that3120573-HSD protein (human genes are HSD3B1 and HSD3B2) isone of the determination factors for the dynamic propertyof adrenal steroidogenesis Our results support the clinicalevidence of Doi and colleagues [69] and we believe that theHSD3B1 enzyme has a promising potential as novel drugtarget for endocrine hypertension
42 Metabolic Shift of Adrenal Steroidogenesis and the Contri-butions of HSD3B2 and CYP17H The metabolic propertiesof adrenal steroidogenesis in NCI-H295R cells were revealedby dynamic sensitivity analysis using themathematicalmodel(Figure 6) Mineralocorticoids such as DCORTICO COR-TICO and ALDO and intermediate steroids upstream of
the adrenal steroidogenesis pathway such as PREG HPREGDHEA PROG and HPROG were accelerated by reactionsof cholesterol import (119896Cholesterol Transport) StAR protein (119896MTR
119891
and 119896loc119891) and CYP11A1 (119881CYP11A1max ) On the other hand glu-
cocorticoids such as DCORT and CORT and sex hormonessuch as DIONE TESTO and E1 were suppressed by thesemodel parametersTherefore enhancement of the net adrenalsteroidogenesis capacity which supplies PREG precursor tothe pathway causes a production shift from glucocorticoidsto mineralocorticoids by substrate inhibitions of CYP17HHSD3B2 and CYP21A2 caused by accumulation of initialintermediate steroids such as PREG and PROG Sensitivi-ties of CYP17H (119881CYP17HmaxA ) and HSD3B2 (119881HSD3B2
maxA ) for theproducts were dynamically changed and these parametersdetermined the metabolic balance of downstream steroidsin the adrenal steroidogenesis pathway According to theseresults of dynamic sensitivity analysis of StAR CYP11A1CYP17H and HSD3B2 we suggest that the enhancementof CYP17H and HSD3B2 activity during ACTH stimulationwas important to shift the steroidogenic output away fromALDO biosynthesis towards CORT biosynthesis as wellas adrenal androgen production This suggestion partially
16 Journal of Toxicology
supports a comparative animal study in which molecularand cellular variations in CYP17H activity dramatically affectacute cortisol production resulting in distinct physiologicaland behavioral responses [72]
Results of two-dimensional parameter scanning of theenzymatic activities of CYP17H and HSD3B2 quantitativelyshowed the detail of the metabolic relationship betweenmin-eralocorticoid and glucocorticoid biosynthesis (Figure 7)Particularly the results showed that the balance of theseenzymatic activities was very important for the typicalfunction of NCI-H295R cells namely the ability to produceall steroid hormones NCI-H295R cells lost this functionwhen enzymatic activities of CYP17H and HSD3B2 werechanged by over 60 and 30 respectively In addition theybecame mineralocorticoid (ALDO) secreting cells when theenzymatic activity of CYP17H or HSD3B2 was inhibited byover 50 or glucocorticoid (DCORT and CORT) secretingcells when these enzymes were activated by over 50 Inparticular this analysis also showed that HSD3B2 was a keyplayer in the adrenal steroidogenesis of NCI-H295R cellsbecause HSD3B2 inhibition by over 80 almost completelyinhibited the biosynthesis of downstream steroids The ratioof CYP17A1 to HSD3B2 mRNA expression levels has beenrelated to several endocrine diseases with a low level in APAs[73] and high level in cortisol-producing adenomas [74]Furthermore the expression levels or enzymatic activitiesof CYP17A1 and HSD3B1 have been related to androgenproduction in polycystic ovary syndrome [75 76] Theseclinical studies support our simulation results indicating thatthe balance of enzymatic activity of CYP17H and HSD3B2determines the shift in steroidogenic output to mineralocor-ticoids glucocorticoids or androgens
43 Methodologies of Quantitative Mechanistic Analysis forDrug Discovery According to our results obtained using themathematical model of steroidogenesis in NCI-H295R cellssuch as sensitivity analysis comprehensive analysis based onsystems biology is available to quantitatively estimate themechanism of action of steroidogenic disrupting compoundsfrom differential profiling of adrenal steroid hormonesbecause dynamic patterns of steroid hormones in adrenalsteroidogenesis pathway are highly complex Our proposedmethod of quantitative mechanistic analysis of steroidogenicinhibitors was able to predict known action sites in theadrenal steroidogenesis pathway at only one time point (72 hafter drug exposure) Moreover according to the resultsof sensitivity analysis (Figure 6) 119881max of all steroidogenicenzymes was more sensitive than the 119870
119898 because the
intracellular concentrations of steroid hormones were almostmaintained at sufficiently high levels compared with 119870
119898
values of steroidogenic enzymesThese results suggested thatestimation of the mechanism of action of drugs is moreeffective and detectable when using the influences of 119881maxas the searching parameters such as our proposed methodOur data showed that the proposed method based on asystems biology model is a very powerful tool for exploratoryscreening of steroidogenic disrupting compounds
RCGA as a solver of parameter estimation problemsin systems biology has been applied to biological network
identification of gene regulatory networks and metabolicpathways and optimization of biological processes usingexperimentally observed time-course data [77ndash82] In thisstudy RCGA was useful to estimate the mechanism ofaction of novel pharmaceutical drug candidates for adrenalsteroidogenesis as a new application of RCGA in systemsbiology We had two issues when applying RCGA to thequantitative mechanistic analysis of drug actions Theseissues were the vast calculation cost and multimodality ofquasi-optimum solutions in solving the optimization prob-lem because the mathematical model in systems biologyconsists of many equations and parameters A proposedoptimization strategy using RCGA based on REXJGG wasa highly stable and efficient calculation method for a betterquasi-optimum solution than the unimodal normal distri-bution crossover (UNDX)minimum generation gap (MGG)method that is well applied in the engineering field In addi-tion we expanded the RCGA optimization program basedon REXJGG to a hybrid method of GA and then applieda local search as recommended by Harada and Kobayashi[28 83] A final optimal solution was obtained with a goodconvergence property Because these problems are generalin systems biology studies we suggest that the proposedhybrid method based on REXJGG is a very useful tool forquantitative mechanistic analysis of novel pharmaceuticaldrugs not limited to steroidogenic disrupting compounds
5 Conclusions
Thenovelmathematicalmodel of adrenal steroidogenesis wasconstructed in this study including cholesterol transport anddistribution the C
21-steroid hormone pathway steroid trans-
port and cell proliferation which could reproduce adrenalsteroidogenesis in NCI-H295R cells According to the resultsof dynamic sensitivity analysis using the newmodel HSD3B2plays the most important role in the metabolic balance ofadrenal steroidogenesis in NCI-H295R cells Moreover toquantitatively estimate mechanisms of action of adrenal toxiccompounds we analyzed differential metabolic profiles of12 steroid hormones at 3 days after exposure to 11 adrenaltoxic compounds by using the new mathematical modeland a hybrid optimization method of the RCGA and a localsearch (nonlinear least squares) We could estimate whichsteroidogenic enzymes were affected by these compoundsusing the hybrid optimization method Vasculotoxic agentswere estimated to have no effect according to the resultsobtained by our method In terms of adrenal steroidogenicinhibitors the predicted action sites were approximatelymatched to the target enzymes as reported in the literatureThus our computer-aided method based on a systems biol-ogy approach may be useful to analyze the mechanism ofaction of endocrine-disrupting compounds and to assess thehuman adrenal toxicity of novel pharmaceutical drugs basedon steroid hormone profiling
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Journal of Toxicology 17
Authorsrsquo Contribution
Ryuta Saito Natsuko Terasaki and Naohisa Tsutsui designedthe study Naohisa Tsutsui and Naoya Masutomi supervisedthe experiments Natsuko Terasaki performed all in vitroexperiments and collected dataMakotoYamazaki performedall measurements of steroids Ryuta Saito conceived the insilico strategies analyzed data developed all programs andthe mathematical model and performed all simulation anal-yses with Masahiro Okamoto Ryuta Saito Natsuko TerasakiMakoto Yamazaki NaoyaMasutomi andMasahiroOkamotowrote the paper
Acknowledgments
The authors would like to thank Dr Michael S Breenand Dr Miyuki Breen for helpful discussions concerningprevious mathematical models of adrenal steroidogenesisThis work was partially supported by the Grant-in-Aid forScientific Research on Innovative Areas ldquoSynthetic Biologyrdquo(No 23119001 (MO)) from Ministry of Education CultureSports Science and Technology in Japan
References
[1] P W Harvey D J Everett and C J Springall ldquoAdrenal toxi-cology a strategy for assessment of functional toxicity to theadrenal cortex and steroidogenesisrdquo Journal of Applied Toxi-cology vol 27 no 2 pp 103ndash115 2007
[2] W E Ribelin ldquoThe effects of drugs and chemicals upon thestructure of the adrenal glandrdquo Fundamental and Applied Toxi-cology vol 4 no 1 pp 105ndash119 1984
[3] H D Colby and P A Longhurst ldquoToxicology of the adrenalglandrdquo in Endocrine Toxicology C K Atterwill and J D FlackEds pp 243ndash281 CambridgeUniversity Press Cambridge UK1992
[4] H D Colby ldquoIn vitro assessment of adrenocortical toxicityrdquoJournal of Pharmacological and Toxicological Methods vol 32no 1 pp 1ndash6 1994
[5] T J Rosol J T Yarrington J Latendresse and C C CapenldquoAdrenal gland structure function and mechanisms of toxic-ityrdquo Toxicologic Pathology vol 29 no 1 pp 41ndash48 2001
[6] P W Harvey and D J Everett ldquoThe adrenal cortex and ster-oidogenesis as cellular and molecular targets for toxicity crit-ical omissions from regulatory endocrine disrupter screeningstrategies for human healthrdquo Journal of Applied Toxicology vol23 no 2 pp 81ndash87 2003
[7] A F Gazdar H K Oie C H Shackleton et al ldquoEstablishmentand characterization of a human adrenocortical carcinoma cellline that expresses multiple pathways of steroid biosynthesisrdquoCancer Research vol 50 no 17 pp 5488ndash5496 1990
[8] W E Rainey I M Bird C Sawetawan et al ldquoRegulation ofhuman adrenal carcinoma cell (NCI-H295) production of C19steroidsrdquo The Journal of Clinical Endocrinology amp Metabolismvol 77 no 3 pp 731ndash737 1993
[9] B Staels D W Hum andW L Miller ldquoRegulation of steroido-genesis in NCI-H295 cells a cellular model of the human fetaladrenalrdquo Molecular Endocrinology vol 7 no 3 pp 423ndash4331993
[10] M Hecker J L Newsted M B Murphy et al ldquoHuman adreno-carcinoma (H295R) cells for rapid in vitro determination of
effects on steroidogenesis hormone productionrdquo Toxicologyand Applied Pharmacology vol 217 no 1 pp 114ndash124 2006
[11] A Oskarsson E Ulleras K E Plant J P Hinson and P SGoldfarb ldquoSteroidogenic gene expression in H295R cells andthe human adrenal gland adrenotoxic effects of lindane invitrordquo Journal of Applied Toxicology vol 26 no 6 pp 484ndash4922006
[12] UMuller-Vieira M Angotti and RW Hartman ldquoThe adreno-cortical tumor cell line NCI-H295R as an in vitro screeningsystem for the evaluation of CYP11B2 (aldosterone synthase)and CYP11B1 (steroid-11120573-hydroxylase) inhibitorsrdquoThe Journalof Steroid Biochemistry and Molecular Biology vol 96 no 3-4pp 259ndash270 2005
[13] K Imagawa S Okayama M Takaoka et al ldquoInhibitory effectof efonidipine on aldosterone synthesis and secretion in humanadrenocarcinoma (H295R) cellsrdquo Journal of CardiovascularPharmacology vol 47 no 1 pp 133ndash138 2006
[14] M Voets UMuller-Vieira S Marchais-Oberwinkler and RWHartmann ldquoSynthesis of amidinohydrazones and evaluation oftheir inhibitory effect towards aldosterone synthase (CYP11B2)and the formation of selected steroidsrdquo Archiv der Pharmazievol 337 no 7 pp 411ndash416 2004
[15] F K Nielsen C H Hansen J A Fey et al ldquoH295R cells as amodel for steroidogenic disruption a broader perspective usingsimultaneous chemical analysis of 7 key steroid hormonesrdquoToxicology in Vitro vol 26 no 2 pp 343ndash350 2012
[16] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cortisolproduction and on activities of steroidogenic enzymes inhuman adrenocortical H295R cellsrdquo Journal of Steroid Biochem-istry and Molecular Biology vol 80 no 3 pp 355ndash363 2002
[17] R F Canton J T Sanderson S Nijmeijer A Bergman R JLetcher and M van den Berg ldquoIn vitro effects of brominatedflame retardants and metabolites on CYP17 catalytic activity anovel mechanism of actionrdquo Toxicology and Applied Pharma-cology vol 216 no 2 pp 274ndash281 2006
[18] K Hilscherova P D Jones T Gracia et al ldquoAssessment ofthe effects of chemicals on the expression of ten steroidogenicgenes in the H295R cell line using real-time PCRrdquo ToxicologicalSciences vol 81 no 1 pp 78ndash89 2004
[19] T Gracia K Hilscherova P D Jones et al ldquoThe H295R systemfor evaluation of endocrine-disrupting effectsrdquo Ecotoxicologyand Environmental Safety vol 65 no 3 pp 293ndash305 2006
[20] X Zhang J L NewstedM Hecker E B Higley P D Jones andJ P Giesy ldquoClassification of chemicals based on concentration-dependent toxicological data using ToxClustrdquo EnvironmentalScience and Technology vol 43 no 10 pp 3926ndash3932 2009
[21] M S Breen M Breen N Terasaki M Yamazaki and R BConolly ldquoComputational model of steroidogenesis in humanH295R cells to predict biochemical response to endocrine-active chemicalsmodel development formetyraponerdquoEnviron-mental Health Perspectives vol 118 no 2 pp 265ndash272 2010
[22] M Breen M S Breen N Terasaki M Yamazaki A LLloyd and R B Conolly ldquoMechanistic computational model ofsteroidogenesis in H295R cells role of oxysterols and cell pro-liferation to improve predictability of biochemical response toendocrine active chemical-metyraponerdquo Toxicological Sciencesvol 123 no 1 pp 80ndash93 2011
[23] F B Kraemer ldquoAdrenal cholesterol utilizationrdquo Molecular andCellular Endocrinology vol 265-266 pp 42ndash45 2007
[24] W L Miller ldquoSteroidogenic acute regulatory protein (StAR)a novel mitochondrial cholesterol transporterrdquo Biochimica et
18 Journal of Toxicology
Biophysica Acta (BBA)mdashMolecular and Cell Biology of Lipidsvol 1771 no 6 pp 663ndash676 2007
[25] S Szabo and I T Lippe ldquoAdrenal gland chemically inducedstructural and functional changes in the cortexrdquo ToxicologicPathology vol 17 no 2 pp 317ndash329 1989
[26] C Chen-Pan I-J Pan Y Yamamoto T Sakogawa J Yamadaand YHayashi ldquoPrompt recovery of damaged adrenalmedullaeinduced by salinomycinrdquoToxicologic Pathology vol 27 no 5 pp563ndash572 1999
[27] R J Santen and R I Misbin ldquoAminoglutethimide review ofpharmacology and clinical userdquo Pharmacotherapy vol 1 no 2pp 95ndash120 1981
[28] M J Carella N V Dimitrov V V Gossain L Srivastava andD R Rovner ldquoAdrenal effects of low-dose aminoglutethimidewhen used alone in postmenopausal women with advancedbreast cancerrdquoMetabolism vol 43 no 6 pp 723ndash727 1994
[29] M K Johansson J T Sanderson and B-O Lund ldquoEffects of3-MeSO2-DDE and some CYP inhibitors on glucocorticoidsteroidogenesis in the H295R human adrenocortical carcinomacell linerdquo Toxicology in Vitro vol 16 no 2 pp 113ndash121 2002
[30] Y Touitou A Bogdan J C Legrand and P Desgrez ldquoAmino-glutethimide and glutethimide effects on 18-hydroxycorti-costerone biosynthesis by human and sheep adrenals in vitrordquoActa Endocrinologica vol 80 no 3 pp 517ndash526 1975
[31] M M Hart and J A Straw ldquoStudies on the site of action ofop1015840-DDD in the dog adrenal cortex 1 Inhibition of ACTH-mediated pregnenolone synthesisrdquo Steroids vol 17 no 1ndash5 pp559ndash574 1971
[32] M Ojima M Saitoh N Itoh Y Kusano S Fukuchi and HNaganuma ldquoThe effects of op1015840-DDD on adrenal steroidogen-esis and hepatic steroid metabolismrdquo Nippon Naibunpi Gakkaizasshi vol 61 no 3 pp 168ndash178 1985
[33] L F Canosa and N R Ceballos ldquoEffects of different steroid-biosynthesis inhibitors on the testicular steroidogenesis of thetoad Bufo arenarumrdquo Journal of Comparative Physiology B vol171 no 6 pp 519ndash526 2001
[34] D C Kossor S Kominami S Takemori and H D ColbyldquoRole of the steroid 17 alpha-hydroxylase in spironolactone-mediated destruction of adrenal cytochrome P-450rdquoMolecularPharmacology vol 40 no 2 pp 321ndash325 1991
[35] S C Cheng K Suzuki W Sadee and B W Harding ldquoEffectsof spironolactone canrenone and canrenoate K on cytochromeP450 and 11120573 and 18 hydroxylation in bovine and humanadrenal corticalmitochondriardquo Endocrinology vol 99 no 4 pp1097ndash1106 1976
[36] K Yanagibashi M Haniu J E Shively W H Shen and P HallldquoThe synthesis of aldosterone by the adrenal cortex Two zones(fasciculata and glomerulosa) possess one enzyme for 11120573- 18-hydroxylation and aldehyde synthesisrdquoThe Journal of BiologicalChemistry vol 261 no 8 pp 3556ndash3562 1986
[37] K Morishita H Okumura N Ito and N Takahashi ldquoPrimaryculture system of adrenocortical cells from dogs to evaluatedirect effects of chemicals on steroidogenesisrdquo Toxicology vol165 no 2-3 pp 171ndash178 2001
[38] S W J Lamberts E G Bons H A Bruining and F H de JongldquoDifferential effects of the imidazole derivatives etomidateketoconazole and miconazole and of metyrapone on the secre-tion of cortisol and its precursors by human adrenocorticalcellsrdquo Journal of Pharmacology and Experimental Therapeuticsvol 240 no 1 pp 259ndash264 1987
[39] L Yin S Lucas F Maurer U Kazmaier Q Hu and RW Hartmann ldquoNovel imidazol-1-ylmethyl substituted 1256-tetrahydropyrrolo[321-ij]quinolin-4-ones as potent and selec-tive CYP11B1 inhibitors for the treatment of Cushingrsquos syn-dromerdquo Journal ofMedicinal Chemistry vol 55 no 14 pp 6629ndash6633 2012
[40] M DiMattina N Maronian H Ashby D L Loriaux andB D Albertson ldquoKetoconazole inhibits multiple steroidogenicenzymes involved in androgen biosynthesis in the humanovaryrdquo Fertility and Sterility vol 49 no 1 pp 62ndash65 1988
[41] M Ayub and M J Levell ldquoInhibition of human adrenal steroi-dogenic enzymes in vitro by imidazole drugs including keto-conazolerdquo Journal of Steroid Biochemistry vol 32 no 4 pp 515ndash524 1989
[42] D Engelhardt M M Weber T Miksch F Abedinpour andC Jaspers ldquoThe influence of ketoconazole on human adrenalsteroidogenesis incubation studies with tissue slicesrdquo ClinicalEndocrinology vol 35 no 2 pp 163ndash168 1991
[43] S Bhasin S Sikka T Fielder et al ldquoHormonal effects of keto-conazole in vivo in the male rat mechanism of actionrdquo Endo-crinology vol 118 no 3 pp 1229ndash1232 1986
[44] D M Zisterer and D C Williams ldquoCalmidazolium and otherimidazole compounds affect steroidogenesis in Y1 cells lack ofinvolvement of the peripheral-type benzodiazepine receptorrdquoJournal of Steroid Biochemistry and Molecular Biology vol 60no 3-4 pp 189ndash195 1997
[45] L P Walsh C N Kuratko and D M Stocco ldquoEconazoleand miconazole inhibit steroidogenesis and disrupt steroido-genic acute regulatory (StAR) protein expression post-trans-criptionallyrdquo Journal of Steroid Biochemistry and Molecular Bio-logy vol 75 no 4-5 pp 229ndash236 2000
[46] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cor-tisol production and on activities of steroidogenic enzymesin human adrenocortical H295R cellsrdquo The Journal of SteroidBiochemistry and Molecular Biology vol 80 no 3 pp 355ndash3632002
[47] D P Dempsher D S Gann and R D Phair ldquoA mechanisticmodel of ACTH-stimulated cortisol secretionrdquo American Jour-nal of Physiology vol 246 no 4 part 2 pp R587ndashR596 1984
[48] N Sarai S Matsuoka and A Noma ldquosimBio a Java package forthe development of detailed cell modelsrdquo Progress in Biophysicsand Molecular Biology vol 90 no 1ndash3 pp 360ndash377 2006
[49] K Levenberg ldquoA method for the solution of certain non-linearproblems in least squaresrdquo Quarterly of Applied Mathematicsvol 2 pp 164ndash168 1944
[50] D W Marquardt ldquoAn algorithm for least-squares estimation ofnonlinear parametersrdquo SIAM Journal on Applied Mathematicsvol 11 no 2 pp 431ndash441 1963
[51] P E Gill and W Murray ldquoAlgorithms for the solution of thenonlinear least-squares problemrdquo SIAM Journal on NumericalAnalysis vol 15 no 5 pp 977ndash992 1978
[52] Y Akimoto R Hasada J Sakuma I Ono and S KobayashildquoGeneration alternation model for real-coded GA using multi-parent proposal and evaluation of just generation gap (JGG)rdquoin Proceedings of the 19th SICE Symposium on DecentralizedAutonomous Systems pp 341ndash346 Tokyo Japan January 2007
[53] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoProposal and evaluation of adaptive real-coded crossoverAREXrdquo Transactions of the Japanese Society for Artificial Intel-ligence vol 24 no 6 pp 446ndash458 2009
Journal of Toxicology 19
[54] S Kobayashi ldquoThe frontiers of real-coded genetic algorithmsrdquoJournal of Japanese Society for Artificial Intelligence vol 24 no1 pp 147ndash162 2009
[55] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoAnalysis of the behavior of MGG and JGG as a selection modelfor real-coded genetic algorithmsrdquo Transactions of the JapaneseSociety for Artificial Intelligence vol 25 no 2 pp 281ndash289 2010
[56] A Struthers H Krum and G H Williams ldquoA comparisonof the aldosterone-blocking agents eplerenone and spironolac-tonerdquo Clinical Cardiology vol 31 no 4 pp 153ndash158 2008
[57] K Yamasaki M Sawaki S Noda et al ldquoComparison of theHershberger assay and androgen receptor binding assay oftwelve chemicalsrdquo Toxicology vol 195 no 2-3 pp 177ndash1862004
[58] C Eil and S K Edelson ldquoThe use of human skin fibroblasts toobtain potency estimates of drug binding to androgen recept-orsrdquo The Journal of Clinical Endocrinology amp Metabolism vol59 no 1 pp 51ndash55 1984
[59] P Ye T Yamashita D M Pollock H Sasano and W ERainey ldquoContrasting effects of eplerenone and spironolac-tone on adrenal cell steroidogenesisrdquo Hormone and MetabolicResearch vol 41 no 1 pp 35ndash39 2009
[60] B D Albertson K L Frederick N C Maronian et al ldquoTheeffect of ketoconazole on steroidogenesis I Leydig cell enzymeactivity in vitrordquo Research Communications in Chemical Patho-logy and Pharmacology vol 61 no 1 pp 17ndash26 1988
[61] B D Albertson N C Maronian K L Frederick et al ldquoTheeffect of ketoconazole on steroidogenesis II Adrenocorticalenzyme activity in vitrordquoResearch Communications in ChemicalPathology and Pharmacology vol 61 no 1 pp 27ndash34 1988
[62] DM Stocco andB J Clark ldquoRegulation of the acute productionof steroids in steroidogenic cellsrdquo Endocrine Reviews vol 17 no3 pp 221ndash244 1996
[63] L K Christenson and J F Strauss III ldquoSteroidogenic acute regu-latory protein (StAR) and the intramitochondrial translocationof cholesterolrdquo Biochimica et Biophysica Acta (BBA)mdashMolecularand Cell Biology of Lipids vol 1529 no 1ndash3 pp 175ndash187 2000
[64] E R Simpson ldquoCholesterol side-chain cleavage cytochromeP450 and the control of steroidogenesisrdquo Molecular and Cel-lular Endocrinology vol 13 no 3 pp 213ndash227 1979
[65] H A LaVoie and S R King ldquoTranscriptional regulation of ster-oidogenic genes STARD1 CYP11A1 and HSD3Brdquo ExperimentalBiology and Medicine vol 234 no 8 pp 880ndash907 2009
[66] F Raucci A DrsquoAniello and M M Di Fiore ldquoStimulation ofandrogen production by d-aspartate through the enhancementof StAR P450scc and 3120573-HSD mRNA levels in vivo rat testisand in culture of immature rat Leydig cellsrdquo Steroids vol 84pp 103ndash110 2014
[67] M Doi Y Takahashi R Komatsu et al ldquoSalt-sensitive hyper-tension in circadian clock-deficient Cry-null mice involves dys-regulated adrenal Hsd3b6rdquo Nature Medicine vol 16 no 1 pp67ndash74 2010
[68] H Okamura M Doi Y Yamaguchi and J-M Fustin ldquoHyper-tension due to loss of clock novel insight from the molecu-lar analysis of Cry1Cry2-deleted micerdquo Current HypertensionReports vol 13 no 2 pp 103ndash108 2011
[69] M Doi F Satoh T Maekawa et al ldquoIsoform-specific mono-clonal antibodies against 3120573-hydroxysteroid dehydrogenaseisomerase family provide markers for subclassification ofhuman primary aldosteronismrdquo The Journal of Clinical Endo-crinology amp Metabolism vol 99 no 2 pp E257ndashE262 2014
[70] S Konosu-Fukaya Y Nakamura F Satoh et al ldquo3120573-hydroxy-steroid dehydrogenase isoforms in human aldosterone-produc-ing adenomardquo Molecular and Cellular Endocrinology vol 408pp 205ndash212 2015
[71] G C Verwoert J Hofland N Amin et al ldquoExpression and genevariation studies deny association of humanHSD3B1 gene withaldosterone production or blood pressurerdquoAmerican Journal ofHypertension vol 28 no 1 pp 113ndash120 2015
[72] D Hough K Storbeck SW P Cloete A C Swart and P SwartldquoRelative contribution of P450c17 towards the acute cortisolresponse lessons from sheep and goatsrdquoMolecular and CellularEndocrinology vol 408 pp 107ndash113 2015
[73] I Sakuma S Suematsu Y Matsuzawa et al ldquoCharacterizationof steroidogenic enzyme expression in aldosterone-producingadenoma a comparison with various human adrenal tumorsrdquoEndocrine Journal vol 60 no 3 pp 329ndash336 2013
[74] A Tong A Jia S Yan Y Zhang Y Xie and G Liu ldquoEctopiccortisol-producing adrenocortical adenoma in the renal hilumhistopathological features and steroidogenic enzyme profilerdquoInternational Journal of Clinical and Experimental Pathologyvol 7 no 7 pp 4415ndash4421 2014
[75] A Hirsch D Hahn P Kempna et al ldquoMetformin inhibitshuman androgen production by regulating steroidogenicenzymes HSD3B2 and CYP17A1 and complex I activity of therespiratory chainrdquo Endocrinology vol 153 no 9 pp 4354ndash43662012
[76] B Xu L Gao Y Cui et al ldquoSET protein up-regulated test-osterone production in the cultured preantral folliclesrdquo Repro-ductive Biology and Endocrinology vol 11 article 9 2013
[77] S Kikuchi D Tominaga M Arita M Takahashi and MTomita ldquoDynamic modeling of genetic networks using geneticalgorithm and S-systemrdquo Bioinformatics vol 19 no 5 pp 643ndash650 2003
[78] N Shikata YMaki M Nakatsui et al ldquoDetermining importantregulatory relations of amino acids from dynamic networkanalysis of plasma amino acidsrdquo Amino Acids vol 38 no 1 pp179ndash187 2010
[79] M Nakatsui T Ueda Y Maki I Ono and M OkamotoldquoMethod for inferring and extracting reliable genetic interac-tions from time-series profile of gene expressionrdquoMathematicalBiosciences vol 215 no 1 pp 105ndash114 2008
[80] A Komori Y Maki M Nakatsui I Ono and M OkamotoldquoEfficient numerical optimization algorithm based on new real-coded genetic algorithm AREX + JGG and application to theinverse problem in systems biologyrdquo Applied Mathematics vol3 no 10 pp 1463ndash1470 2012
[81] J A Roubos G van Straten and A J B van Boxtel ldquoAn evolu-tionary strategy for fed-batch bioreactor optimization conceptsand performancerdquo Journal of Biotechnology vol 67 no 2-3 pp173ndash187 1999
[82] D Sarkar and J M Modak ldquoGenetic algorithms with filters foroptimal control problems in fed-batch bioreactorsrdquo Bioprocessand Biosystems Engineering vol 26 no 5 pp 295ndash306 2004
[83] K Harada K Ikeda J Sakuma I Ono and S KobayashildquoHybridization of genetic algorithm with local search in mul-tiobjective function optimization recommendation of GA thenLSrdquoTransactions of the Japanese Society for Artificial Intelligencevol 21 no 6 pp 482ndash492 2006
Submit your manuscripts athttpwwwhindawicom
PainResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
ToxinsJournal of
VaccinesJournal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AntibioticsInternational Journal of
ToxicologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
StrokeResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Drug DeliveryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Pharmacological Sciences
Tropical MedicineJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AddictionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Emergency Medicine InternationalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Autoimmune Diseases
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anesthesiology Research and Practice
ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Pharmaceutics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
MEDIATORSINFLAMMATION
of
4 Journal of Toxicology
Industries Ltd Osaka Japan) All chemicals were dissolvedin DMSO (Wako Pure Chemical Industries Ltd OsakaJapan) and added to the culture medium at 1 1000 dilutions
217 Validation Study Using Adrenal Toxicants NCI-H295Rcells were cultured for 3 days in 6-well plates and thenstimulated with the above-mentioned compounds Upon thestart of stimulation various concentrations of test chemicalswere added to the cultures After a further 3 days of culturewith the chemicals the concentrations of 12 steroids (PREGHPREG DHEA PROG HPROG DIONE DCORTICODCORT CORTICO CORT ALDO and TESTO) in theculture medium were measured by LCMSMSThe test con-centrations of the chemicalswere determined by dose-findingcytotoxicity assays The cytotoxicity assay was conducted in96-well plates using ATP content in cells as an endpoint(CellTiter-Glotrade Luminescent Cell Viability Assay Promega)Concentrations that caused more than 20 cytotoxicity werenot used in the steroidogenesis assay The test concentrationsof adrenal steroidogenesis inhibitors and other compoundsare shown in Table 1
218 Statistical Analysis Comparisons were performed bythe two-sample Welchrsquos 119905-test with Bonferroni multiple test-ing correction for each steroid hormone species Statisticallysignificant steroid hormones were considered at adjusted119901 values of less than 001 Differential metabolic steroidprofiles were classified by hierarchical cluster analysis Pair-wise distances between all compounds and all steroids werecalculated by standardized Euclidean metric This distancematrix was analyzed with Wardrsquos method for hierarchicalclustering Statistical analysis was performed usingMATLABsoftware (MathWorks Inc Natick MA)
22 The Computational Part
221 Mathematical Modeling of Adrenal Steroidogenesis inNCI-H295R Cells Steroid hormones secreted from humanadrenal corticocarcinoma NCI-H295R cells are synthesizedfrom cholesterol through the C
21-steroid hormone biosyn-
thesis pathway A mathematical model of adrenal steroido-genesis in NCI-H295R cells was constructed with cholesteroltransport and the intracellular localization pathway theoxysterol synthesis pathway as a bypass of steroidogenesisthe C
21-steroid hormone biosynthesis pathway as the main
steroidogenesis pathway passive transport of steroid hor-mones and cell proliferation (Figure 1) In this model twocompartments the intracellular space and culture mediumwere incorporated as the available region Equations andparameters of the cell proliferation and diffusional trans-port of steroid hormones have been proposed by previousstudies [21 22] Cholesterol transport and the intracellularlocalization pathway including the oxysterol bypass wereintegrated using a part of the ACTH-stimulated cortisolsecretion model described by Dempsher and colleagues [47]The C
21-steroid hormone biosynthesis pathway includes 14
steroid hormones PREG HPREG DHEA PROG HPROGDIONE TESTO DCORTICO DCORT CORTICO CORTALDO E1 and E2 and 17 enzymatic reactions catalyzed by
nine steroidogenic enzymes cholesterol side chain cleavageenzyme (CYP11A1) 17120572-hydroxylase (CYP17H) C
1720-lyase
(CYP17L) 3120573-hydroxysteroid dehydrogenase (HSD3B2) 21-hydroxylase (CYP21A2) 11120573-hydroxylase (CYP11B1) 18-hydroxylase (CYP11B2) 17120573-hydroxysteroid dehydrogenase(HSD17B3) and aromatase (CYP19A1) In this mathematicalmodel of adrenal steroidogenesis inNCI-H295R cells the fluxvelocities ofmolecular transportation and enzymatic reactionrates of steroidogenic enzymes were defined based on thefirst-order reaction and rapid-equilibrium enzyme kineticsrespectively All equations in the mathematical model ofadrenal steroidogenesis of NCI-H295R cells were describedin a supplementary document (see Supplementary Materialavailable online at httpdxdoiorg10115520164041827)The rate constants and the maximum activities were esti-mated by fitting to experimental time-course data of theconcentrations of cholesterol and all steroids Initial valuesof cholesterol and the 14 steroid concentrations were usedin each experimentally measured value and every steroidconcentration was assumed to rapidly reach the equilibriumstate between the culture medium and intracellular space Allfixed values of static parameters and initial values of variableparameters in this model were described in Tables S1 and S2in a supplementary document respectively
222 Modeling and Simulation Environment This compu-tational model of adrenal steroidogenesis in NCI-H295Rcells was developed on the simBio platform which is ageneral environment of biological dynamic simulation andcomputational model development [48] ODEs were solvedby the fourth-order Runge-Kuttamethodwith a variable timestepThe time step (119889119905) was adjusted to refer to themaximumabsolute value of flux velocities or enzymatic reaction ratesat each time point and the range of the time step was from1 times 10minus5 to 10minus2 To confirm whether the range of the timestep was suitable the numerical error ratio was calculated bycertain fixed time steps in the range of the time step whichwas under 1 times 10minus8 in every time step The duration time ofcomputational simulation of adrenal steroidogenesis in NCI-H295R cells was set at 72 h
223 Parameter Optimization To reconstruct experimentaltime-course patterns of the concentrations of cholesterol andthe 14 steroids in the culture medium and intracellular spacewe optimized every rate constant and maximum velocityof the steroidogenic enzymes This parameter optimizationproblem was solved by the Levenberg-Marquardt methodwhich is one of the nonlinear least squares methods [49ndash51] The objective function of optimization was used as thefollowing normalized least squares distance (NLSD)
NLSD = sumℎ
sum
119894
sum
119895
(119883expℎ119894119895minus 119883
simℎ119894119895)
2
119883maxℎ119894
2 (1)
where ℎ is the compartment (culture medium or intracellularspace) 119894 is the molecular species (cholesterol and the 14steroids) 119895 is the time point (0 8 24 48 and 72 h) 119883exp
ℎ119894119895
is the experimentally measured concentration of molecule
Journal of Toxicology 5
Table1Ad
renaltoxicities
andactio
nsof
teste
dcompo
unds
Testchem
ical
Testconcentrations
(120583M)
Pathologicalfeatures
ofadrenaltoxicity
Repo
rted
enzymeinh
ibition
sRe
ferences
Acrylonitrile
(AN)
01110and
100
Hem
orrhagicadrenaln
ecrosis
Not
repo
rted
[25]
Salin
omycin
(SM)
000
00100101
1and10
Dam
agetoadrenalm
edullae
Not
repo
rted
[26]
Thioguanine(TG
)001110and100
Hem
orrhagicadrenaln
ecrosis
Not
repo
rted
[25]
Fumaron
itrile
(FN)
01110and
100
Hem
orrhagicadrenaln
ecrosis
Not
repo
rted
[25]
Aminoglutethim
ide(AG
T)01110and
100
Hypertro
phyvascular
degeneratio
nCY
P11A
1CY
P21A
2CY
P11B1andCY
P11B2
[627ndash30]
op1015840-D
DD(D
DD)
0111025and100
Atroph
yCY
P11A
1HSD
3B2CY
P21A
2CY
P11B1andCY
P11B2
[293132]
Spiro
nolacton
e(SP
)11050and100
Hypertro
phy
CYP17H
CYP
17LCY
P11B1andCY
P11B2
[633ndash35]
Metyrapon
e(MP)
01110and
100
Hypertro
phyvascular
degeneratio
nCY
P11A
1CY
P11B1andCY
P11B2
[62936ndash39]
Ketoconazole(KC)
01110and
100
Hypertro
phy
CYP11A
1CY
P17H
CYP
17LHSD
3B2CY
P21A
2and
CYP11B1
[62940
ndash43]
Micon
azole(MC)
011102550and
100
Hypertro
phy
CYP11A
1CY
P17H
CYP
17LCY
P21A
2andCY
P11B1
[4144
45]
Daidzein(D
Z)01110and
100
Unk
nown
HSD
3B2andCY
P21A
2[46]
6 Journal of Toxicology
CHOC
PREG
PROG
DCORTICO
CORTICO
ALDO
CORT
DCORT
HPROG
HPREG DHEA
DIONE
TESTO
CYP11A1
E1
E2
HSD3B2
CYP1
7L
CYP21A2 HSD17B3
CYP1
9A1
CYP11B2
OXYOxysterol synthesis
PREGHPREG DHEAPROG
HPROG DIONE E1DCORTICODCORT TESTO E2CORTICO
CORTALDO
NCI-H295R cell
Medium culture
CYP11B1
CHOL
StAR
CHOM
CHOSCEH
CHOR
CHOLCholesterol transport
Total intracellular cholesterol
CYP1
7H
Vloc Vacc
VMTR
VCEH
Figure 1 Schematic diagram of the mathematical model of adrenal steroidogenesis in NCI-H295R cells Overview of the mathematicalmodel of adrenal steroidogenesis in NCI-H295R cells including cholesterol transport and intracellular localization oxysterol synthesisthe C
21-steroid hormone biosynthesis pathway passive diffusional transport of steroid hormones and cell proliferation CHOL total
cholesterol in medium culture CHOS stored cholesterol esters in the endoplasmic reticulum CHOC intracellular free cholesterolCHOM mitochondrial free cholesterol CHON mitochondrial free cholesterol close to CYP11A1 enzymes CHOR mitochondrial freecholesterol remote from CYP11A1 enzymes PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosteronePROG progesterone HPROG 17120572-hydroxyprogesterone DIONE androstenedione DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICO corticosterone CORT cortisol ALDO aldosterone TESTO testosterone E1 estrone E2 17120573-estradiol OXYoxysterol CEH cholesterol ester hydrolase StAR steroidogenic acute regulatory protein CYP11A1 P450 side chain cleavage enzymeCYP17H 17120572-hydroxylase of CYP17 CYP17L C
17ndash20 lyase of CYP17 HSD3B2 3120573-hydroxysteroid dehydrogenase CYP21A2 21-hydroxylaseCYP11B1 11120573-hydroxylase CYP11B2 18-hydroxylase HSD17B3 17120573-hydroxysteroid dehydrogenase and CYP19A1 aromatase
119894 in compartment ℎ at time point 119895 119883simℎ119894119895
is the simulatedconcentration of molecule 119894 in compartment ℎ at time point119895 and 119883max
ℎ119894is the maximum concentration of molecule 119894 in
compartment ℎ over all time points Data points under thelower quantitation limit were excluded from the evaluationby the objective function
Effects of every static model parameter for parameteroptimization were calculated from differences of fitting theobjective function using sensitivity analysis
224 Quantitative Estimation of the Mechanism of Action ofAdrenal Toxicants Metabolic steroid profiling and differen-tial patterns of the adrenal steroid hormones by chemicalperturbation were reconstructed to optimize the relativeactivities of the steroidogenic enzymes The input datafor the quantitative mechanistic analysis of adrenal toxiccompounds was a fold change (ratio) of the measured 12
steroid concentrations induced by drug exposure for 72 hThe two-step optimization method of the real-coded geneticalgorithm (RCGA) was adopted as a global optimizationmethod in the quantitative mechanistic analysis of adrenaltoxic compoundsThe operations of the crossover and gener-ation alterationmodel in RCGAwere used for the real-codedensemble crossover (REX) and just generation gap (JGG)[52ndash55] As the initial parameters of RCGA maximum gen-eration population size selection size of parent individualspopulation size of child individuals and termination criteriawere 1000 100 6 25 and under 01 of NLSD respectivelyThe search space for the relative activities of the steroidogenicenzymes was from 1100 to 100 To evaluate the fitness ofeach individual the sum of squared residuals for fold changesof measured 12 steroid concentrations was used as theobjective function Nonlinear least squares optimization bythe Levenberg-Marquardt method was used as a local
Journal of Toxicology 7
search [49ndash51] As the estimated mechanisms of actionsof the adrenal toxic compounds the relative activities ofeight steroidogenic enzymes (CYP11A1 CYP17H CYP17LHSD3B2 CYP21A2 CYP11B1 CYP11B2 and HSD17B3)were optimized by the above-mentioned 2-step optimizationmethod Every optimization calculation was duplicated tocheck the numerical stability of the optimal parameters
225 Global Dynamic Sensitivity Analysis The propertyof every kinetic parameter in this computational modelof steroidogenesis in NCI-H295R cells was evaluated bydynamic sensitivity analysis The sensitivity (119878
119909119910) of kinetic
parameter 119909 for variable parameter 119910 was defined by the fol-lowing equation
119878119909119910(119905)=
Δ119910(119905)119910(119905)
Δ119909119909 (2)
where variable parameter119910was the concentration of a steroidhormone in the cytosolic space of NCI-H295R cells Theperturbation for kinetic parameters was +10 (Δ119909119909 = 01)
3 Results
31 Experimental Data on Adrenal Steroidogenesis
311 Adrenal Steroidogenesis of NCI-H295R Cells and theMass Balance All steroid hormones in the culture mediumwere significantly increased after 72 h of stimulation with50 nM ACTH 20120583M forskolin and 100 nM angiotensinII (Figure 2(a)) Mass balances in steroidogenesis of NCI-H295R cells under nontreatment and control (stimulated)conditions are shown in Figures 2(b) and 2(c) respectivelyUnder stimulation the dynamics of net mass in these exper-iments were unchanged and accumulated cholesterol wasconverted to adrenal steroids
312 Cytotoxicity of Adrenal Toxicants Viabilities of cellstreated with each compound were expressed as a relativevalue to the ATP level of the control Effects of AN SM TGFN AGT DDD SP MP KC MC and DZ on cell viabilitywere determined to be valid under 80 of the relative ATPlevel at 7 days after treatment AN SM TG and FN showedcytotoxicity at over 100 1 10 and 10120583M respectively AGTMP andDZ did not affect cell viability at up to 100 120583MDDDSP KC andMC induced less than 80 of cell viability at over100 50 100 and 25 120583M respectively
313 Differentially Steroid Profiling of Adrenal ToxicantsAfter NCI-H295R cells were exposed to each test compoundduring three days the concentrations of 12 steroid hor-mones in the culture mediumwere simultaneously measuredby LCMSMS All effects of the compounds on adrenalsteroidogenesis were evaluated at the concentration withoutany overt cytotoxicity The differential metabolic steroidprofiles of 11 adrenal toxic compounds were classified andvisualized by using hierarchical clustering analysis (Figure 3)
Four adrenal toxicants without steroidogenic inhibitionAN SM TG and FN did not change the medium concen-trations of all steroid hormones by more than 2-fold Above-mentioned 4 compounds at every condition and 7 adrenalsteroidogenic inhibitors at the low exposure concentrationwere gathered into a big cluster as nonchange group The7 steroidogenic inhibitors at the maximum exposure con-centration showed the characteristic steroid profiles eachbut 100 120583M DZ and 10 120583M SP were classified as a clusterAGT drastically decreased the medium concentrations ofPREG HPREG DHEA PROG DCORTICO CORTICOand ALDO at 100 120583M DDD dose-dependently decreased themedium concentrations of PROG DCORTICO CORTICOCORT and ALDO at gt10 120583M and decreased PREG HPREGDHEA PROG HPROG DIONE and DCORT at the maxi-mum exposure concentration of 25120583M SP increased PREGHPREG and DHEA and decreased PROG DIONE DCOR-TICODCORTCORTICOALDO andTESTOat 10120583MMPdose-dependently decreased CORTICO CORT and ALDOand decreased DHEA HPROG DIONE and TESTO at themaximum exposure concentration of 100 120583M KC drasticallydecreased the medium concentrations of PREG HPREGDHEAHPROGDIONEDCORTICODCORTCORTICOCORTO ALDO and TESTO at 10 120583M MC increased themedium concentrations of PROG and decreased DIONEDCORT CORT and TESTO at 10 120583M DZ increased PREGHPREG and DHEA and decreased DIONE DCORTICODCORT CORTICO CORT ALDO and TESTO at 100 120583M
32 The Mathematical Modeling
321 Optimization of the Mathematical Model of AdrenalSteroidogenesis in NCI-H295R Cells The mathematicalmodel of adrenal steroidogenesis in NCI-H295R cells wasoptimized for several kinetic parameters of cholesteroltransport intracellular localization the oxysterol pathwayand maximum velocity of steroidogenic enzymes to fit theexperimental time-course data All optimized kinetic para-meters are shown in Table S1 in a supplementary documentThe optimized mathematical model was reconstructedwith the experimental dynamic patterns of cholesteroland the 14 steroid hormones in the intracellular space andculture medium The fitness was 0621761 of NLSD valuesas the fitting objective function The simulation results andexperimental data are shown in Figure 4
Optimized kinetic parameters were calculated sensitivi-ties for the NLSD value as the fitting score and are shown inTable S1 in a supplementary document The highly sensitiveparameters for fitting the NLSD score were the extractednine kinetic parameters 119896Cholesterol Transport 119896
acc119891 119896loc119891 119896loc119887
119881CYP11A1max 119870CYP11A1
119898 119881CYP17HmaxA 119881HSD3B2
maxA and 119881CYP21A2maxA whichhad higher than 30 fitting sensitivity
33 The Validation Using the Adrenal Toxicants
331 Mechanistic Analysis of Adrenal Toxicants Effects ofadrenal toxic compounds on steroidogenic enzymes werequantitatively predicted from the change in the ratio ofthe measured medium concentrations of the 12 steroid
8 Journal of Toxicology
0001
001
01
1
10
PREG
HPR
EG
DH
EA
PRO
G
HPR
OG
AN
DRO
DCO
RTIC
O
DCO
RT
CORT
ICO
CORT
ALD
O
TEST
O E1 E2C
once
ntra
tion
in m
ediu
m (120583
M)
lowast
lowast
lowast
lowast
lowast
lowast
lowast lowast lowastlowast
lowast
lowast
lowast
lowast
(a)
0 8 24 48 72Time (hours)
0
40
80
120
160
200
240
Num
ber o
f mol
ecul
es (n
mol
)
(b)
0 8 24 48 72Time (hours)
0
40
80
120
160
200
240
Num
ber o
f mol
ecul
es (n
mol
)
(c)
Figure 2 Experimental data of metabolic profiling of adrenal steroid hormones and the mass balance Concentrations of steroid hormonessecreted from NCI-H295R cells in the culture medium at 72 h after stimulation were compared with the untreated condition and thestimulated condition by 50 nM ACTH 20 120583M forskolin and 100 nM angiotensin II (a) Net molecular amounts including cholesterol andsteroid hormones in the culture medium and intracellular space are plotted at five time points (0 8 24 48 and 72 h after stimulation)under the untreated condition (b) and the stimulated condition (c) In the bar graphs dark and light bars indicate the amount of cholesteroland adrenal steroids respectively All data are shown as the mean plusmn SD (119873 = 4) lowast119901 values corrected by the familywise error ratelt001 PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-hydroxyprogesteroneDIONE androstenedioneDCORTICO 11-deoxycorticosteroneDCORT 11-deoxycortisol CORTICO corticosteroneCORT cortisol ALDO aldosterone and TESTO testosterone
hormones at 72 h after drug exposure using themathematicalmodel of adrenal steroidogenesis in NCI-H295R cells Thereproducibility of the estimated results was confirmed byperforming the test twice The estimated effects of 11 adrenaltoxic compounds on eight steroidogenic enzymes are shownin Figure 5 The adrenal toxic compounds without steroido-genic inhibition such as vasculotoxic agents (AN SM TGand FN) were not estimated for the target steroidogenicenzymes under noncytotoxic conditions Every fitness valueswere under 005 of NLSD values used as the fitting objectivefunction (Figures 5(a)ndash5(d)) Other steroidogenic inhibitors(AGT DDD SPMP KCMC andDZ) are described in detailbelow
332 AGT The mechanism of action of AGT in adrenalsteroidogenesis was estimated by inhibition of CYP11A1HSD3B2 CYP21A2 and CYP11B1 at 100 120583M (estimatedinhibitions were 770 780 811 and 598 resp)(Figure 5(e)) AGT has been reported to inhibit CYP11A1
CYP21A2 CYP11B1 and CYP11B2 [6 27ndash30] Our resultswere mostly consistent with the previous reports In particu-lar CYP11A1 appeared to be inhibited strongly byAGT In ourstudy HSD3B2 inhibition of AGT was shown by mechanisticanalysis based on systems biology approaches as a novelmechanism of action of AGT Inhibition of AGT by CYP11B2was not estimated in our study However the concentrationof ALDO in the culture medium decreased to 38 ofthe normal stimulated condition Inhibition of AGT byCYP11B2 has been shown using sheep adrenal homogenatesas well as a human adrenal homogenate from a patientwith Cushingrsquos syndrome [30]The activity of 18-hydroxylaseinduced by CYP11B2 was determined as the conversion ofcorticosterone to 18-hydroxycorticosterone in the previousstudy The cause of the discrepancy regarding the effect ofAGT on CYP11B2 was suggested to be substrate inhibitionbecause the intracellular concentration of CORTICO wasincreased by over 10 times of that in the culture mediumto reach 50 120583M Another possibility was poor quantitative
Journal of Toxicology 9
log2
ratio
4
3
2
1
0
minus1
minus2
minus3
minus4
ALDO
CORTICO
CORT
TESTO
DIONE
HPROG
DCORTICO
DCORT
PROG
PREG
HPREG
DHEA
DZ(
100)
SP(1
0)SP
(1)
DZ(
10)
DD
D(0
1)
AN
(01
)A
N(1
)A
N(1
0)SM
(00
1)TG
(00
1)M
P(0
1)FN
(01
)TG
(1)
AGT(
1)AG
T(0
1)SM
(00
0001
)D
DD
(1)
SM(0
1)
AGT(
10)
MP(
1)FN
(1)
MC(
1)D
Z(1)
KC(0
1)
MC(
01)
DZ(
01)
KC( 1
)D
DD
(10)
MP(
10)
MC(
10)
MP(
100)
AGT(
100)
DD
D(2
5)KC
(10)
Figure 3 Hierarchical cluster analysis of differential metabolic profiles of 12 steroid hormones by exposure to adrenal toxicity compoundsAdrenal toxicants were classified by using the differential metabolic profiling of 12 steroid hormones Concentrations of 12 adrenalsteroids secreted from NCI-H295R cells were drastically changed by the exposure of adrenal steroidogenesis inhibitors The 12 adrenalsteroid hormones were quantitatively measured by LC-MSMS simultaneously Four adrenal vasculotoxic compounds acrylonitrile ANfumaronitrile FN salinomycin SM and thioguanine TG were used as negative control compounds for adrenal steroidogenesis inhibitorsSeven adrenal steroidogenesis inhibitors aminoglutethimide AGT op1015840-DDD DDD spironolactone SP metyrapone MP ketoconazoleKC miconazole MC and daidzein DZ showed a characteristic steroid profile each and were classified as each independent singletonat the maximum exposure condition Exposure concentrations of adrenal toxic compounds were described in brackets of the samplename and the units were prepared as 120583M A blue cluster was classified as a group of nonchange samples including negative controlcompounds and low exposure conditions of adrenal steroidogenesis inhibitors PREG pregnenolone HPREG 17120572-hydroxypregnenoloneDHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-hydroxyprogesterone DIONE androstenedione DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICO corticosterone CORT cortisol ALDO aldosterone and TESTO testosterone
reliability of the experimental data because the ALDOconcentration was under the lower limit of quantification at100 120583MAGT Hecker and colleagues reported that 3 120583MAGTdecreases PREG and PROG concentrations and increases theTESTO concentration [10] However AGT did not increasethe TESTO concentration in our study One possibility isthat the concentration of TESTO was already enhanced byabout 33-fold through stimulation with ACTH forskolinand angiotensin II
333 DDD The mechanism of action of DDD in adrenalsteroidogenesis was estimated by dose-dependent inhibitionof CYP11A1 HSD3B2 CYP21A2 and CYP11B1 (estimatedinhibitions at 25120583M were 870 869 769 and 849resp) (Figure 5(f)) DDD has been reported to inhibitCYP11A1 HSD3B2 CYP21A2 CYP11B1 and CYP11B2 [29 3132] Inhibition of DDD by CYP11B2 was not estimated in ourstudy However the concentration of ALDO in the culturemedium decreased to 3 of that in the normal stimulatedcondition Inhibition of DDD by CYP11B2 has been shownusing mitochondrial and microsomal fractions prepared bystandard centrifugation procedures from a bovine adrenal
cortex homogenate [32]The cause of the discrepancy regard-ing the inhibition ofDDDbyCYP11B2 could not be explainedby same effect in the case of AGT
334 SP Themechanism of action of SP in adrenal steroido-genesis was estimated by inhibition of HSD3B2 CYP21A2and HSD17B3 (estimated inhibitions at 10120583M were 702595 and 593 resp) (Figure 5(g)) SP has been reportedto inhibit CYP17H CYP17L CYP11B1 and CYP11B2 [6 33ndash35] The inhibitory effect of SP on the HSD3B2 enzyme wasa novel mechanism of action The main action of SP is asa mineralocorticoid receptor (MR) antagonist SP has alsobeen reported to exert some off-target effects by binding toandrogen glucocorticoid and progesterone receptors [56ndash58] SP has been shown to inhibit the production of ALDOand CORT from PREG induced by angiotensin II in H295Rcells but the specific MR antagonist eplerenone did not showthe inhibitory effects [59] Therefore HSD3B2 inhibition bySP is not mediated via MR and the action might be directinhibition of HSD3B2 enzymes or a part of known off-targeteffects mediated through other nuclear hormone receptorsRegarding CYP17H and CYP17L our results were consistent
10 Journal of Toxicology
CHOL
0
20
40
60
80
100C
once
ntra
tion
in m
ediu
m (120583
M)
10 20 30 40 50 60 70 800Time (hours)
(a)
PREG
DCORT
DCORTICO
CORTICO
CORT
Con
cent
ratio
n in
med
ium
(120583M
)
10
4
2
0
8
6
10 20 30 40 50 60 70 800Time (hours)
(b)
Con
cent
ratio
n in
med
ium
(120583M
)
0
020406
081
1214
0 10 20 30 40 50 60 70 80Time (hours)
HPREG
DIONE
PROGHPROG
E1
(c)
DHEAE2
ALDO
TESTO
0
005
01
015
02
025
Con
cent
ratio
n in
med
ium
(120583M
)
0 10 20 30Time (hours)
40 50 60 70 80
(d)
CHOL
0
5000
10000
15000
20000
25000
30000
Con
cent
ratio
n in
cell
(120583M
)
10 20 30 40 50 60 70 800Time (hours)
(e)
Con
cent
ratio
n in
cell
(120583M
)
0
50
100
150
200
250
0 10 20 30 40 50 60 70 80Time (hours)
PREG
DCORT
DCORTICO
CORTICOCORT
(f)
0 10 20 30 40 50 60 70 80Time (hours)
HPREG
DIONE
PROGHPROG
E1
Con
cent
ratio
n in
cell
(120583M
)
3025201510
35
05
(g)
005
115
2
253
35
Con
cent
ratio
n in
cell
(120583M
)
0 10 20 30 40 50 60 70 80Time (hours)
DHEA
E2ALDO
TESTO
(h)
Figure 4 Comparison of time-course profiles of cholesterol and adrenal steroids produced by NCI-H295R cells between experimentallymeasured and simulated data To intuitively confirm reconstruction of the measured experimental data in the developed simulation modelof NCI-H295R cells dynamics of cholesterol and adrenal steroids produced by NCI-H295R cells were plotted to overlay experimentaldata with the simulated results Graphs show the dynamics of medium concentrations of cholesterol (a) and adrenal steroids ((b)ndash(d))and intracellular concentrations of cholesterol (e) and adrenal steroids ((f)ndash(h)) Steroid hormones were categorized into three groups byconcentration levels Major steroids were PREG pregnenolone DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICOcorticosterone and CORT cortisol ((b) and (f)) Moderate steroids were HPREG 17120572-hydroxypregnenolone PROG progesterone HPROG17120572-hydroxyprogesterone DIONE androstenedione and E1 estrone ((c) and (g)) Minor steroids were DHEA dehydroepiandrosteroneALDO aldosterone TESTO testosterone and E2 17120573-estradiol ((d) and (h)) Experimental data are shown as symbols with dotted lines Alldata represent the mean plusmn SD (119873 = 4) Simulation data are shown as solid lines
Journal of Toxicology 11
0110
10
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
AN (120583M)
(a)
0110
FN (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(b)
00101
SM (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
100E minus 05
(c)
0110
TG (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B1
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(d)
0110
10100
AGT (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(e)
0110
1025
DDD (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(f)
1010
SP (120583M)
2750005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(g)
0110
10100
MP (120583M)
2110005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(h)
Figure 5 Continued
12 Journal of Toxicology
0110
10
KC (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(i)
0110
10
MC (120583M)
4430005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(j)
0110
10100
DZ (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(k)
Figure 5 Estimated mechanism of action of adrenal toxicants by using the mathematical model of adrenal steroidogenesis in NCI-H295Rcells Mechanisms of action of adrenal toxicity compounds were quantitatively estimated from experimental results of differential steroidprofiling using the mathematical model of adrenal steroidogenesis in NCI-H295R cells The drug action was defined as a scaling factorof enzymatic activity in the simulation model These scaling factors were optimized to fit experimental data by a hybrid optimizationmethod of the RCGA and nonlinear least squares Estimated drug actions by the exposure of vasculotoxic agents acrylonitrile AN (a)fumaronitrile FN (b) salinomycin SM (c) and thioguanine TG (d) and the steroidogenic inhibitors aminoglutethimide AGT (e) op1015840-DDD DDD (f) spironolactone SP (g) metyrapone MP (h) ketoconazole KC (i) miconazole MC (j) and daidzein DZ (k) are shownas a spider radar chart CYP11A1 P450 side chain cleavage enzyme CYP17H 17120572-hydroxylase of CYP17 CYP17L C
17ndash20 lyase of CYP17HSD3B2 3120573-hydroxysteroid dehydrogenase CYP21A2 21-hydroxylase CYP11B1 11120573-hydroxylase CYP11B2 18-hydroxylase and HSD17B317120573-hydroxysteroid dehydrogenase
with previous reports [33 34] 7120572-thiospironolactone whichis synthesized by deacetylation of SP inhibits CYP17H andCYP17L [34] The fact that there were no inhibitions ofCYP17HorCYP17L in our study suggests that SPmight not bedeacetylated to 7120572-thiospironolactone in NCI-H295R cellsRegarding CYP11B1 and CYP11B2 our results were unclearcompared with a previous study It has been shown that30 120583M SP inhibits CYP11B1 and CYP11B2 in human andbovine adrenal mitochondria [35]The cause of CYP11B1 andCYP11B2 inhibition by SP could not be determined in ourstudy which might be due to the lower maximum exposureconcentration of SP than that in the previous reportWe couldnot examine SP concentrations over 10 120583M because theseconcentrations were cytotoxic in NCI-H295R cells
335 MP The mechanism of action of MP in adrenalsteroidogenesis was estimated by dose-dependent inhibitionof CYP11B1 (estimated inhibitions at 1 10 and 100 120583M were571 827 and 982 resp) (Figure 5(h)) MP has beenreported to inhibit CYP11B1 as its major effect and CYP11A1and CYP11B2 as a weak effect [6 29 36ndash39] The results wereable to show that MP is a selective inhibitor of CYP11B1 inthe previous report However the estimated effect of MPat a high concentration 100 120583M as the maximum exposureconcentration was unclear According to the previous reportselectivity of MP for CYP11B1CYP11B2 is about five times[39] In addition 20 120583M MP has a slight inhibition effect onCYP11A1 in H295R cells [29]
336 KC The mechanism of action of KC in adrenalsteroidogenesis was estimated by inhibition of CYP11A1CYP17H CYP17L HSD3B2 CYP21A2 CYP11B1 and
CYP11B2 (estimated inhibitions at 10 120583Mwere 926 943518 830 882 974 and 798 resp) (Figure 5(i))KC has been reported to inhibit CYP11A1 CYP17H CYP17LHSD3B2 CYP21A2 and CYP11B1 [6 29 40ndash43] Our resultswere almost consistent with the previous reports KC inhibitsCYP11A1 CYP17H CYP21A2 and CYP11B1 in NCI-H295Rcells at 10 120583M [29] and CYP17H CYP17L CYP21A2 andCYP11B1 in human adrenal mitochondria and Leydig cellmicrosomes at 2ndash5120583M [60 61] However KC has shownonly weak inhibition of HSD3B2 and HSD17B3 in Leydigcells at the millimolar level [60 61] Regarding CYP11B2 andHSD17B3 we considered that these estimated inhibitions ofKC did have sufficient reliability in terms of quantitative pre-diction precision because ALDO and TESTO concentrationswere less than the lower limit of quantification at 10 120583MKC
337 MC Themechanism of action of MC in adrenal stero-idogenesis was estimated by inhibition of CYP17H CYP17LCYP11B1 and HSD17B3 (estimated inhibitions at 10120583Mwere691 530 764 and 571 resp) (Figure 5(j)) MC hasbeen reported to inhibit not only CYP17H and CYP17L butalsoCYP11A1 CYP21A2 andCYP11B1 [41 44 45]The resultsin the previous reports were able to estimate that MC isa CYP17 inhibitor However CYP11A1 inhibition by MCprobably instead of a reduction in StAR expression wasnot clearly detected in our study using NCI-H295R cellsbecause there were no decreases in the concentrations ofPREG and PROG in the culture medium Indirect inhibitionof CYP11A1 via the peripheral-type benzodiazepine receptorhas been reported in mouse adrenocortical Y-1 cells treatedwith MC in the absence of stimuli by measuring PREGproduction [44] On the other hand reductions of StAR
Journal of Toxicology 13
protein expression andor transport activity without affectingtotal steroid synthesis or CYP11A1 and HSD3B2 enzymeexpression or activities have been reported in (BU)
2cAMP-
stimulated MA-10 Leydig tumor cells treated with MC bymeasuring PROG production [45] Therefore the effect ofMC on the initial reaction in adrenal steroidogenesis fromcholesterol should be different according to the cell type andstimulation condition Inhibition of CYP21A2 and CYP11B1by MC has been reported as decreases in the consumptionof PROG and DCORTICO respectively [41] Inhibition ofCYP11B1 was estimated by the action of MC in this studybut that of CYP21A2 was not detected In the previousexperimental report inhibitory sites by MCmight have beenreflected by inhibition of CYP17H activity because CYP21A2activity was measured as a decrease in labeled PROG
338 DZ The mechanism of action of DZ in adrenalsteroidogenesis was estimated by inhibition of CYP11A1HSD3B2 CYP21A2 CYP11B1 and HSD17B3 (estimated inhi-bitions at 100 120583M were 586 941 965 872 and981 resp) (Figure 5(k)) DZ has been reported to inhibitHSD3B2 and CYP21A2 [46] The results of HSD3B2 andCYP21A2 were consistent with the previous report Howeverinhibitions have not been reported for CYP11A1 CYP11B1and HSD17B3These estimated effects of DZ on CYP11B1 andHSD17B3 were unclear because the concentrations of ALDOand TESTO were less than the lower limit of quantificationat 100 120583M In addition these enzymes act downstream of thestrong action points of DZ such as HSD3B2 and CYP21A2
34 The Simulations and the Systematic Model Analysis
341 Dynamic Sensitivity Analysis of Adrenal SteroidogenesisTo comprehensively understand the dynamics of adrenalsteroidogenesis dynamic sensitivities were calculated forsteroid concentrations secreted by NCI-H295R cells usingour constructed mathematical model of steroidogenesis Theresults of dynamic sensitivity analysis at 72 h of duration and6 h of interval time are presented as a heat-map in Figure 6
The top 10 parameters of the total area under the curveof dynamic sensitivity for cholesterol and the 14 steroids inculture medium were 119881HSD3B2
maxA 119881CYP11A1max 119881CYP21A2maxA 119896loc119891 119896loc119887
119870CYP11A1119898
119881CYP17HmaxA 119896Cholesterol Transport 119896acc119891 and 119881CYP21A2maxB in
order from the top Cholesterol uptake (119896Cholesterol Transport)StAR protein (119896loc
119891) and CYP11A1 (119881CYP11A1max ) which are
determining factors of the capacity for steroidogenesis pro-moted the production of mineralocorticoids (DCORTICOCORTICO and ALDO) and restrained the synthesis of glu-cocorticoids (DCORT and CORT) and sex steroids (DIONETESTO and E1) because of the accumulation of interme-diate molecules in steroidogenesis (PREG HPREG DHEAPROG and HPROG) only by self-activation The dynamicpatterns of the intermediate molecules in steroidogenesiswere mainly dependent on the activity of CYP17H andHSD3B2 with PREG as the substrate of these enzymes inwhich the dynamic sensitivities of 119881CYP17HmaxA for HPREG andHPROG and 119881HSD3B2
maxA for PROG HPROG and DCORTICO
reversed the direction of sensitivity at 49ndash66 h after stimu-lation The dynamic sensitivities of the maximum activitiesof HSD3B2 for PREG (119881HSD3B2
maxA ) and CYP21A2 for PROG(119881CYP21A2maxA ) were related to all steroids at 72 h Almost allmodel parameters had positive sensitivity for downstreamsteroids in the adrenal steroidogenic pathway and negativesensitivity for direct-binding steroids as substrates of thesteroidogenic enzyme The sensitivity of 119881max in all steroido-genic enzymes was relatively higher than 119870
119898for the same
steroid substrate
342 Simulation of the Metabolic Balance of AdrenalSteroidogenesis Pathway To clearly show the property of themetabolic shift between mineralocorticoid and glucocorti-coid biosynthesis we performed two-dimensional parameterscanning of the enzymatic activities of CYP17H and HSD3B2(Figure 7) NCI-H295R cells lost the ability to produce allsteroid hormones when enzymatic activities of CYP17H andHSD3B2 were changed by over 60 and 30 respectivelyActivation ofCYP17HandorHSD3B2 induced themetabolicshift that enhanced the glucocorticoid biosynthesis and devi-ated from the mineralocorticoid biosynthesis On the otherhand inhibition of CYP17H andor HSD3B2 induced themetabolic shift that enhanced the mineralocorticoid biosyn-thesis and deviated from the glucocorticoid biosynthesisMoreover the enzymatic activity of HSD3B2 regulated themetabolic balance of sex steroids and the precursors onadrenal steroidogenesis of NCI-H295R cells E1 TESTO andDIONE were produced by NCI-H295R cells when activatingthe enzymatic activity of HSD3B2 Conversely E2 andDHEAwere produced by NCI-H295R cells when suppressing theenzymatic activity of HSD3B2 The biosynthesis of down-stream steroids in adrenal steroidogenesis pathway such asmineralocorticoids and glucocorticoids was almost com-pletely terminated when the enzymatic activity of HSD3B2was decreased by over 80
4 Discussion
41 Importance of 3120573-HSDActivity in Adrenal SteroidogenesisOur systematic analysis using the mathematical model ofadrenal steroidogenesis in NCI-H295R cells revealed thatthe enzymatic activity of 3120573-HSD controls the dynamics ofadrenal steroidogenesis The activity of the StAR proteincontrols the net capacity of steroidogenesis in steroidogeniccells which is the transport of cholesterol from the outer toinner mitochondrial membranes Both the expression levelsof StAR protein and mRNA are rapidly elevated in responseto stimulation by tropic hormones such as ACTH [62 63]Another important factor in adrenal steroidogenesisis is thecholesterol side chain cleavage enzyme CYP11A1 the firstrate-limiting and hormonally regulated step in the synthesisof all steroid hormones which is conversion of cholesterolto pregnenolone in mitochondria [64] According to ourresults of global sensitivity analysis (Supplementary Table 1and Figure 6(d)) in addition to CYP11A1 and StAR proteins3120573-HSD was one of the key regulators in adrenal steroido-genesis of NCI-H295R cells And also this result suggeststhat a significant regulatory mechanism in steroidogenesis
14 Journal of Toxicology
60544830 4236 66 7218 246 120
Time (hours)
0
1
2
3C
once
ntra
tion
(a)
60544830 4236 66 7218 246 120
Time (hours)
minus05
0
05
minus1
1
Sens
itivi
ty
(b)
0
6
12
18
24
30
36
42
48
54
60
66
72
Sens
itivi
ty fo
r med
ium
conc
entr
atio
ns
of ch
oles
tero
l or s
tero
id h
orm
ones
Dur
atio
n tim
e (ho
urs)
Model parameters
+10
+05
00
minus05
minus10
(c)
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
Parameter number in the mathematical model of steroidogenesis in NCI-H295R cells
CHOL
PREG
HPREG
DHEA
PROG
HPROG
DIONE
DCORTICO
DCORT
CORTICO
CORT
ALDO
TESTO
E1
E2
Chol
este
rol o
r ste
roid
hor
mon
es
0
05
10
15
20
minus05
minus10
minus15
minus20
(d)
Figure 6 Heat-map of the global dynamic sensitivity analysis of adrenal steroid concentrations produced by NCI-H295R cells The globaldynamic sensitivity analysis is a powerful tool to comprehensively understand the dependencies of themodel parameters in themathematicalmodel of the biological complex system Dynamic sensitivities of model parameters in the mathematical model of adrenal steroidogenesisin NCI-H295R cells were calculated for all steroid concentrations in the culture medium every 6 h until 72 h after stimulation To clarify theview of heat-map of global dynamic sensitivity analysis imaginary data of the dynamics of steroid concentrations in the original model (blueline) and perturbed model for sensitivity analysis (red line) were prepared (a) Using this imaginary data the calculated dynamic sensitivities(b) and the visualized dynamic sensitivity as one block of the heat-map (c) were shown respectively By the same method that explainedusing imaginary data the large-scale data of the global dynamic sensitivity analysis on the mathematical model of adrenal steroidogenesisin NCI-H295R cells was comprehensively visualized as a big graph of heat-map (d) Parameter numbers in the horizontal axis are (1)119896Choresterol Transport (2) 119896CEH (3) 119896MTR
119891 (4) 119896MTR
119887 (5) 119896acc
119891 (6) 119896acc
119887 (7) 119896loc
119891 (8) 119896loc
119887 (9) 119896Oxysterol Synthesis (10)119870
CYP11A1119898
(11) 119881CYP11A1max (12)119870CYP17H119898A
(13) 119870CYP17H119898B (14) 119881CYP17HmaxA (15) 119881CYP17HmaxB (16) 119881CYP17HmaxB (17) 119870CYP17L
119898B (18) 119881CYP17LmaxA (19) 119881CYP17LmaxB (20) 119870HSD3B2119898A (21) 119870HSD3B2
119898B (22) 119870HSD3B2119898C (23)
119881HSD3B2maxA (24) 119881HSD3B2
maxB (25) 119881HSD3B2maxB (26) 119870CYP21A2
119898A (27) 119870CYP21A2119898B (28) 119881CYP21A2maxA (29) 119881CYP21A2maxB (30) 119870CYP11B1
119898A (31) 119870CYP11B1119898B (32) 119881CYP11B1maxA
(33) 119881CYP11B1maxB (34) 119896CYP11B2 (35) 119870HSD17B3119898A (36) 119870HSD17B3
119898B (37) 119881HSD17B3maxA (38) 119881HSD17B3
maxB (39) 119870CYP19A1119898A (40) 119870CYP19A1
119898B (41) 119881CYP19A1maxA and (42)119881
CYP19A1maxB PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-
hydroxyprogesteroneDIONE androstenedioneDCORTICO 11-deoxycorticosteroneDCORT 11-deoxycortisol CORTICO corticosteroneCORT cortisol ALDO aldosterone and TESTO testosterone
pathway is very reasonable StAR CYP11A1 and 3120573-HSD(isoforms 1 or 2 in humans) proteins generally respond to thesame hormones that stimulate steroid production throughcommon pathways such as cAMP signaling in adrenal glandsand testes [65 66] Moreover our data also support recentexperimental evidence from clinical and in vivo studiessuggesting that the enzymatic activity of 3120573-HSD plays
an important role in the regulation of mineralocorticoidsynthesis in adrenal steroidogenesis and contributes to hyper-tension caused by abnormal overproduction of aldosterone[67ndash70] Circadian clock-deficient Cry-null mice show salt-sensitive hypertension due to abnormally high synthesis ofaldosterone which is caused by constitutively high expressionof HSD3B6mRNA and protein in the adrenal cortex [67 68]
Journal of Toxicology 15
All steroid producing cells
Mineralocorticoid producing cells Glucocorticoid producing cells
Mineralocorticoid and glucocorticoid producing cells
Terminated steroidogenesis
Sex steroids and precursor sex steroidsDIONE TESTO E1DEHA E2
NCI-H295Rcells at controlcondition
Normalized activities of HSD3B2
Nor
mal
ized
activ
ities
of C
YP17
H
2
18
16
14
12
1
08
06
04
02
00 05 1 15 2
Figure 7 Metabolic categories of the steroidogenic cells determined by the balance of HSD3B2 and CYP17H activitiesThe two-dimensionalparameter scanning analysis by the perturbation of two focused parameters clarifies the interaction and the relationship between the twoparameters in the complex system Functional cellular categories of steroidogenic cells were defined by the levels of mineralocorticoid(ALDO) glucocorticoids (DCORTandCORT) and androgens (DHEAandDIONE) at 72 h after stimulation Enzymatic activities ofHSD3B2andCYP17were normalized by standard values of the simulationmodel inNCI-H295R cells Scanning ranges ofHSD3B2 andCYP17 activitieswere 0ndash200 each 10 Green regions are all 14 steroids produced by NCI-H295R cells Red blue and purple regions are mineralocorticoidproducing cells glucocorticoids-producing cells and both corticoid-producing cells respectively Yellow regions are steroidogenesis of NCI-H295R cells terminated upstream of the adrenal steroidogenesis pathway
Recent clinical observations have revealed predominantexpression of HSD3B2 mRNA and protein in tumor cellsof aldosterone-producing adenoma (APA) and HSD3B1mRNA significantly correlated with CYP11B2 mRNA levelsand plasma aldosterone concentrations in APA patients [6970] However the relationship is unclear and disputed ina small-scale clinical study indicating that genetic variationin HSD3B1 affects blood pressure and hypertension in APApatients [71] The results of our simulation study suggest that3120573-HSD protein (human genes are HSD3B1 and HSD3B2) isone of the determination factors for the dynamic propertyof adrenal steroidogenesis Our results support the clinicalevidence of Doi and colleagues [69] and we believe that theHSD3B1 enzyme has a promising potential as novel drugtarget for endocrine hypertension
42 Metabolic Shift of Adrenal Steroidogenesis and the Contri-butions of HSD3B2 and CYP17H The metabolic propertiesof adrenal steroidogenesis in NCI-H295R cells were revealedby dynamic sensitivity analysis using themathematicalmodel(Figure 6) Mineralocorticoids such as DCORTICO COR-TICO and ALDO and intermediate steroids upstream of
the adrenal steroidogenesis pathway such as PREG HPREGDHEA PROG and HPROG were accelerated by reactionsof cholesterol import (119896Cholesterol Transport) StAR protein (119896MTR
119891
and 119896loc119891) and CYP11A1 (119881CYP11A1max ) On the other hand glu-
cocorticoids such as DCORT and CORT and sex hormonessuch as DIONE TESTO and E1 were suppressed by thesemodel parametersTherefore enhancement of the net adrenalsteroidogenesis capacity which supplies PREG precursor tothe pathway causes a production shift from glucocorticoidsto mineralocorticoids by substrate inhibitions of CYP17HHSD3B2 and CYP21A2 caused by accumulation of initialintermediate steroids such as PREG and PROG Sensitivi-ties of CYP17H (119881CYP17HmaxA ) and HSD3B2 (119881HSD3B2
maxA ) for theproducts were dynamically changed and these parametersdetermined the metabolic balance of downstream steroidsin the adrenal steroidogenesis pathway According to theseresults of dynamic sensitivity analysis of StAR CYP11A1CYP17H and HSD3B2 we suggest that the enhancementof CYP17H and HSD3B2 activity during ACTH stimulationwas important to shift the steroidogenic output away fromALDO biosynthesis towards CORT biosynthesis as wellas adrenal androgen production This suggestion partially
16 Journal of Toxicology
supports a comparative animal study in which molecularand cellular variations in CYP17H activity dramatically affectacute cortisol production resulting in distinct physiologicaland behavioral responses [72]
Results of two-dimensional parameter scanning of theenzymatic activities of CYP17H and HSD3B2 quantitativelyshowed the detail of the metabolic relationship betweenmin-eralocorticoid and glucocorticoid biosynthesis (Figure 7)Particularly the results showed that the balance of theseenzymatic activities was very important for the typicalfunction of NCI-H295R cells namely the ability to produceall steroid hormones NCI-H295R cells lost this functionwhen enzymatic activities of CYP17H and HSD3B2 werechanged by over 60 and 30 respectively In addition theybecame mineralocorticoid (ALDO) secreting cells when theenzymatic activity of CYP17H or HSD3B2 was inhibited byover 50 or glucocorticoid (DCORT and CORT) secretingcells when these enzymes were activated by over 50 Inparticular this analysis also showed that HSD3B2 was a keyplayer in the adrenal steroidogenesis of NCI-H295R cellsbecause HSD3B2 inhibition by over 80 almost completelyinhibited the biosynthesis of downstream steroids The ratioof CYP17A1 to HSD3B2 mRNA expression levels has beenrelated to several endocrine diseases with a low level in APAs[73] and high level in cortisol-producing adenomas [74]Furthermore the expression levels or enzymatic activitiesof CYP17A1 and HSD3B1 have been related to androgenproduction in polycystic ovary syndrome [75 76] Theseclinical studies support our simulation results indicating thatthe balance of enzymatic activity of CYP17H and HSD3B2determines the shift in steroidogenic output to mineralocor-ticoids glucocorticoids or androgens
43 Methodologies of Quantitative Mechanistic Analysis forDrug Discovery According to our results obtained using themathematical model of steroidogenesis in NCI-H295R cellssuch as sensitivity analysis comprehensive analysis based onsystems biology is available to quantitatively estimate themechanism of action of steroidogenic disrupting compoundsfrom differential profiling of adrenal steroid hormonesbecause dynamic patterns of steroid hormones in adrenalsteroidogenesis pathway are highly complex Our proposedmethod of quantitative mechanistic analysis of steroidogenicinhibitors was able to predict known action sites in theadrenal steroidogenesis pathway at only one time point (72 hafter drug exposure) Moreover according to the resultsof sensitivity analysis (Figure 6) 119881max of all steroidogenicenzymes was more sensitive than the 119870
119898 because the
intracellular concentrations of steroid hormones were almostmaintained at sufficiently high levels compared with 119870
119898
values of steroidogenic enzymesThese results suggested thatestimation of the mechanism of action of drugs is moreeffective and detectable when using the influences of 119881maxas the searching parameters such as our proposed methodOur data showed that the proposed method based on asystems biology model is a very powerful tool for exploratoryscreening of steroidogenic disrupting compounds
RCGA as a solver of parameter estimation problemsin systems biology has been applied to biological network
identification of gene regulatory networks and metabolicpathways and optimization of biological processes usingexperimentally observed time-course data [77ndash82] In thisstudy RCGA was useful to estimate the mechanism ofaction of novel pharmaceutical drug candidates for adrenalsteroidogenesis as a new application of RCGA in systemsbiology We had two issues when applying RCGA to thequantitative mechanistic analysis of drug actions Theseissues were the vast calculation cost and multimodality ofquasi-optimum solutions in solving the optimization prob-lem because the mathematical model in systems biologyconsists of many equations and parameters A proposedoptimization strategy using RCGA based on REXJGG wasa highly stable and efficient calculation method for a betterquasi-optimum solution than the unimodal normal distri-bution crossover (UNDX)minimum generation gap (MGG)method that is well applied in the engineering field In addi-tion we expanded the RCGA optimization program basedon REXJGG to a hybrid method of GA and then applieda local search as recommended by Harada and Kobayashi[28 83] A final optimal solution was obtained with a goodconvergence property Because these problems are generalin systems biology studies we suggest that the proposedhybrid method based on REXJGG is a very useful tool forquantitative mechanistic analysis of novel pharmaceuticaldrugs not limited to steroidogenic disrupting compounds
5 Conclusions
Thenovelmathematicalmodel of adrenal steroidogenesis wasconstructed in this study including cholesterol transport anddistribution the C
21-steroid hormone pathway steroid trans-
port and cell proliferation which could reproduce adrenalsteroidogenesis in NCI-H295R cells According to the resultsof dynamic sensitivity analysis using the newmodel HSD3B2plays the most important role in the metabolic balance ofadrenal steroidogenesis in NCI-H295R cells Moreover toquantitatively estimate mechanisms of action of adrenal toxiccompounds we analyzed differential metabolic profiles of12 steroid hormones at 3 days after exposure to 11 adrenaltoxic compounds by using the new mathematical modeland a hybrid optimization method of the RCGA and a localsearch (nonlinear least squares) We could estimate whichsteroidogenic enzymes were affected by these compoundsusing the hybrid optimization method Vasculotoxic agentswere estimated to have no effect according to the resultsobtained by our method In terms of adrenal steroidogenicinhibitors the predicted action sites were approximatelymatched to the target enzymes as reported in the literatureThus our computer-aided method based on a systems biol-ogy approach may be useful to analyze the mechanism ofaction of endocrine-disrupting compounds and to assess thehuman adrenal toxicity of novel pharmaceutical drugs basedon steroid hormone profiling
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Journal of Toxicology 17
Authorsrsquo Contribution
Ryuta Saito Natsuko Terasaki and Naohisa Tsutsui designedthe study Naohisa Tsutsui and Naoya Masutomi supervisedthe experiments Natsuko Terasaki performed all in vitroexperiments and collected dataMakotoYamazaki performedall measurements of steroids Ryuta Saito conceived the insilico strategies analyzed data developed all programs andthe mathematical model and performed all simulation anal-yses with Masahiro Okamoto Ryuta Saito Natsuko TerasakiMakoto Yamazaki NaoyaMasutomi andMasahiroOkamotowrote the paper
Acknowledgments
The authors would like to thank Dr Michael S Breenand Dr Miyuki Breen for helpful discussions concerningprevious mathematical models of adrenal steroidogenesisThis work was partially supported by the Grant-in-Aid forScientific Research on Innovative Areas ldquoSynthetic Biologyrdquo(No 23119001 (MO)) from Ministry of Education CultureSports Science and Technology in Japan
References
[1] P W Harvey D J Everett and C J Springall ldquoAdrenal toxi-cology a strategy for assessment of functional toxicity to theadrenal cortex and steroidogenesisrdquo Journal of Applied Toxi-cology vol 27 no 2 pp 103ndash115 2007
[2] W E Ribelin ldquoThe effects of drugs and chemicals upon thestructure of the adrenal glandrdquo Fundamental and Applied Toxi-cology vol 4 no 1 pp 105ndash119 1984
[3] H D Colby and P A Longhurst ldquoToxicology of the adrenalglandrdquo in Endocrine Toxicology C K Atterwill and J D FlackEds pp 243ndash281 CambridgeUniversity Press Cambridge UK1992
[4] H D Colby ldquoIn vitro assessment of adrenocortical toxicityrdquoJournal of Pharmacological and Toxicological Methods vol 32no 1 pp 1ndash6 1994
[5] T J Rosol J T Yarrington J Latendresse and C C CapenldquoAdrenal gland structure function and mechanisms of toxic-ityrdquo Toxicologic Pathology vol 29 no 1 pp 41ndash48 2001
[6] P W Harvey and D J Everett ldquoThe adrenal cortex and ster-oidogenesis as cellular and molecular targets for toxicity crit-ical omissions from regulatory endocrine disrupter screeningstrategies for human healthrdquo Journal of Applied Toxicology vol23 no 2 pp 81ndash87 2003
[7] A F Gazdar H K Oie C H Shackleton et al ldquoEstablishmentand characterization of a human adrenocortical carcinoma cellline that expresses multiple pathways of steroid biosynthesisrdquoCancer Research vol 50 no 17 pp 5488ndash5496 1990
[8] W E Rainey I M Bird C Sawetawan et al ldquoRegulation ofhuman adrenal carcinoma cell (NCI-H295) production of C19steroidsrdquo The Journal of Clinical Endocrinology amp Metabolismvol 77 no 3 pp 731ndash737 1993
[9] B Staels D W Hum andW L Miller ldquoRegulation of steroido-genesis in NCI-H295 cells a cellular model of the human fetaladrenalrdquo Molecular Endocrinology vol 7 no 3 pp 423ndash4331993
[10] M Hecker J L Newsted M B Murphy et al ldquoHuman adreno-carcinoma (H295R) cells for rapid in vitro determination of
effects on steroidogenesis hormone productionrdquo Toxicologyand Applied Pharmacology vol 217 no 1 pp 114ndash124 2006
[11] A Oskarsson E Ulleras K E Plant J P Hinson and P SGoldfarb ldquoSteroidogenic gene expression in H295R cells andthe human adrenal gland adrenotoxic effects of lindane invitrordquo Journal of Applied Toxicology vol 26 no 6 pp 484ndash4922006
[12] UMuller-Vieira M Angotti and RW Hartman ldquoThe adreno-cortical tumor cell line NCI-H295R as an in vitro screeningsystem for the evaluation of CYP11B2 (aldosterone synthase)and CYP11B1 (steroid-11120573-hydroxylase) inhibitorsrdquoThe Journalof Steroid Biochemistry and Molecular Biology vol 96 no 3-4pp 259ndash270 2005
[13] K Imagawa S Okayama M Takaoka et al ldquoInhibitory effectof efonidipine on aldosterone synthesis and secretion in humanadrenocarcinoma (H295R) cellsrdquo Journal of CardiovascularPharmacology vol 47 no 1 pp 133ndash138 2006
[14] M Voets UMuller-Vieira S Marchais-Oberwinkler and RWHartmann ldquoSynthesis of amidinohydrazones and evaluation oftheir inhibitory effect towards aldosterone synthase (CYP11B2)and the formation of selected steroidsrdquo Archiv der Pharmazievol 337 no 7 pp 411ndash416 2004
[15] F K Nielsen C H Hansen J A Fey et al ldquoH295R cells as amodel for steroidogenic disruption a broader perspective usingsimultaneous chemical analysis of 7 key steroid hormonesrdquoToxicology in Vitro vol 26 no 2 pp 343ndash350 2012
[16] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cortisolproduction and on activities of steroidogenic enzymes inhuman adrenocortical H295R cellsrdquo Journal of Steroid Biochem-istry and Molecular Biology vol 80 no 3 pp 355ndash363 2002
[17] R F Canton J T Sanderson S Nijmeijer A Bergman R JLetcher and M van den Berg ldquoIn vitro effects of brominatedflame retardants and metabolites on CYP17 catalytic activity anovel mechanism of actionrdquo Toxicology and Applied Pharma-cology vol 216 no 2 pp 274ndash281 2006
[18] K Hilscherova P D Jones T Gracia et al ldquoAssessment ofthe effects of chemicals on the expression of ten steroidogenicgenes in the H295R cell line using real-time PCRrdquo ToxicologicalSciences vol 81 no 1 pp 78ndash89 2004
[19] T Gracia K Hilscherova P D Jones et al ldquoThe H295R systemfor evaluation of endocrine-disrupting effectsrdquo Ecotoxicologyand Environmental Safety vol 65 no 3 pp 293ndash305 2006
[20] X Zhang J L NewstedM Hecker E B Higley P D Jones andJ P Giesy ldquoClassification of chemicals based on concentration-dependent toxicological data using ToxClustrdquo EnvironmentalScience and Technology vol 43 no 10 pp 3926ndash3932 2009
[21] M S Breen M Breen N Terasaki M Yamazaki and R BConolly ldquoComputational model of steroidogenesis in humanH295R cells to predict biochemical response to endocrine-active chemicalsmodel development formetyraponerdquoEnviron-mental Health Perspectives vol 118 no 2 pp 265ndash272 2010
[22] M Breen M S Breen N Terasaki M Yamazaki A LLloyd and R B Conolly ldquoMechanistic computational model ofsteroidogenesis in H295R cells role of oxysterols and cell pro-liferation to improve predictability of biochemical response toendocrine active chemical-metyraponerdquo Toxicological Sciencesvol 123 no 1 pp 80ndash93 2011
[23] F B Kraemer ldquoAdrenal cholesterol utilizationrdquo Molecular andCellular Endocrinology vol 265-266 pp 42ndash45 2007
[24] W L Miller ldquoSteroidogenic acute regulatory protein (StAR)a novel mitochondrial cholesterol transporterrdquo Biochimica et
18 Journal of Toxicology
Biophysica Acta (BBA)mdashMolecular and Cell Biology of Lipidsvol 1771 no 6 pp 663ndash676 2007
[25] S Szabo and I T Lippe ldquoAdrenal gland chemically inducedstructural and functional changes in the cortexrdquo ToxicologicPathology vol 17 no 2 pp 317ndash329 1989
[26] C Chen-Pan I-J Pan Y Yamamoto T Sakogawa J Yamadaand YHayashi ldquoPrompt recovery of damaged adrenalmedullaeinduced by salinomycinrdquoToxicologic Pathology vol 27 no 5 pp563ndash572 1999
[27] R J Santen and R I Misbin ldquoAminoglutethimide review ofpharmacology and clinical userdquo Pharmacotherapy vol 1 no 2pp 95ndash120 1981
[28] M J Carella N V Dimitrov V V Gossain L Srivastava andD R Rovner ldquoAdrenal effects of low-dose aminoglutethimidewhen used alone in postmenopausal women with advancedbreast cancerrdquoMetabolism vol 43 no 6 pp 723ndash727 1994
[29] M K Johansson J T Sanderson and B-O Lund ldquoEffects of3-MeSO2-DDE and some CYP inhibitors on glucocorticoidsteroidogenesis in the H295R human adrenocortical carcinomacell linerdquo Toxicology in Vitro vol 16 no 2 pp 113ndash121 2002
[30] Y Touitou A Bogdan J C Legrand and P Desgrez ldquoAmino-glutethimide and glutethimide effects on 18-hydroxycorti-costerone biosynthesis by human and sheep adrenals in vitrordquoActa Endocrinologica vol 80 no 3 pp 517ndash526 1975
[31] M M Hart and J A Straw ldquoStudies on the site of action ofop1015840-DDD in the dog adrenal cortex 1 Inhibition of ACTH-mediated pregnenolone synthesisrdquo Steroids vol 17 no 1ndash5 pp559ndash574 1971
[32] M Ojima M Saitoh N Itoh Y Kusano S Fukuchi and HNaganuma ldquoThe effects of op1015840-DDD on adrenal steroidogen-esis and hepatic steroid metabolismrdquo Nippon Naibunpi Gakkaizasshi vol 61 no 3 pp 168ndash178 1985
[33] L F Canosa and N R Ceballos ldquoEffects of different steroid-biosynthesis inhibitors on the testicular steroidogenesis of thetoad Bufo arenarumrdquo Journal of Comparative Physiology B vol171 no 6 pp 519ndash526 2001
[34] D C Kossor S Kominami S Takemori and H D ColbyldquoRole of the steroid 17 alpha-hydroxylase in spironolactone-mediated destruction of adrenal cytochrome P-450rdquoMolecularPharmacology vol 40 no 2 pp 321ndash325 1991
[35] S C Cheng K Suzuki W Sadee and B W Harding ldquoEffectsof spironolactone canrenone and canrenoate K on cytochromeP450 and 11120573 and 18 hydroxylation in bovine and humanadrenal corticalmitochondriardquo Endocrinology vol 99 no 4 pp1097ndash1106 1976
[36] K Yanagibashi M Haniu J E Shively W H Shen and P HallldquoThe synthesis of aldosterone by the adrenal cortex Two zones(fasciculata and glomerulosa) possess one enzyme for 11120573- 18-hydroxylation and aldehyde synthesisrdquoThe Journal of BiologicalChemistry vol 261 no 8 pp 3556ndash3562 1986
[37] K Morishita H Okumura N Ito and N Takahashi ldquoPrimaryculture system of adrenocortical cells from dogs to evaluatedirect effects of chemicals on steroidogenesisrdquo Toxicology vol165 no 2-3 pp 171ndash178 2001
[38] S W J Lamberts E G Bons H A Bruining and F H de JongldquoDifferential effects of the imidazole derivatives etomidateketoconazole and miconazole and of metyrapone on the secre-tion of cortisol and its precursors by human adrenocorticalcellsrdquo Journal of Pharmacology and Experimental Therapeuticsvol 240 no 1 pp 259ndash264 1987
[39] L Yin S Lucas F Maurer U Kazmaier Q Hu and RW Hartmann ldquoNovel imidazol-1-ylmethyl substituted 1256-tetrahydropyrrolo[321-ij]quinolin-4-ones as potent and selec-tive CYP11B1 inhibitors for the treatment of Cushingrsquos syn-dromerdquo Journal ofMedicinal Chemistry vol 55 no 14 pp 6629ndash6633 2012
[40] M DiMattina N Maronian H Ashby D L Loriaux andB D Albertson ldquoKetoconazole inhibits multiple steroidogenicenzymes involved in androgen biosynthesis in the humanovaryrdquo Fertility and Sterility vol 49 no 1 pp 62ndash65 1988
[41] M Ayub and M J Levell ldquoInhibition of human adrenal steroi-dogenic enzymes in vitro by imidazole drugs including keto-conazolerdquo Journal of Steroid Biochemistry vol 32 no 4 pp 515ndash524 1989
[42] D Engelhardt M M Weber T Miksch F Abedinpour andC Jaspers ldquoThe influence of ketoconazole on human adrenalsteroidogenesis incubation studies with tissue slicesrdquo ClinicalEndocrinology vol 35 no 2 pp 163ndash168 1991
[43] S Bhasin S Sikka T Fielder et al ldquoHormonal effects of keto-conazole in vivo in the male rat mechanism of actionrdquo Endo-crinology vol 118 no 3 pp 1229ndash1232 1986
[44] D M Zisterer and D C Williams ldquoCalmidazolium and otherimidazole compounds affect steroidogenesis in Y1 cells lack ofinvolvement of the peripheral-type benzodiazepine receptorrdquoJournal of Steroid Biochemistry and Molecular Biology vol 60no 3-4 pp 189ndash195 1997
[45] L P Walsh C N Kuratko and D M Stocco ldquoEconazoleand miconazole inhibit steroidogenesis and disrupt steroido-genic acute regulatory (StAR) protein expression post-trans-criptionallyrdquo Journal of Steroid Biochemistry and Molecular Bio-logy vol 75 no 4-5 pp 229ndash236 2000
[46] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cor-tisol production and on activities of steroidogenic enzymesin human adrenocortical H295R cellsrdquo The Journal of SteroidBiochemistry and Molecular Biology vol 80 no 3 pp 355ndash3632002
[47] D P Dempsher D S Gann and R D Phair ldquoA mechanisticmodel of ACTH-stimulated cortisol secretionrdquo American Jour-nal of Physiology vol 246 no 4 part 2 pp R587ndashR596 1984
[48] N Sarai S Matsuoka and A Noma ldquosimBio a Java package forthe development of detailed cell modelsrdquo Progress in Biophysicsand Molecular Biology vol 90 no 1ndash3 pp 360ndash377 2006
[49] K Levenberg ldquoA method for the solution of certain non-linearproblems in least squaresrdquo Quarterly of Applied Mathematicsvol 2 pp 164ndash168 1944
[50] D W Marquardt ldquoAn algorithm for least-squares estimation ofnonlinear parametersrdquo SIAM Journal on Applied Mathematicsvol 11 no 2 pp 431ndash441 1963
[51] P E Gill and W Murray ldquoAlgorithms for the solution of thenonlinear least-squares problemrdquo SIAM Journal on NumericalAnalysis vol 15 no 5 pp 977ndash992 1978
[52] Y Akimoto R Hasada J Sakuma I Ono and S KobayashildquoGeneration alternation model for real-coded GA using multi-parent proposal and evaluation of just generation gap (JGG)rdquoin Proceedings of the 19th SICE Symposium on DecentralizedAutonomous Systems pp 341ndash346 Tokyo Japan January 2007
[53] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoProposal and evaluation of adaptive real-coded crossoverAREXrdquo Transactions of the Japanese Society for Artificial Intel-ligence vol 24 no 6 pp 446ndash458 2009
Journal of Toxicology 19
[54] S Kobayashi ldquoThe frontiers of real-coded genetic algorithmsrdquoJournal of Japanese Society for Artificial Intelligence vol 24 no1 pp 147ndash162 2009
[55] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoAnalysis of the behavior of MGG and JGG as a selection modelfor real-coded genetic algorithmsrdquo Transactions of the JapaneseSociety for Artificial Intelligence vol 25 no 2 pp 281ndash289 2010
[56] A Struthers H Krum and G H Williams ldquoA comparisonof the aldosterone-blocking agents eplerenone and spironolac-tonerdquo Clinical Cardiology vol 31 no 4 pp 153ndash158 2008
[57] K Yamasaki M Sawaki S Noda et al ldquoComparison of theHershberger assay and androgen receptor binding assay oftwelve chemicalsrdquo Toxicology vol 195 no 2-3 pp 177ndash1862004
[58] C Eil and S K Edelson ldquoThe use of human skin fibroblasts toobtain potency estimates of drug binding to androgen recept-orsrdquo The Journal of Clinical Endocrinology amp Metabolism vol59 no 1 pp 51ndash55 1984
[59] P Ye T Yamashita D M Pollock H Sasano and W ERainey ldquoContrasting effects of eplerenone and spironolac-tone on adrenal cell steroidogenesisrdquo Hormone and MetabolicResearch vol 41 no 1 pp 35ndash39 2009
[60] B D Albertson K L Frederick N C Maronian et al ldquoTheeffect of ketoconazole on steroidogenesis I Leydig cell enzymeactivity in vitrordquo Research Communications in Chemical Patho-logy and Pharmacology vol 61 no 1 pp 17ndash26 1988
[61] B D Albertson N C Maronian K L Frederick et al ldquoTheeffect of ketoconazole on steroidogenesis II Adrenocorticalenzyme activity in vitrordquoResearch Communications in ChemicalPathology and Pharmacology vol 61 no 1 pp 27ndash34 1988
[62] DM Stocco andB J Clark ldquoRegulation of the acute productionof steroids in steroidogenic cellsrdquo Endocrine Reviews vol 17 no3 pp 221ndash244 1996
[63] L K Christenson and J F Strauss III ldquoSteroidogenic acute regu-latory protein (StAR) and the intramitochondrial translocationof cholesterolrdquo Biochimica et Biophysica Acta (BBA)mdashMolecularand Cell Biology of Lipids vol 1529 no 1ndash3 pp 175ndash187 2000
[64] E R Simpson ldquoCholesterol side-chain cleavage cytochromeP450 and the control of steroidogenesisrdquo Molecular and Cel-lular Endocrinology vol 13 no 3 pp 213ndash227 1979
[65] H A LaVoie and S R King ldquoTranscriptional regulation of ster-oidogenic genes STARD1 CYP11A1 and HSD3Brdquo ExperimentalBiology and Medicine vol 234 no 8 pp 880ndash907 2009
[66] F Raucci A DrsquoAniello and M M Di Fiore ldquoStimulation ofandrogen production by d-aspartate through the enhancementof StAR P450scc and 3120573-HSD mRNA levels in vivo rat testisand in culture of immature rat Leydig cellsrdquo Steroids vol 84pp 103ndash110 2014
[67] M Doi Y Takahashi R Komatsu et al ldquoSalt-sensitive hyper-tension in circadian clock-deficient Cry-null mice involves dys-regulated adrenal Hsd3b6rdquo Nature Medicine vol 16 no 1 pp67ndash74 2010
[68] H Okamura M Doi Y Yamaguchi and J-M Fustin ldquoHyper-tension due to loss of clock novel insight from the molecu-lar analysis of Cry1Cry2-deleted micerdquo Current HypertensionReports vol 13 no 2 pp 103ndash108 2011
[69] M Doi F Satoh T Maekawa et al ldquoIsoform-specific mono-clonal antibodies against 3120573-hydroxysteroid dehydrogenaseisomerase family provide markers for subclassification ofhuman primary aldosteronismrdquo The Journal of Clinical Endo-crinology amp Metabolism vol 99 no 2 pp E257ndashE262 2014
[70] S Konosu-Fukaya Y Nakamura F Satoh et al ldquo3120573-hydroxy-steroid dehydrogenase isoforms in human aldosterone-produc-ing adenomardquo Molecular and Cellular Endocrinology vol 408pp 205ndash212 2015
[71] G C Verwoert J Hofland N Amin et al ldquoExpression and genevariation studies deny association of humanHSD3B1 gene withaldosterone production or blood pressurerdquoAmerican Journal ofHypertension vol 28 no 1 pp 113ndash120 2015
[72] D Hough K Storbeck SW P Cloete A C Swart and P SwartldquoRelative contribution of P450c17 towards the acute cortisolresponse lessons from sheep and goatsrdquoMolecular and CellularEndocrinology vol 408 pp 107ndash113 2015
[73] I Sakuma S Suematsu Y Matsuzawa et al ldquoCharacterizationof steroidogenic enzyme expression in aldosterone-producingadenoma a comparison with various human adrenal tumorsrdquoEndocrine Journal vol 60 no 3 pp 329ndash336 2013
[74] A Tong A Jia S Yan Y Zhang Y Xie and G Liu ldquoEctopiccortisol-producing adrenocortical adenoma in the renal hilumhistopathological features and steroidogenic enzyme profilerdquoInternational Journal of Clinical and Experimental Pathologyvol 7 no 7 pp 4415ndash4421 2014
[75] A Hirsch D Hahn P Kempna et al ldquoMetformin inhibitshuman androgen production by regulating steroidogenicenzymes HSD3B2 and CYP17A1 and complex I activity of therespiratory chainrdquo Endocrinology vol 153 no 9 pp 4354ndash43662012
[76] B Xu L Gao Y Cui et al ldquoSET protein up-regulated test-osterone production in the cultured preantral folliclesrdquo Repro-ductive Biology and Endocrinology vol 11 article 9 2013
[77] S Kikuchi D Tominaga M Arita M Takahashi and MTomita ldquoDynamic modeling of genetic networks using geneticalgorithm and S-systemrdquo Bioinformatics vol 19 no 5 pp 643ndash650 2003
[78] N Shikata YMaki M Nakatsui et al ldquoDetermining importantregulatory relations of amino acids from dynamic networkanalysis of plasma amino acidsrdquo Amino Acids vol 38 no 1 pp179ndash187 2010
[79] M Nakatsui T Ueda Y Maki I Ono and M OkamotoldquoMethod for inferring and extracting reliable genetic interac-tions from time-series profile of gene expressionrdquoMathematicalBiosciences vol 215 no 1 pp 105ndash114 2008
[80] A Komori Y Maki M Nakatsui I Ono and M OkamotoldquoEfficient numerical optimization algorithm based on new real-coded genetic algorithm AREX + JGG and application to theinverse problem in systems biologyrdquo Applied Mathematics vol3 no 10 pp 1463ndash1470 2012
[81] J A Roubos G van Straten and A J B van Boxtel ldquoAn evolu-tionary strategy for fed-batch bioreactor optimization conceptsand performancerdquo Journal of Biotechnology vol 67 no 2-3 pp173ndash187 1999
[82] D Sarkar and J M Modak ldquoGenetic algorithms with filters foroptimal control problems in fed-batch bioreactorsrdquo Bioprocessand Biosystems Engineering vol 26 no 5 pp 295ndash306 2004
[83] K Harada K Ikeda J Sakuma I Ono and S KobayashildquoHybridization of genetic algorithm with local search in mul-tiobjective function optimization recommendation of GA thenLSrdquoTransactions of the Japanese Society for Artificial Intelligencevol 21 no 6 pp 482ndash492 2006
Submit your manuscripts athttpwwwhindawicom
PainResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
ToxinsJournal of
VaccinesJournal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AntibioticsInternational Journal of
ToxicologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
StrokeResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Drug DeliveryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Pharmacological Sciences
Tropical MedicineJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AddictionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Emergency Medicine InternationalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Autoimmune Diseases
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anesthesiology Research and Practice
ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Pharmaceutics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
MEDIATORSINFLAMMATION
of
Journal of Toxicology 5
Table1Ad
renaltoxicities
andactio
nsof
teste
dcompo
unds
Testchem
ical
Testconcentrations
(120583M)
Pathologicalfeatures
ofadrenaltoxicity
Repo
rted
enzymeinh
ibition
sRe
ferences
Acrylonitrile
(AN)
01110and
100
Hem
orrhagicadrenaln
ecrosis
Not
repo
rted
[25]
Salin
omycin
(SM)
000
00100101
1and10
Dam
agetoadrenalm
edullae
Not
repo
rted
[26]
Thioguanine(TG
)001110and100
Hem
orrhagicadrenaln
ecrosis
Not
repo
rted
[25]
Fumaron
itrile
(FN)
01110and
100
Hem
orrhagicadrenaln
ecrosis
Not
repo
rted
[25]
Aminoglutethim
ide(AG
T)01110and
100
Hypertro
phyvascular
degeneratio
nCY
P11A
1CY
P21A
2CY
P11B1andCY
P11B2
[627ndash30]
op1015840-D
DD(D
DD)
0111025and100
Atroph
yCY
P11A
1HSD
3B2CY
P21A
2CY
P11B1andCY
P11B2
[293132]
Spiro
nolacton
e(SP
)11050and100
Hypertro
phy
CYP17H
CYP
17LCY
P11B1andCY
P11B2
[633ndash35]
Metyrapon
e(MP)
01110and
100
Hypertro
phyvascular
degeneratio
nCY
P11A
1CY
P11B1andCY
P11B2
[62936ndash39]
Ketoconazole(KC)
01110and
100
Hypertro
phy
CYP11A
1CY
P17H
CYP
17LHSD
3B2CY
P21A
2and
CYP11B1
[62940
ndash43]
Micon
azole(MC)
011102550and
100
Hypertro
phy
CYP11A
1CY
P17H
CYP
17LCY
P21A
2andCY
P11B1
[4144
45]
Daidzein(D
Z)01110and
100
Unk
nown
HSD
3B2andCY
P21A
2[46]
6 Journal of Toxicology
CHOC
PREG
PROG
DCORTICO
CORTICO
ALDO
CORT
DCORT
HPROG
HPREG DHEA
DIONE
TESTO
CYP11A1
E1
E2
HSD3B2
CYP1
7L
CYP21A2 HSD17B3
CYP1
9A1
CYP11B2
OXYOxysterol synthesis
PREGHPREG DHEAPROG
HPROG DIONE E1DCORTICODCORT TESTO E2CORTICO
CORTALDO
NCI-H295R cell
Medium culture
CYP11B1
CHOL
StAR
CHOM
CHOSCEH
CHOR
CHOLCholesterol transport
Total intracellular cholesterol
CYP1
7H
Vloc Vacc
VMTR
VCEH
Figure 1 Schematic diagram of the mathematical model of adrenal steroidogenesis in NCI-H295R cells Overview of the mathematicalmodel of adrenal steroidogenesis in NCI-H295R cells including cholesterol transport and intracellular localization oxysterol synthesisthe C
21-steroid hormone biosynthesis pathway passive diffusional transport of steroid hormones and cell proliferation CHOL total
cholesterol in medium culture CHOS stored cholesterol esters in the endoplasmic reticulum CHOC intracellular free cholesterolCHOM mitochondrial free cholesterol CHON mitochondrial free cholesterol close to CYP11A1 enzymes CHOR mitochondrial freecholesterol remote from CYP11A1 enzymes PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosteronePROG progesterone HPROG 17120572-hydroxyprogesterone DIONE androstenedione DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICO corticosterone CORT cortisol ALDO aldosterone TESTO testosterone E1 estrone E2 17120573-estradiol OXYoxysterol CEH cholesterol ester hydrolase StAR steroidogenic acute regulatory protein CYP11A1 P450 side chain cleavage enzymeCYP17H 17120572-hydroxylase of CYP17 CYP17L C
17ndash20 lyase of CYP17 HSD3B2 3120573-hydroxysteroid dehydrogenase CYP21A2 21-hydroxylaseCYP11B1 11120573-hydroxylase CYP11B2 18-hydroxylase HSD17B3 17120573-hydroxysteroid dehydrogenase and CYP19A1 aromatase
119894 in compartment ℎ at time point 119895 119883simℎ119894119895
is the simulatedconcentration of molecule 119894 in compartment ℎ at time point119895 and 119883max
ℎ119894is the maximum concentration of molecule 119894 in
compartment ℎ over all time points Data points under thelower quantitation limit were excluded from the evaluationby the objective function
Effects of every static model parameter for parameteroptimization were calculated from differences of fitting theobjective function using sensitivity analysis
224 Quantitative Estimation of the Mechanism of Action ofAdrenal Toxicants Metabolic steroid profiling and differen-tial patterns of the adrenal steroid hormones by chemicalperturbation were reconstructed to optimize the relativeactivities of the steroidogenic enzymes The input datafor the quantitative mechanistic analysis of adrenal toxiccompounds was a fold change (ratio) of the measured 12
steroid concentrations induced by drug exposure for 72 hThe two-step optimization method of the real-coded geneticalgorithm (RCGA) was adopted as a global optimizationmethod in the quantitative mechanistic analysis of adrenaltoxic compoundsThe operations of the crossover and gener-ation alterationmodel in RCGAwere used for the real-codedensemble crossover (REX) and just generation gap (JGG)[52ndash55] As the initial parameters of RCGA maximum gen-eration population size selection size of parent individualspopulation size of child individuals and termination criteriawere 1000 100 6 25 and under 01 of NLSD respectivelyThe search space for the relative activities of the steroidogenicenzymes was from 1100 to 100 To evaluate the fitness ofeach individual the sum of squared residuals for fold changesof measured 12 steroid concentrations was used as theobjective function Nonlinear least squares optimization bythe Levenberg-Marquardt method was used as a local
Journal of Toxicology 7
search [49ndash51] As the estimated mechanisms of actionsof the adrenal toxic compounds the relative activities ofeight steroidogenic enzymes (CYP11A1 CYP17H CYP17LHSD3B2 CYP21A2 CYP11B1 CYP11B2 and HSD17B3)were optimized by the above-mentioned 2-step optimizationmethod Every optimization calculation was duplicated tocheck the numerical stability of the optimal parameters
225 Global Dynamic Sensitivity Analysis The propertyof every kinetic parameter in this computational modelof steroidogenesis in NCI-H295R cells was evaluated bydynamic sensitivity analysis The sensitivity (119878
119909119910) of kinetic
parameter 119909 for variable parameter 119910 was defined by the fol-lowing equation
119878119909119910(119905)=
Δ119910(119905)119910(119905)
Δ119909119909 (2)
where variable parameter119910was the concentration of a steroidhormone in the cytosolic space of NCI-H295R cells Theperturbation for kinetic parameters was +10 (Δ119909119909 = 01)
3 Results
31 Experimental Data on Adrenal Steroidogenesis
311 Adrenal Steroidogenesis of NCI-H295R Cells and theMass Balance All steroid hormones in the culture mediumwere significantly increased after 72 h of stimulation with50 nM ACTH 20120583M forskolin and 100 nM angiotensinII (Figure 2(a)) Mass balances in steroidogenesis of NCI-H295R cells under nontreatment and control (stimulated)conditions are shown in Figures 2(b) and 2(c) respectivelyUnder stimulation the dynamics of net mass in these exper-iments were unchanged and accumulated cholesterol wasconverted to adrenal steroids
312 Cytotoxicity of Adrenal Toxicants Viabilities of cellstreated with each compound were expressed as a relativevalue to the ATP level of the control Effects of AN SM TGFN AGT DDD SP MP KC MC and DZ on cell viabilitywere determined to be valid under 80 of the relative ATPlevel at 7 days after treatment AN SM TG and FN showedcytotoxicity at over 100 1 10 and 10120583M respectively AGTMP andDZ did not affect cell viability at up to 100 120583MDDDSP KC andMC induced less than 80 of cell viability at over100 50 100 and 25 120583M respectively
313 Differentially Steroid Profiling of Adrenal ToxicantsAfter NCI-H295R cells were exposed to each test compoundduring three days the concentrations of 12 steroid hor-mones in the culture mediumwere simultaneously measuredby LCMSMS All effects of the compounds on adrenalsteroidogenesis were evaluated at the concentration withoutany overt cytotoxicity The differential metabolic steroidprofiles of 11 adrenal toxic compounds were classified andvisualized by using hierarchical clustering analysis (Figure 3)
Four adrenal toxicants without steroidogenic inhibitionAN SM TG and FN did not change the medium concen-trations of all steroid hormones by more than 2-fold Above-mentioned 4 compounds at every condition and 7 adrenalsteroidogenic inhibitors at the low exposure concentrationwere gathered into a big cluster as nonchange group The7 steroidogenic inhibitors at the maximum exposure con-centration showed the characteristic steroid profiles eachbut 100 120583M DZ and 10 120583M SP were classified as a clusterAGT drastically decreased the medium concentrations ofPREG HPREG DHEA PROG DCORTICO CORTICOand ALDO at 100 120583M DDD dose-dependently decreased themedium concentrations of PROG DCORTICO CORTICOCORT and ALDO at gt10 120583M and decreased PREG HPREGDHEA PROG HPROG DIONE and DCORT at the maxi-mum exposure concentration of 25120583M SP increased PREGHPREG and DHEA and decreased PROG DIONE DCOR-TICODCORTCORTICOALDO andTESTOat 10120583MMPdose-dependently decreased CORTICO CORT and ALDOand decreased DHEA HPROG DIONE and TESTO at themaximum exposure concentration of 100 120583M KC drasticallydecreased the medium concentrations of PREG HPREGDHEAHPROGDIONEDCORTICODCORTCORTICOCORTO ALDO and TESTO at 10 120583M MC increased themedium concentrations of PROG and decreased DIONEDCORT CORT and TESTO at 10 120583M DZ increased PREGHPREG and DHEA and decreased DIONE DCORTICODCORT CORTICO CORT ALDO and TESTO at 100 120583M
32 The Mathematical Modeling
321 Optimization of the Mathematical Model of AdrenalSteroidogenesis in NCI-H295R Cells The mathematicalmodel of adrenal steroidogenesis in NCI-H295R cells wasoptimized for several kinetic parameters of cholesteroltransport intracellular localization the oxysterol pathwayand maximum velocity of steroidogenic enzymes to fit theexperimental time-course data All optimized kinetic para-meters are shown in Table S1 in a supplementary documentThe optimized mathematical model was reconstructedwith the experimental dynamic patterns of cholesteroland the 14 steroid hormones in the intracellular space andculture medium The fitness was 0621761 of NLSD valuesas the fitting objective function The simulation results andexperimental data are shown in Figure 4
Optimized kinetic parameters were calculated sensitivi-ties for the NLSD value as the fitting score and are shown inTable S1 in a supplementary document The highly sensitiveparameters for fitting the NLSD score were the extractednine kinetic parameters 119896Cholesterol Transport 119896
acc119891 119896loc119891 119896loc119887
119881CYP11A1max 119870CYP11A1
119898 119881CYP17HmaxA 119881HSD3B2
maxA and 119881CYP21A2maxA whichhad higher than 30 fitting sensitivity
33 The Validation Using the Adrenal Toxicants
331 Mechanistic Analysis of Adrenal Toxicants Effects ofadrenal toxic compounds on steroidogenic enzymes werequantitatively predicted from the change in the ratio ofthe measured medium concentrations of the 12 steroid
8 Journal of Toxicology
0001
001
01
1
10
PREG
HPR
EG
DH
EA
PRO
G
HPR
OG
AN
DRO
DCO
RTIC
O
DCO
RT
CORT
ICO
CORT
ALD
O
TEST
O E1 E2C
once
ntra
tion
in m
ediu
m (120583
M)
lowast
lowast
lowast
lowast
lowast
lowast
lowast lowast lowastlowast
lowast
lowast
lowast
lowast
(a)
0 8 24 48 72Time (hours)
0
40
80
120
160
200
240
Num
ber o
f mol
ecul
es (n
mol
)
(b)
0 8 24 48 72Time (hours)
0
40
80
120
160
200
240
Num
ber o
f mol
ecul
es (n
mol
)
(c)
Figure 2 Experimental data of metabolic profiling of adrenal steroid hormones and the mass balance Concentrations of steroid hormonessecreted from NCI-H295R cells in the culture medium at 72 h after stimulation were compared with the untreated condition and thestimulated condition by 50 nM ACTH 20 120583M forskolin and 100 nM angiotensin II (a) Net molecular amounts including cholesterol andsteroid hormones in the culture medium and intracellular space are plotted at five time points (0 8 24 48 and 72 h after stimulation)under the untreated condition (b) and the stimulated condition (c) In the bar graphs dark and light bars indicate the amount of cholesteroland adrenal steroids respectively All data are shown as the mean plusmn SD (119873 = 4) lowast119901 values corrected by the familywise error ratelt001 PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-hydroxyprogesteroneDIONE androstenedioneDCORTICO 11-deoxycorticosteroneDCORT 11-deoxycortisol CORTICO corticosteroneCORT cortisol ALDO aldosterone and TESTO testosterone
hormones at 72 h after drug exposure using themathematicalmodel of adrenal steroidogenesis in NCI-H295R cells Thereproducibility of the estimated results was confirmed byperforming the test twice The estimated effects of 11 adrenaltoxic compounds on eight steroidogenic enzymes are shownin Figure 5 The adrenal toxic compounds without steroido-genic inhibition such as vasculotoxic agents (AN SM TGand FN) were not estimated for the target steroidogenicenzymes under noncytotoxic conditions Every fitness valueswere under 005 of NLSD values used as the fitting objectivefunction (Figures 5(a)ndash5(d)) Other steroidogenic inhibitors(AGT DDD SPMP KCMC andDZ) are described in detailbelow
332 AGT The mechanism of action of AGT in adrenalsteroidogenesis was estimated by inhibition of CYP11A1HSD3B2 CYP21A2 and CYP11B1 at 100 120583M (estimatedinhibitions were 770 780 811 and 598 resp)(Figure 5(e)) AGT has been reported to inhibit CYP11A1
CYP21A2 CYP11B1 and CYP11B2 [6 27ndash30] Our resultswere mostly consistent with the previous reports In particu-lar CYP11A1 appeared to be inhibited strongly byAGT In ourstudy HSD3B2 inhibition of AGT was shown by mechanisticanalysis based on systems biology approaches as a novelmechanism of action of AGT Inhibition of AGT by CYP11B2was not estimated in our study However the concentrationof ALDO in the culture medium decreased to 38 ofthe normal stimulated condition Inhibition of AGT byCYP11B2 has been shown using sheep adrenal homogenatesas well as a human adrenal homogenate from a patientwith Cushingrsquos syndrome [30]The activity of 18-hydroxylaseinduced by CYP11B2 was determined as the conversion ofcorticosterone to 18-hydroxycorticosterone in the previousstudy The cause of the discrepancy regarding the effect ofAGT on CYP11B2 was suggested to be substrate inhibitionbecause the intracellular concentration of CORTICO wasincreased by over 10 times of that in the culture mediumto reach 50 120583M Another possibility was poor quantitative
Journal of Toxicology 9
log2
ratio
4
3
2
1
0
minus1
minus2
minus3
minus4
ALDO
CORTICO
CORT
TESTO
DIONE
HPROG
DCORTICO
DCORT
PROG
PREG
HPREG
DHEA
DZ(
100)
SP(1
0)SP
(1)
DZ(
10)
DD
D(0
1)
AN
(01
)A
N(1
)A
N(1
0)SM
(00
1)TG
(00
1)M
P(0
1)FN
(01
)TG
(1)
AGT(
1)AG
T(0
1)SM
(00
0001
)D
DD
(1)
SM(0
1)
AGT(
10)
MP(
1)FN
(1)
MC(
1)D
Z(1)
KC(0
1)
MC(
01)
DZ(
01)
KC( 1
)D
DD
(10)
MP(
10)
MC(
10)
MP(
100)
AGT(
100)
DD
D(2
5)KC
(10)
Figure 3 Hierarchical cluster analysis of differential metabolic profiles of 12 steroid hormones by exposure to adrenal toxicity compoundsAdrenal toxicants were classified by using the differential metabolic profiling of 12 steroid hormones Concentrations of 12 adrenalsteroids secreted from NCI-H295R cells were drastically changed by the exposure of adrenal steroidogenesis inhibitors The 12 adrenalsteroid hormones were quantitatively measured by LC-MSMS simultaneously Four adrenal vasculotoxic compounds acrylonitrile ANfumaronitrile FN salinomycin SM and thioguanine TG were used as negative control compounds for adrenal steroidogenesis inhibitorsSeven adrenal steroidogenesis inhibitors aminoglutethimide AGT op1015840-DDD DDD spironolactone SP metyrapone MP ketoconazoleKC miconazole MC and daidzein DZ showed a characteristic steroid profile each and were classified as each independent singletonat the maximum exposure condition Exposure concentrations of adrenal toxic compounds were described in brackets of the samplename and the units were prepared as 120583M A blue cluster was classified as a group of nonchange samples including negative controlcompounds and low exposure conditions of adrenal steroidogenesis inhibitors PREG pregnenolone HPREG 17120572-hydroxypregnenoloneDHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-hydroxyprogesterone DIONE androstenedione DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICO corticosterone CORT cortisol ALDO aldosterone and TESTO testosterone
reliability of the experimental data because the ALDOconcentration was under the lower limit of quantification at100 120583MAGT Hecker and colleagues reported that 3 120583MAGTdecreases PREG and PROG concentrations and increases theTESTO concentration [10] However AGT did not increasethe TESTO concentration in our study One possibility isthat the concentration of TESTO was already enhanced byabout 33-fold through stimulation with ACTH forskolinand angiotensin II
333 DDD The mechanism of action of DDD in adrenalsteroidogenesis was estimated by dose-dependent inhibitionof CYP11A1 HSD3B2 CYP21A2 and CYP11B1 (estimatedinhibitions at 25120583M were 870 869 769 and 849resp) (Figure 5(f)) DDD has been reported to inhibitCYP11A1 HSD3B2 CYP21A2 CYP11B1 and CYP11B2 [29 3132] Inhibition of DDD by CYP11B2 was not estimated in ourstudy However the concentration of ALDO in the culturemedium decreased to 3 of that in the normal stimulatedcondition Inhibition of DDD by CYP11B2 has been shownusing mitochondrial and microsomal fractions prepared bystandard centrifugation procedures from a bovine adrenal
cortex homogenate [32]The cause of the discrepancy regard-ing the inhibition ofDDDbyCYP11B2 could not be explainedby same effect in the case of AGT
334 SP Themechanism of action of SP in adrenal steroido-genesis was estimated by inhibition of HSD3B2 CYP21A2and HSD17B3 (estimated inhibitions at 10120583M were 702595 and 593 resp) (Figure 5(g)) SP has been reportedto inhibit CYP17H CYP17L CYP11B1 and CYP11B2 [6 33ndash35] The inhibitory effect of SP on the HSD3B2 enzyme wasa novel mechanism of action The main action of SP is asa mineralocorticoid receptor (MR) antagonist SP has alsobeen reported to exert some off-target effects by binding toandrogen glucocorticoid and progesterone receptors [56ndash58] SP has been shown to inhibit the production of ALDOand CORT from PREG induced by angiotensin II in H295Rcells but the specific MR antagonist eplerenone did not showthe inhibitory effects [59] Therefore HSD3B2 inhibition bySP is not mediated via MR and the action might be directinhibition of HSD3B2 enzymes or a part of known off-targeteffects mediated through other nuclear hormone receptorsRegarding CYP17H and CYP17L our results were consistent
10 Journal of Toxicology
CHOL
0
20
40
60
80
100C
once
ntra
tion
in m
ediu
m (120583
M)
10 20 30 40 50 60 70 800Time (hours)
(a)
PREG
DCORT
DCORTICO
CORTICO
CORT
Con
cent
ratio
n in
med
ium
(120583M
)
10
4
2
0
8
6
10 20 30 40 50 60 70 800Time (hours)
(b)
Con
cent
ratio
n in
med
ium
(120583M
)
0
020406
081
1214
0 10 20 30 40 50 60 70 80Time (hours)
HPREG
DIONE
PROGHPROG
E1
(c)
DHEAE2
ALDO
TESTO
0
005
01
015
02
025
Con
cent
ratio
n in
med
ium
(120583M
)
0 10 20 30Time (hours)
40 50 60 70 80
(d)
CHOL
0
5000
10000
15000
20000
25000
30000
Con
cent
ratio
n in
cell
(120583M
)
10 20 30 40 50 60 70 800Time (hours)
(e)
Con
cent
ratio
n in
cell
(120583M
)
0
50
100
150
200
250
0 10 20 30 40 50 60 70 80Time (hours)
PREG
DCORT
DCORTICO
CORTICOCORT
(f)
0 10 20 30 40 50 60 70 80Time (hours)
HPREG
DIONE
PROGHPROG
E1
Con
cent
ratio
n in
cell
(120583M
)
3025201510
35
05
(g)
005
115
2
253
35
Con
cent
ratio
n in
cell
(120583M
)
0 10 20 30 40 50 60 70 80Time (hours)
DHEA
E2ALDO
TESTO
(h)
Figure 4 Comparison of time-course profiles of cholesterol and adrenal steroids produced by NCI-H295R cells between experimentallymeasured and simulated data To intuitively confirm reconstruction of the measured experimental data in the developed simulation modelof NCI-H295R cells dynamics of cholesterol and adrenal steroids produced by NCI-H295R cells were plotted to overlay experimentaldata with the simulated results Graphs show the dynamics of medium concentrations of cholesterol (a) and adrenal steroids ((b)ndash(d))and intracellular concentrations of cholesterol (e) and adrenal steroids ((f)ndash(h)) Steroid hormones were categorized into three groups byconcentration levels Major steroids were PREG pregnenolone DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICOcorticosterone and CORT cortisol ((b) and (f)) Moderate steroids were HPREG 17120572-hydroxypregnenolone PROG progesterone HPROG17120572-hydroxyprogesterone DIONE androstenedione and E1 estrone ((c) and (g)) Minor steroids were DHEA dehydroepiandrosteroneALDO aldosterone TESTO testosterone and E2 17120573-estradiol ((d) and (h)) Experimental data are shown as symbols with dotted lines Alldata represent the mean plusmn SD (119873 = 4) Simulation data are shown as solid lines
Journal of Toxicology 11
0110
10
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
AN (120583M)
(a)
0110
FN (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(b)
00101
SM (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
100E minus 05
(c)
0110
TG (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B1
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(d)
0110
10100
AGT (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(e)
0110
1025
DDD (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(f)
1010
SP (120583M)
2750005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(g)
0110
10100
MP (120583M)
2110005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(h)
Figure 5 Continued
12 Journal of Toxicology
0110
10
KC (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(i)
0110
10
MC (120583M)
4430005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(j)
0110
10100
DZ (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(k)
Figure 5 Estimated mechanism of action of adrenal toxicants by using the mathematical model of adrenal steroidogenesis in NCI-H295Rcells Mechanisms of action of adrenal toxicity compounds were quantitatively estimated from experimental results of differential steroidprofiling using the mathematical model of adrenal steroidogenesis in NCI-H295R cells The drug action was defined as a scaling factorof enzymatic activity in the simulation model These scaling factors were optimized to fit experimental data by a hybrid optimizationmethod of the RCGA and nonlinear least squares Estimated drug actions by the exposure of vasculotoxic agents acrylonitrile AN (a)fumaronitrile FN (b) salinomycin SM (c) and thioguanine TG (d) and the steroidogenic inhibitors aminoglutethimide AGT (e) op1015840-DDD DDD (f) spironolactone SP (g) metyrapone MP (h) ketoconazole KC (i) miconazole MC (j) and daidzein DZ (k) are shownas a spider radar chart CYP11A1 P450 side chain cleavage enzyme CYP17H 17120572-hydroxylase of CYP17 CYP17L C
17ndash20 lyase of CYP17HSD3B2 3120573-hydroxysteroid dehydrogenase CYP21A2 21-hydroxylase CYP11B1 11120573-hydroxylase CYP11B2 18-hydroxylase and HSD17B317120573-hydroxysteroid dehydrogenase
with previous reports [33 34] 7120572-thiospironolactone whichis synthesized by deacetylation of SP inhibits CYP17H andCYP17L [34] The fact that there were no inhibitions ofCYP17HorCYP17L in our study suggests that SPmight not bedeacetylated to 7120572-thiospironolactone in NCI-H295R cellsRegarding CYP11B1 and CYP11B2 our results were unclearcompared with a previous study It has been shown that30 120583M SP inhibits CYP11B1 and CYP11B2 in human andbovine adrenal mitochondria [35]The cause of CYP11B1 andCYP11B2 inhibition by SP could not be determined in ourstudy which might be due to the lower maximum exposureconcentration of SP than that in the previous reportWe couldnot examine SP concentrations over 10 120583M because theseconcentrations were cytotoxic in NCI-H295R cells
335 MP The mechanism of action of MP in adrenalsteroidogenesis was estimated by dose-dependent inhibitionof CYP11B1 (estimated inhibitions at 1 10 and 100 120583M were571 827 and 982 resp) (Figure 5(h)) MP has beenreported to inhibit CYP11B1 as its major effect and CYP11A1and CYP11B2 as a weak effect [6 29 36ndash39] The results wereable to show that MP is a selective inhibitor of CYP11B1 inthe previous report However the estimated effect of MPat a high concentration 100 120583M as the maximum exposureconcentration was unclear According to the previous reportselectivity of MP for CYP11B1CYP11B2 is about five times[39] In addition 20 120583M MP has a slight inhibition effect onCYP11A1 in H295R cells [29]
336 KC The mechanism of action of KC in adrenalsteroidogenesis was estimated by inhibition of CYP11A1CYP17H CYP17L HSD3B2 CYP21A2 CYP11B1 and
CYP11B2 (estimated inhibitions at 10 120583Mwere 926 943518 830 882 974 and 798 resp) (Figure 5(i))KC has been reported to inhibit CYP11A1 CYP17H CYP17LHSD3B2 CYP21A2 and CYP11B1 [6 29 40ndash43] Our resultswere almost consistent with the previous reports KC inhibitsCYP11A1 CYP17H CYP21A2 and CYP11B1 in NCI-H295Rcells at 10 120583M [29] and CYP17H CYP17L CYP21A2 andCYP11B1 in human adrenal mitochondria and Leydig cellmicrosomes at 2ndash5120583M [60 61] However KC has shownonly weak inhibition of HSD3B2 and HSD17B3 in Leydigcells at the millimolar level [60 61] Regarding CYP11B2 andHSD17B3 we considered that these estimated inhibitions ofKC did have sufficient reliability in terms of quantitative pre-diction precision because ALDO and TESTO concentrationswere less than the lower limit of quantification at 10 120583MKC
337 MC Themechanism of action of MC in adrenal stero-idogenesis was estimated by inhibition of CYP17H CYP17LCYP11B1 and HSD17B3 (estimated inhibitions at 10120583Mwere691 530 764 and 571 resp) (Figure 5(j)) MC hasbeen reported to inhibit not only CYP17H and CYP17L butalsoCYP11A1 CYP21A2 andCYP11B1 [41 44 45]The resultsin the previous reports were able to estimate that MC isa CYP17 inhibitor However CYP11A1 inhibition by MCprobably instead of a reduction in StAR expression wasnot clearly detected in our study using NCI-H295R cellsbecause there were no decreases in the concentrations ofPREG and PROG in the culture medium Indirect inhibitionof CYP11A1 via the peripheral-type benzodiazepine receptorhas been reported in mouse adrenocortical Y-1 cells treatedwith MC in the absence of stimuli by measuring PREGproduction [44] On the other hand reductions of StAR
Journal of Toxicology 13
protein expression andor transport activity without affectingtotal steroid synthesis or CYP11A1 and HSD3B2 enzymeexpression or activities have been reported in (BU)
2cAMP-
stimulated MA-10 Leydig tumor cells treated with MC bymeasuring PROG production [45] Therefore the effect ofMC on the initial reaction in adrenal steroidogenesis fromcholesterol should be different according to the cell type andstimulation condition Inhibition of CYP21A2 and CYP11B1by MC has been reported as decreases in the consumptionof PROG and DCORTICO respectively [41] Inhibition ofCYP11B1 was estimated by the action of MC in this studybut that of CYP21A2 was not detected In the previousexperimental report inhibitory sites by MCmight have beenreflected by inhibition of CYP17H activity because CYP21A2activity was measured as a decrease in labeled PROG
338 DZ The mechanism of action of DZ in adrenalsteroidogenesis was estimated by inhibition of CYP11A1HSD3B2 CYP21A2 CYP11B1 and HSD17B3 (estimated inhi-bitions at 100 120583M were 586 941 965 872 and981 resp) (Figure 5(k)) DZ has been reported to inhibitHSD3B2 and CYP21A2 [46] The results of HSD3B2 andCYP21A2 were consistent with the previous report Howeverinhibitions have not been reported for CYP11A1 CYP11B1and HSD17B3These estimated effects of DZ on CYP11B1 andHSD17B3 were unclear because the concentrations of ALDOand TESTO were less than the lower limit of quantificationat 100 120583M In addition these enzymes act downstream of thestrong action points of DZ such as HSD3B2 and CYP21A2
34 The Simulations and the Systematic Model Analysis
341 Dynamic Sensitivity Analysis of Adrenal SteroidogenesisTo comprehensively understand the dynamics of adrenalsteroidogenesis dynamic sensitivities were calculated forsteroid concentrations secreted by NCI-H295R cells usingour constructed mathematical model of steroidogenesis Theresults of dynamic sensitivity analysis at 72 h of duration and6 h of interval time are presented as a heat-map in Figure 6
The top 10 parameters of the total area under the curveof dynamic sensitivity for cholesterol and the 14 steroids inculture medium were 119881HSD3B2
maxA 119881CYP11A1max 119881CYP21A2maxA 119896loc119891 119896loc119887
119870CYP11A1119898
119881CYP17HmaxA 119896Cholesterol Transport 119896acc119891 and 119881CYP21A2maxB in
order from the top Cholesterol uptake (119896Cholesterol Transport)StAR protein (119896loc
119891) and CYP11A1 (119881CYP11A1max ) which are
determining factors of the capacity for steroidogenesis pro-moted the production of mineralocorticoids (DCORTICOCORTICO and ALDO) and restrained the synthesis of glu-cocorticoids (DCORT and CORT) and sex steroids (DIONETESTO and E1) because of the accumulation of interme-diate molecules in steroidogenesis (PREG HPREG DHEAPROG and HPROG) only by self-activation The dynamicpatterns of the intermediate molecules in steroidogenesiswere mainly dependent on the activity of CYP17H andHSD3B2 with PREG as the substrate of these enzymes inwhich the dynamic sensitivities of 119881CYP17HmaxA for HPREG andHPROG and 119881HSD3B2
maxA for PROG HPROG and DCORTICO
reversed the direction of sensitivity at 49ndash66 h after stimu-lation The dynamic sensitivities of the maximum activitiesof HSD3B2 for PREG (119881HSD3B2
maxA ) and CYP21A2 for PROG(119881CYP21A2maxA ) were related to all steroids at 72 h Almost allmodel parameters had positive sensitivity for downstreamsteroids in the adrenal steroidogenic pathway and negativesensitivity for direct-binding steroids as substrates of thesteroidogenic enzyme The sensitivity of 119881max in all steroido-genic enzymes was relatively higher than 119870
119898for the same
steroid substrate
342 Simulation of the Metabolic Balance of AdrenalSteroidogenesis Pathway To clearly show the property of themetabolic shift between mineralocorticoid and glucocorti-coid biosynthesis we performed two-dimensional parameterscanning of the enzymatic activities of CYP17H and HSD3B2(Figure 7) NCI-H295R cells lost the ability to produce allsteroid hormones when enzymatic activities of CYP17H andHSD3B2 were changed by over 60 and 30 respectivelyActivation ofCYP17HandorHSD3B2 induced themetabolicshift that enhanced the glucocorticoid biosynthesis and devi-ated from the mineralocorticoid biosynthesis On the otherhand inhibition of CYP17H andor HSD3B2 induced themetabolic shift that enhanced the mineralocorticoid biosyn-thesis and deviated from the glucocorticoid biosynthesisMoreover the enzymatic activity of HSD3B2 regulated themetabolic balance of sex steroids and the precursors onadrenal steroidogenesis of NCI-H295R cells E1 TESTO andDIONE were produced by NCI-H295R cells when activatingthe enzymatic activity of HSD3B2 Conversely E2 andDHEAwere produced by NCI-H295R cells when suppressing theenzymatic activity of HSD3B2 The biosynthesis of down-stream steroids in adrenal steroidogenesis pathway such asmineralocorticoids and glucocorticoids was almost com-pletely terminated when the enzymatic activity of HSD3B2was decreased by over 80
4 Discussion
41 Importance of 3120573-HSDActivity in Adrenal SteroidogenesisOur systematic analysis using the mathematical model ofadrenal steroidogenesis in NCI-H295R cells revealed thatthe enzymatic activity of 3120573-HSD controls the dynamics ofadrenal steroidogenesis The activity of the StAR proteincontrols the net capacity of steroidogenesis in steroidogeniccells which is the transport of cholesterol from the outer toinner mitochondrial membranes Both the expression levelsof StAR protein and mRNA are rapidly elevated in responseto stimulation by tropic hormones such as ACTH [62 63]Another important factor in adrenal steroidogenesisis is thecholesterol side chain cleavage enzyme CYP11A1 the firstrate-limiting and hormonally regulated step in the synthesisof all steroid hormones which is conversion of cholesterolto pregnenolone in mitochondria [64] According to ourresults of global sensitivity analysis (Supplementary Table 1and Figure 6(d)) in addition to CYP11A1 and StAR proteins3120573-HSD was one of the key regulators in adrenal steroido-genesis of NCI-H295R cells And also this result suggeststhat a significant regulatory mechanism in steroidogenesis
14 Journal of Toxicology
60544830 4236 66 7218 246 120
Time (hours)
0
1
2
3C
once
ntra
tion
(a)
60544830 4236 66 7218 246 120
Time (hours)
minus05
0
05
minus1
1
Sens
itivi
ty
(b)
0
6
12
18
24
30
36
42
48
54
60
66
72
Sens
itivi
ty fo
r med
ium
conc
entr
atio
ns
of ch
oles
tero
l or s
tero
id h
orm
ones
Dur
atio
n tim
e (ho
urs)
Model parameters
+10
+05
00
minus05
minus10
(c)
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
Parameter number in the mathematical model of steroidogenesis in NCI-H295R cells
CHOL
PREG
HPREG
DHEA
PROG
HPROG
DIONE
DCORTICO
DCORT
CORTICO
CORT
ALDO
TESTO
E1
E2
Chol
este
rol o
r ste
roid
hor
mon
es
0
05
10
15
20
minus05
minus10
minus15
minus20
(d)
Figure 6 Heat-map of the global dynamic sensitivity analysis of adrenal steroid concentrations produced by NCI-H295R cells The globaldynamic sensitivity analysis is a powerful tool to comprehensively understand the dependencies of themodel parameters in themathematicalmodel of the biological complex system Dynamic sensitivities of model parameters in the mathematical model of adrenal steroidogenesisin NCI-H295R cells were calculated for all steroid concentrations in the culture medium every 6 h until 72 h after stimulation To clarify theview of heat-map of global dynamic sensitivity analysis imaginary data of the dynamics of steroid concentrations in the original model (blueline) and perturbed model for sensitivity analysis (red line) were prepared (a) Using this imaginary data the calculated dynamic sensitivities(b) and the visualized dynamic sensitivity as one block of the heat-map (c) were shown respectively By the same method that explainedusing imaginary data the large-scale data of the global dynamic sensitivity analysis on the mathematical model of adrenal steroidogenesisin NCI-H295R cells was comprehensively visualized as a big graph of heat-map (d) Parameter numbers in the horizontal axis are (1)119896Choresterol Transport (2) 119896CEH (3) 119896MTR
119891 (4) 119896MTR
119887 (5) 119896acc
119891 (6) 119896acc
119887 (7) 119896loc
119891 (8) 119896loc
119887 (9) 119896Oxysterol Synthesis (10)119870
CYP11A1119898
(11) 119881CYP11A1max (12)119870CYP17H119898A
(13) 119870CYP17H119898B (14) 119881CYP17HmaxA (15) 119881CYP17HmaxB (16) 119881CYP17HmaxB (17) 119870CYP17L
119898B (18) 119881CYP17LmaxA (19) 119881CYP17LmaxB (20) 119870HSD3B2119898A (21) 119870HSD3B2
119898B (22) 119870HSD3B2119898C (23)
119881HSD3B2maxA (24) 119881HSD3B2
maxB (25) 119881HSD3B2maxB (26) 119870CYP21A2
119898A (27) 119870CYP21A2119898B (28) 119881CYP21A2maxA (29) 119881CYP21A2maxB (30) 119870CYP11B1
119898A (31) 119870CYP11B1119898B (32) 119881CYP11B1maxA
(33) 119881CYP11B1maxB (34) 119896CYP11B2 (35) 119870HSD17B3119898A (36) 119870HSD17B3
119898B (37) 119881HSD17B3maxA (38) 119881HSD17B3
maxB (39) 119870CYP19A1119898A (40) 119870CYP19A1
119898B (41) 119881CYP19A1maxA and (42)119881
CYP19A1maxB PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-
hydroxyprogesteroneDIONE androstenedioneDCORTICO 11-deoxycorticosteroneDCORT 11-deoxycortisol CORTICO corticosteroneCORT cortisol ALDO aldosterone and TESTO testosterone
pathway is very reasonable StAR CYP11A1 and 3120573-HSD(isoforms 1 or 2 in humans) proteins generally respond to thesame hormones that stimulate steroid production throughcommon pathways such as cAMP signaling in adrenal glandsand testes [65 66] Moreover our data also support recentexperimental evidence from clinical and in vivo studiessuggesting that the enzymatic activity of 3120573-HSD plays
an important role in the regulation of mineralocorticoidsynthesis in adrenal steroidogenesis and contributes to hyper-tension caused by abnormal overproduction of aldosterone[67ndash70] Circadian clock-deficient Cry-null mice show salt-sensitive hypertension due to abnormally high synthesis ofaldosterone which is caused by constitutively high expressionof HSD3B6mRNA and protein in the adrenal cortex [67 68]
Journal of Toxicology 15
All steroid producing cells
Mineralocorticoid producing cells Glucocorticoid producing cells
Mineralocorticoid and glucocorticoid producing cells
Terminated steroidogenesis
Sex steroids and precursor sex steroidsDIONE TESTO E1DEHA E2
NCI-H295Rcells at controlcondition
Normalized activities of HSD3B2
Nor
mal
ized
activ
ities
of C
YP17
H
2
18
16
14
12
1
08
06
04
02
00 05 1 15 2
Figure 7 Metabolic categories of the steroidogenic cells determined by the balance of HSD3B2 and CYP17H activitiesThe two-dimensionalparameter scanning analysis by the perturbation of two focused parameters clarifies the interaction and the relationship between the twoparameters in the complex system Functional cellular categories of steroidogenic cells were defined by the levels of mineralocorticoid(ALDO) glucocorticoids (DCORTandCORT) and androgens (DHEAandDIONE) at 72 h after stimulation Enzymatic activities ofHSD3B2andCYP17were normalized by standard values of the simulationmodel inNCI-H295R cells Scanning ranges ofHSD3B2 andCYP17 activitieswere 0ndash200 each 10 Green regions are all 14 steroids produced by NCI-H295R cells Red blue and purple regions are mineralocorticoidproducing cells glucocorticoids-producing cells and both corticoid-producing cells respectively Yellow regions are steroidogenesis of NCI-H295R cells terminated upstream of the adrenal steroidogenesis pathway
Recent clinical observations have revealed predominantexpression of HSD3B2 mRNA and protein in tumor cellsof aldosterone-producing adenoma (APA) and HSD3B1mRNA significantly correlated with CYP11B2 mRNA levelsand plasma aldosterone concentrations in APA patients [6970] However the relationship is unclear and disputed ina small-scale clinical study indicating that genetic variationin HSD3B1 affects blood pressure and hypertension in APApatients [71] The results of our simulation study suggest that3120573-HSD protein (human genes are HSD3B1 and HSD3B2) isone of the determination factors for the dynamic propertyof adrenal steroidogenesis Our results support the clinicalevidence of Doi and colleagues [69] and we believe that theHSD3B1 enzyme has a promising potential as novel drugtarget for endocrine hypertension
42 Metabolic Shift of Adrenal Steroidogenesis and the Contri-butions of HSD3B2 and CYP17H The metabolic propertiesof adrenal steroidogenesis in NCI-H295R cells were revealedby dynamic sensitivity analysis using themathematicalmodel(Figure 6) Mineralocorticoids such as DCORTICO COR-TICO and ALDO and intermediate steroids upstream of
the adrenal steroidogenesis pathway such as PREG HPREGDHEA PROG and HPROG were accelerated by reactionsof cholesterol import (119896Cholesterol Transport) StAR protein (119896MTR
119891
and 119896loc119891) and CYP11A1 (119881CYP11A1max ) On the other hand glu-
cocorticoids such as DCORT and CORT and sex hormonessuch as DIONE TESTO and E1 were suppressed by thesemodel parametersTherefore enhancement of the net adrenalsteroidogenesis capacity which supplies PREG precursor tothe pathway causes a production shift from glucocorticoidsto mineralocorticoids by substrate inhibitions of CYP17HHSD3B2 and CYP21A2 caused by accumulation of initialintermediate steroids such as PREG and PROG Sensitivi-ties of CYP17H (119881CYP17HmaxA ) and HSD3B2 (119881HSD3B2
maxA ) for theproducts were dynamically changed and these parametersdetermined the metabolic balance of downstream steroidsin the adrenal steroidogenesis pathway According to theseresults of dynamic sensitivity analysis of StAR CYP11A1CYP17H and HSD3B2 we suggest that the enhancementof CYP17H and HSD3B2 activity during ACTH stimulationwas important to shift the steroidogenic output away fromALDO biosynthesis towards CORT biosynthesis as wellas adrenal androgen production This suggestion partially
16 Journal of Toxicology
supports a comparative animal study in which molecularand cellular variations in CYP17H activity dramatically affectacute cortisol production resulting in distinct physiologicaland behavioral responses [72]
Results of two-dimensional parameter scanning of theenzymatic activities of CYP17H and HSD3B2 quantitativelyshowed the detail of the metabolic relationship betweenmin-eralocorticoid and glucocorticoid biosynthesis (Figure 7)Particularly the results showed that the balance of theseenzymatic activities was very important for the typicalfunction of NCI-H295R cells namely the ability to produceall steroid hormones NCI-H295R cells lost this functionwhen enzymatic activities of CYP17H and HSD3B2 werechanged by over 60 and 30 respectively In addition theybecame mineralocorticoid (ALDO) secreting cells when theenzymatic activity of CYP17H or HSD3B2 was inhibited byover 50 or glucocorticoid (DCORT and CORT) secretingcells when these enzymes were activated by over 50 Inparticular this analysis also showed that HSD3B2 was a keyplayer in the adrenal steroidogenesis of NCI-H295R cellsbecause HSD3B2 inhibition by over 80 almost completelyinhibited the biosynthesis of downstream steroids The ratioof CYP17A1 to HSD3B2 mRNA expression levels has beenrelated to several endocrine diseases with a low level in APAs[73] and high level in cortisol-producing adenomas [74]Furthermore the expression levels or enzymatic activitiesof CYP17A1 and HSD3B1 have been related to androgenproduction in polycystic ovary syndrome [75 76] Theseclinical studies support our simulation results indicating thatthe balance of enzymatic activity of CYP17H and HSD3B2determines the shift in steroidogenic output to mineralocor-ticoids glucocorticoids or androgens
43 Methodologies of Quantitative Mechanistic Analysis forDrug Discovery According to our results obtained using themathematical model of steroidogenesis in NCI-H295R cellssuch as sensitivity analysis comprehensive analysis based onsystems biology is available to quantitatively estimate themechanism of action of steroidogenic disrupting compoundsfrom differential profiling of adrenal steroid hormonesbecause dynamic patterns of steroid hormones in adrenalsteroidogenesis pathway are highly complex Our proposedmethod of quantitative mechanistic analysis of steroidogenicinhibitors was able to predict known action sites in theadrenal steroidogenesis pathway at only one time point (72 hafter drug exposure) Moreover according to the resultsof sensitivity analysis (Figure 6) 119881max of all steroidogenicenzymes was more sensitive than the 119870
119898 because the
intracellular concentrations of steroid hormones were almostmaintained at sufficiently high levels compared with 119870
119898
values of steroidogenic enzymesThese results suggested thatestimation of the mechanism of action of drugs is moreeffective and detectable when using the influences of 119881maxas the searching parameters such as our proposed methodOur data showed that the proposed method based on asystems biology model is a very powerful tool for exploratoryscreening of steroidogenic disrupting compounds
RCGA as a solver of parameter estimation problemsin systems biology has been applied to biological network
identification of gene regulatory networks and metabolicpathways and optimization of biological processes usingexperimentally observed time-course data [77ndash82] In thisstudy RCGA was useful to estimate the mechanism ofaction of novel pharmaceutical drug candidates for adrenalsteroidogenesis as a new application of RCGA in systemsbiology We had two issues when applying RCGA to thequantitative mechanistic analysis of drug actions Theseissues were the vast calculation cost and multimodality ofquasi-optimum solutions in solving the optimization prob-lem because the mathematical model in systems biologyconsists of many equations and parameters A proposedoptimization strategy using RCGA based on REXJGG wasa highly stable and efficient calculation method for a betterquasi-optimum solution than the unimodal normal distri-bution crossover (UNDX)minimum generation gap (MGG)method that is well applied in the engineering field In addi-tion we expanded the RCGA optimization program basedon REXJGG to a hybrid method of GA and then applieda local search as recommended by Harada and Kobayashi[28 83] A final optimal solution was obtained with a goodconvergence property Because these problems are generalin systems biology studies we suggest that the proposedhybrid method based on REXJGG is a very useful tool forquantitative mechanistic analysis of novel pharmaceuticaldrugs not limited to steroidogenic disrupting compounds
5 Conclusions
Thenovelmathematicalmodel of adrenal steroidogenesis wasconstructed in this study including cholesterol transport anddistribution the C
21-steroid hormone pathway steroid trans-
port and cell proliferation which could reproduce adrenalsteroidogenesis in NCI-H295R cells According to the resultsof dynamic sensitivity analysis using the newmodel HSD3B2plays the most important role in the metabolic balance ofadrenal steroidogenesis in NCI-H295R cells Moreover toquantitatively estimate mechanisms of action of adrenal toxiccompounds we analyzed differential metabolic profiles of12 steroid hormones at 3 days after exposure to 11 adrenaltoxic compounds by using the new mathematical modeland a hybrid optimization method of the RCGA and a localsearch (nonlinear least squares) We could estimate whichsteroidogenic enzymes were affected by these compoundsusing the hybrid optimization method Vasculotoxic agentswere estimated to have no effect according to the resultsobtained by our method In terms of adrenal steroidogenicinhibitors the predicted action sites were approximatelymatched to the target enzymes as reported in the literatureThus our computer-aided method based on a systems biol-ogy approach may be useful to analyze the mechanism ofaction of endocrine-disrupting compounds and to assess thehuman adrenal toxicity of novel pharmaceutical drugs basedon steroid hormone profiling
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Journal of Toxicology 17
Authorsrsquo Contribution
Ryuta Saito Natsuko Terasaki and Naohisa Tsutsui designedthe study Naohisa Tsutsui and Naoya Masutomi supervisedthe experiments Natsuko Terasaki performed all in vitroexperiments and collected dataMakotoYamazaki performedall measurements of steroids Ryuta Saito conceived the insilico strategies analyzed data developed all programs andthe mathematical model and performed all simulation anal-yses with Masahiro Okamoto Ryuta Saito Natsuko TerasakiMakoto Yamazaki NaoyaMasutomi andMasahiroOkamotowrote the paper
Acknowledgments
The authors would like to thank Dr Michael S Breenand Dr Miyuki Breen for helpful discussions concerningprevious mathematical models of adrenal steroidogenesisThis work was partially supported by the Grant-in-Aid forScientific Research on Innovative Areas ldquoSynthetic Biologyrdquo(No 23119001 (MO)) from Ministry of Education CultureSports Science and Technology in Japan
References
[1] P W Harvey D J Everett and C J Springall ldquoAdrenal toxi-cology a strategy for assessment of functional toxicity to theadrenal cortex and steroidogenesisrdquo Journal of Applied Toxi-cology vol 27 no 2 pp 103ndash115 2007
[2] W E Ribelin ldquoThe effects of drugs and chemicals upon thestructure of the adrenal glandrdquo Fundamental and Applied Toxi-cology vol 4 no 1 pp 105ndash119 1984
[3] H D Colby and P A Longhurst ldquoToxicology of the adrenalglandrdquo in Endocrine Toxicology C K Atterwill and J D FlackEds pp 243ndash281 CambridgeUniversity Press Cambridge UK1992
[4] H D Colby ldquoIn vitro assessment of adrenocortical toxicityrdquoJournal of Pharmacological and Toxicological Methods vol 32no 1 pp 1ndash6 1994
[5] T J Rosol J T Yarrington J Latendresse and C C CapenldquoAdrenal gland structure function and mechanisms of toxic-ityrdquo Toxicologic Pathology vol 29 no 1 pp 41ndash48 2001
[6] P W Harvey and D J Everett ldquoThe adrenal cortex and ster-oidogenesis as cellular and molecular targets for toxicity crit-ical omissions from regulatory endocrine disrupter screeningstrategies for human healthrdquo Journal of Applied Toxicology vol23 no 2 pp 81ndash87 2003
[7] A F Gazdar H K Oie C H Shackleton et al ldquoEstablishmentand characterization of a human adrenocortical carcinoma cellline that expresses multiple pathways of steroid biosynthesisrdquoCancer Research vol 50 no 17 pp 5488ndash5496 1990
[8] W E Rainey I M Bird C Sawetawan et al ldquoRegulation ofhuman adrenal carcinoma cell (NCI-H295) production of C19steroidsrdquo The Journal of Clinical Endocrinology amp Metabolismvol 77 no 3 pp 731ndash737 1993
[9] B Staels D W Hum andW L Miller ldquoRegulation of steroido-genesis in NCI-H295 cells a cellular model of the human fetaladrenalrdquo Molecular Endocrinology vol 7 no 3 pp 423ndash4331993
[10] M Hecker J L Newsted M B Murphy et al ldquoHuman adreno-carcinoma (H295R) cells for rapid in vitro determination of
effects on steroidogenesis hormone productionrdquo Toxicologyand Applied Pharmacology vol 217 no 1 pp 114ndash124 2006
[11] A Oskarsson E Ulleras K E Plant J P Hinson and P SGoldfarb ldquoSteroidogenic gene expression in H295R cells andthe human adrenal gland adrenotoxic effects of lindane invitrordquo Journal of Applied Toxicology vol 26 no 6 pp 484ndash4922006
[12] UMuller-Vieira M Angotti and RW Hartman ldquoThe adreno-cortical tumor cell line NCI-H295R as an in vitro screeningsystem for the evaluation of CYP11B2 (aldosterone synthase)and CYP11B1 (steroid-11120573-hydroxylase) inhibitorsrdquoThe Journalof Steroid Biochemistry and Molecular Biology vol 96 no 3-4pp 259ndash270 2005
[13] K Imagawa S Okayama M Takaoka et al ldquoInhibitory effectof efonidipine on aldosterone synthesis and secretion in humanadrenocarcinoma (H295R) cellsrdquo Journal of CardiovascularPharmacology vol 47 no 1 pp 133ndash138 2006
[14] M Voets UMuller-Vieira S Marchais-Oberwinkler and RWHartmann ldquoSynthesis of amidinohydrazones and evaluation oftheir inhibitory effect towards aldosterone synthase (CYP11B2)and the formation of selected steroidsrdquo Archiv der Pharmazievol 337 no 7 pp 411ndash416 2004
[15] F K Nielsen C H Hansen J A Fey et al ldquoH295R cells as amodel for steroidogenic disruption a broader perspective usingsimultaneous chemical analysis of 7 key steroid hormonesrdquoToxicology in Vitro vol 26 no 2 pp 343ndash350 2012
[16] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cortisolproduction and on activities of steroidogenic enzymes inhuman adrenocortical H295R cellsrdquo Journal of Steroid Biochem-istry and Molecular Biology vol 80 no 3 pp 355ndash363 2002
[17] R F Canton J T Sanderson S Nijmeijer A Bergman R JLetcher and M van den Berg ldquoIn vitro effects of brominatedflame retardants and metabolites on CYP17 catalytic activity anovel mechanism of actionrdquo Toxicology and Applied Pharma-cology vol 216 no 2 pp 274ndash281 2006
[18] K Hilscherova P D Jones T Gracia et al ldquoAssessment ofthe effects of chemicals on the expression of ten steroidogenicgenes in the H295R cell line using real-time PCRrdquo ToxicologicalSciences vol 81 no 1 pp 78ndash89 2004
[19] T Gracia K Hilscherova P D Jones et al ldquoThe H295R systemfor evaluation of endocrine-disrupting effectsrdquo Ecotoxicologyand Environmental Safety vol 65 no 3 pp 293ndash305 2006
[20] X Zhang J L NewstedM Hecker E B Higley P D Jones andJ P Giesy ldquoClassification of chemicals based on concentration-dependent toxicological data using ToxClustrdquo EnvironmentalScience and Technology vol 43 no 10 pp 3926ndash3932 2009
[21] M S Breen M Breen N Terasaki M Yamazaki and R BConolly ldquoComputational model of steroidogenesis in humanH295R cells to predict biochemical response to endocrine-active chemicalsmodel development formetyraponerdquoEnviron-mental Health Perspectives vol 118 no 2 pp 265ndash272 2010
[22] M Breen M S Breen N Terasaki M Yamazaki A LLloyd and R B Conolly ldquoMechanistic computational model ofsteroidogenesis in H295R cells role of oxysterols and cell pro-liferation to improve predictability of biochemical response toendocrine active chemical-metyraponerdquo Toxicological Sciencesvol 123 no 1 pp 80ndash93 2011
[23] F B Kraemer ldquoAdrenal cholesterol utilizationrdquo Molecular andCellular Endocrinology vol 265-266 pp 42ndash45 2007
[24] W L Miller ldquoSteroidogenic acute regulatory protein (StAR)a novel mitochondrial cholesterol transporterrdquo Biochimica et
18 Journal of Toxicology
Biophysica Acta (BBA)mdashMolecular and Cell Biology of Lipidsvol 1771 no 6 pp 663ndash676 2007
[25] S Szabo and I T Lippe ldquoAdrenal gland chemically inducedstructural and functional changes in the cortexrdquo ToxicologicPathology vol 17 no 2 pp 317ndash329 1989
[26] C Chen-Pan I-J Pan Y Yamamoto T Sakogawa J Yamadaand YHayashi ldquoPrompt recovery of damaged adrenalmedullaeinduced by salinomycinrdquoToxicologic Pathology vol 27 no 5 pp563ndash572 1999
[27] R J Santen and R I Misbin ldquoAminoglutethimide review ofpharmacology and clinical userdquo Pharmacotherapy vol 1 no 2pp 95ndash120 1981
[28] M J Carella N V Dimitrov V V Gossain L Srivastava andD R Rovner ldquoAdrenal effects of low-dose aminoglutethimidewhen used alone in postmenopausal women with advancedbreast cancerrdquoMetabolism vol 43 no 6 pp 723ndash727 1994
[29] M K Johansson J T Sanderson and B-O Lund ldquoEffects of3-MeSO2-DDE and some CYP inhibitors on glucocorticoidsteroidogenesis in the H295R human adrenocortical carcinomacell linerdquo Toxicology in Vitro vol 16 no 2 pp 113ndash121 2002
[30] Y Touitou A Bogdan J C Legrand and P Desgrez ldquoAmino-glutethimide and glutethimide effects on 18-hydroxycorti-costerone biosynthesis by human and sheep adrenals in vitrordquoActa Endocrinologica vol 80 no 3 pp 517ndash526 1975
[31] M M Hart and J A Straw ldquoStudies on the site of action ofop1015840-DDD in the dog adrenal cortex 1 Inhibition of ACTH-mediated pregnenolone synthesisrdquo Steroids vol 17 no 1ndash5 pp559ndash574 1971
[32] M Ojima M Saitoh N Itoh Y Kusano S Fukuchi and HNaganuma ldquoThe effects of op1015840-DDD on adrenal steroidogen-esis and hepatic steroid metabolismrdquo Nippon Naibunpi Gakkaizasshi vol 61 no 3 pp 168ndash178 1985
[33] L F Canosa and N R Ceballos ldquoEffects of different steroid-biosynthesis inhibitors on the testicular steroidogenesis of thetoad Bufo arenarumrdquo Journal of Comparative Physiology B vol171 no 6 pp 519ndash526 2001
[34] D C Kossor S Kominami S Takemori and H D ColbyldquoRole of the steroid 17 alpha-hydroxylase in spironolactone-mediated destruction of adrenal cytochrome P-450rdquoMolecularPharmacology vol 40 no 2 pp 321ndash325 1991
[35] S C Cheng K Suzuki W Sadee and B W Harding ldquoEffectsof spironolactone canrenone and canrenoate K on cytochromeP450 and 11120573 and 18 hydroxylation in bovine and humanadrenal corticalmitochondriardquo Endocrinology vol 99 no 4 pp1097ndash1106 1976
[36] K Yanagibashi M Haniu J E Shively W H Shen and P HallldquoThe synthesis of aldosterone by the adrenal cortex Two zones(fasciculata and glomerulosa) possess one enzyme for 11120573- 18-hydroxylation and aldehyde synthesisrdquoThe Journal of BiologicalChemistry vol 261 no 8 pp 3556ndash3562 1986
[37] K Morishita H Okumura N Ito and N Takahashi ldquoPrimaryculture system of adrenocortical cells from dogs to evaluatedirect effects of chemicals on steroidogenesisrdquo Toxicology vol165 no 2-3 pp 171ndash178 2001
[38] S W J Lamberts E G Bons H A Bruining and F H de JongldquoDifferential effects of the imidazole derivatives etomidateketoconazole and miconazole and of metyrapone on the secre-tion of cortisol and its precursors by human adrenocorticalcellsrdquo Journal of Pharmacology and Experimental Therapeuticsvol 240 no 1 pp 259ndash264 1987
[39] L Yin S Lucas F Maurer U Kazmaier Q Hu and RW Hartmann ldquoNovel imidazol-1-ylmethyl substituted 1256-tetrahydropyrrolo[321-ij]quinolin-4-ones as potent and selec-tive CYP11B1 inhibitors for the treatment of Cushingrsquos syn-dromerdquo Journal ofMedicinal Chemistry vol 55 no 14 pp 6629ndash6633 2012
[40] M DiMattina N Maronian H Ashby D L Loriaux andB D Albertson ldquoKetoconazole inhibits multiple steroidogenicenzymes involved in androgen biosynthesis in the humanovaryrdquo Fertility and Sterility vol 49 no 1 pp 62ndash65 1988
[41] M Ayub and M J Levell ldquoInhibition of human adrenal steroi-dogenic enzymes in vitro by imidazole drugs including keto-conazolerdquo Journal of Steroid Biochemistry vol 32 no 4 pp 515ndash524 1989
[42] D Engelhardt M M Weber T Miksch F Abedinpour andC Jaspers ldquoThe influence of ketoconazole on human adrenalsteroidogenesis incubation studies with tissue slicesrdquo ClinicalEndocrinology vol 35 no 2 pp 163ndash168 1991
[43] S Bhasin S Sikka T Fielder et al ldquoHormonal effects of keto-conazole in vivo in the male rat mechanism of actionrdquo Endo-crinology vol 118 no 3 pp 1229ndash1232 1986
[44] D M Zisterer and D C Williams ldquoCalmidazolium and otherimidazole compounds affect steroidogenesis in Y1 cells lack ofinvolvement of the peripheral-type benzodiazepine receptorrdquoJournal of Steroid Biochemistry and Molecular Biology vol 60no 3-4 pp 189ndash195 1997
[45] L P Walsh C N Kuratko and D M Stocco ldquoEconazoleand miconazole inhibit steroidogenesis and disrupt steroido-genic acute regulatory (StAR) protein expression post-trans-criptionallyrdquo Journal of Steroid Biochemistry and Molecular Bio-logy vol 75 no 4-5 pp 229ndash236 2000
[46] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cor-tisol production and on activities of steroidogenic enzymesin human adrenocortical H295R cellsrdquo The Journal of SteroidBiochemistry and Molecular Biology vol 80 no 3 pp 355ndash3632002
[47] D P Dempsher D S Gann and R D Phair ldquoA mechanisticmodel of ACTH-stimulated cortisol secretionrdquo American Jour-nal of Physiology vol 246 no 4 part 2 pp R587ndashR596 1984
[48] N Sarai S Matsuoka and A Noma ldquosimBio a Java package forthe development of detailed cell modelsrdquo Progress in Biophysicsand Molecular Biology vol 90 no 1ndash3 pp 360ndash377 2006
[49] K Levenberg ldquoA method for the solution of certain non-linearproblems in least squaresrdquo Quarterly of Applied Mathematicsvol 2 pp 164ndash168 1944
[50] D W Marquardt ldquoAn algorithm for least-squares estimation ofnonlinear parametersrdquo SIAM Journal on Applied Mathematicsvol 11 no 2 pp 431ndash441 1963
[51] P E Gill and W Murray ldquoAlgorithms for the solution of thenonlinear least-squares problemrdquo SIAM Journal on NumericalAnalysis vol 15 no 5 pp 977ndash992 1978
[52] Y Akimoto R Hasada J Sakuma I Ono and S KobayashildquoGeneration alternation model for real-coded GA using multi-parent proposal and evaluation of just generation gap (JGG)rdquoin Proceedings of the 19th SICE Symposium on DecentralizedAutonomous Systems pp 341ndash346 Tokyo Japan January 2007
[53] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoProposal and evaluation of adaptive real-coded crossoverAREXrdquo Transactions of the Japanese Society for Artificial Intel-ligence vol 24 no 6 pp 446ndash458 2009
Journal of Toxicology 19
[54] S Kobayashi ldquoThe frontiers of real-coded genetic algorithmsrdquoJournal of Japanese Society for Artificial Intelligence vol 24 no1 pp 147ndash162 2009
[55] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoAnalysis of the behavior of MGG and JGG as a selection modelfor real-coded genetic algorithmsrdquo Transactions of the JapaneseSociety for Artificial Intelligence vol 25 no 2 pp 281ndash289 2010
[56] A Struthers H Krum and G H Williams ldquoA comparisonof the aldosterone-blocking agents eplerenone and spironolac-tonerdquo Clinical Cardiology vol 31 no 4 pp 153ndash158 2008
[57] K Yamasaki M Sawaki S Noda et al ldquoComparison of theHershberger assay and androgen receptor binding assay oftwelve chemicalsrdquo Toxicology vol 195 no 2-3 pp 177ndash1862004
[58] C Eil and S K Edelson ldquoThe use of human skin fibroblasts toobtain potency estimates of drug binding to androgen recept-orsrdquo The Journal of Clinical Endocrinology amp Metabolism vol59 no 1 pp 51ndash55 1984
[59] P Ye T Yamashita D M Pollock H Sasano and W ERainey ldquoContrasting effects of eplerenone and spironolac-tone on adrenal cell steroidogenesisrdquo Hormone and MetabolicResearch vol 41 no 1 pp 35ndash39 2009
[60] B D Albertson K L Frederick N C Maronian et al ldquoTheeffect of ketoconazole on steroidogenesis I Leydig cell enzymeactivity in vitrordquo Research Communications in Chemical Patho-logy and Pharmacology vol 61 no 1 pp 17ndash26 1988
[61] B D Albertson N C Maronian K L Frederick et al ldquoTheeffect of ketoconazole on steroidogenesis II Adrenocorticalenzyme activity in vitrordquoResearch Communications in ChemicalPathology and Pharmacology vol 61 no 1 pp 27ndash34 1988
[62] DM Stocco andB J Clark ldquoRegulation of the acute productionof steroids in steroidogenic cellsrdquo Endocrine Reviews vol 17 no3 pp 221ndash244 1996
[63] L K Christenson and J F Strauss III ldquoSteroidogenic acute regu-latory protein (StAR) and the intramitochondrial translocationof cholesterolrdquo Biochimica et Biophysica Acta (BBA)mdashMolecularand Cell Biology of Lipids vol 1529 no 1ndash3 pp 175ndash187 2000
[64] E R Simpson ldquoCholesterol side-chain cleavage cytochromeP450 and the control of steroidogenesisrdquo Molecular and Cel-lular Endocrinology vol 13 no 3 pp 213ndash227 1979
[65] H A LaVoie and S R King ldquoTranscriptional regulation of ster-oidogenic genes STARD1 CYP11A1 and HSD3Brdquo ExperimentalBiology and Medicine vol 234 no 8 pp 880ndash907 2009
[66] F Raucci A DrsquoAniello and M M Di Fiore ldquoStimulation ofandrogen production by d-aspartate through the enhancementof StAR P450scc and 3120573-HSD mRNA levels in vivo rat testisand in culture of immature rat Leydig cellsrdquo Steroids vol 84pp 103ndash110 2014
[67] M Doi Y Takahashi R Komatsu et al ldquoSalt-sensitive hyper-tension in circadian clock-deficient Cry-null mice involves dys-regulated adrenal Hsd3b6rdquo Nature Medicine vol 16 no 1 pp67ndash74 2010
[68] H Okamura M Doi Y Yamaguchi and J-M Fustin ldquoHyper-tension due to loss of clock novel insight from the molecu-lar analysis of Cry1Cry2-deleted micerdquo Current HypertensionReports vol 13 no 2 pp 103ndash108 2011
[69] M Doi F Satoh T Maekawa et al ldquoIsoform-specific mono-clonal antibodies against 3120573-hydroxysteroid dehydrogenaseisomerase family provide markers for subclassification ofhuman primary aldosteronismrdquo The Journal of Clinical Endo-crinology amp Metabolism vol 99 no 2 pp E257ndashE262 2014
[70] S Konosu-Fukaya Y Nakamura F Satoh et al ldquo3120573-hydroxy-steroid dehydrogenase isoforms in human aldosterone-produc-ing adenomardquo Molecular and Cellular Endocrinology vol 408pp 205ndash212 2015
[71] G C Verwoert J Hofland N Amin et al ldquoExpression and genevariation studies deny association of humanHSD3B1 gene withaldosterone production or blood pressurerdquoAmerican Journal ofHypertension vol 28 no 1 pp 113ndash120 2015
[72] D Hough K Storbeck SW P Cloete A C Swart and P SwartldquoRelative contribution of P450c17 towards the acute cortisolresponse lessons from sheep and goatsrdquoMolecular and CellularEndocrinology vol 408 pp 107ndash113 2015
[73] I Sakuma S Suematsu Y Matsuzawa et al ldquoCharacterizationof steroidogenic enzyme expression in aldosterone-producingadenoma a comparison with various human adrenal tumorsrdquoEndocrine Journal vol 60 no 3 pp 329ndash336 2013
[74] A Tong A Jia S Yan Y Zhang Y Xie and G Liu ldquoEctopiccortisol-producing adrenocortical adenoma in the renal hilumhistopathological features and steroidogenic enzyme profilerdquoInternational Journal of Clinical and Experimental Pathologyvol 7 no 7 pp 4415ndash4421 2014
[75] A Hirsch D Hahn P Kempna et al ldquoMetformin inhibitshuman androgen production by regulating steroidogenicenzymes HSD3B2 and CYP17A1 and complex I activity of therespiratory chainrdquo Endocrinology vol 153 no 9 pp 4354ndash43662012
[76] B Xu L Gao Y Cui et al ldquoSET protein up-regulated test-osterone production in the cultured preantral folliclesrdquo Repro-ductive Biology and Endocrinology vol 11 article 9 2013
[77] S Kikuchi D Tominaga M Arita M Takahashi and MTomita ldquoDynamic modeling of genetic networks using geneticalgorithm and S-systemrdquo Bioinformatics vol 19 no 5 pp 643ndash650 2003
[78] N Shikata YMaki M Nakatsui et al ldquoDetermining importantregulatory relations of amino acids from dynamic networkanalysis of plasma amino acidsrdquo Amino Acids vol 38 no 1 pp179ndash187 2010
[79] M Nakatsui T Ueda Y Maki I Ono and M OkamotoldquoMethod for inferring and extracting reliable genetic interac-tions from time-series profile of gene expressionrdquoMathematicalBiosciences vol 215 no 1 pp 105ndash114 2008
[80] A Komori Y Maki M Nakatsui I Ono and M OkamotoldquoEfficient numerical optimization algorithm based on new real-coded genetic algorithm AREX + JGG and application to theinverse problem in systems biologyrdquo Applied Mathematics vol3 no 10 pp 1463ndash1470 2012
[81] J A Roubos G van Straten and A J B van Boxtel ldquoAn evolu-tionary strategy for fed-batch bioreactor optimization conceptsand performancerdquo Journal of Biotechnology vol 67 no 2-3 pp173ndash187 1999
[82] D Sarkar and J M Modak ldquoGenetic algorithms with filters foroptimal control problems in fed-batch bioreactorsrdquo Bioprocessand Biosystems Engineering vol 26 no 5 pp 295ndash306 2004
[83] K Harada K Ikeda J Sakuma I Ono and S KobayashildquoHybridization of genetic algorithm with local search in mul-tiobjective function optimization recommendation of GA thenLSrdquoTransactions of the Japanese Society for Artificial Intelligencevol 21 no 6 pp 482ndash492 2006
Submit your manuscripts athttpwwwhindawicom
PainResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
ToxinsJournal of
VaccinesJournal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AntibioticsInternational Journal of
ToxicologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
StrokeResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Drug DeliveryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Pharmacological Sciences
Tropical MedicineJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AddictionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Emergency Medicine InternationalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Autoimmune Diseases
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anesthesiology Research and Practice
ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Pharmaceutics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
MEDIATORSINFLAMMATION
of
6 Journal of Toxicology
CHOC
PREG
PROG
DCORTICO
CORTICO
ALDO
CORT
DCORT
HPROG
HPREG DHEA
DIONE
TESTO
CYP11A1
E1
E2
HSD3B2
CYP1
7L
CYP21A2 HSD17B3
CYP1
9A1
CYP11B2
OXYOxysterol synthesis
PREGHPREG DHEAPROG
HPROG DIONE E1DCORTICODCORT TESTO E2CORTICO
CORTALDO
NCI-H295R cell
Medium culture
CYP11B1
CHOL
StAR
CHOM
CHOSCEH
CHOR
CHOLCholesterol transport
Total intracellular cholesterol
CYP1
7H
Vloc Vacc
VMTR
VCEH
Figure 1 Schematic diagram of the mathematical model of adrenal steroidogenesis in NCI-H295R cells Overview of the mathematicalmodel of adrenal steroidogenesis in NCI-H295R cells including cholesterol transport and intracellular localization oxysterol synthesisthe C
21-steroid hormone biosynthesis pathway passive diffusional transport of steroid hormones and cell proliferation CHOL total
cholesterol in medium culture CHOS stored cholesterol esters in the endoplasmic reticulum CHOC intracellular free cholesterolCHOM mitochondrial free cholesterol CHON mitochondrial free cholesterol close to CYP11A1 enzymes CHOR mitochondrial freecholesterol remote from CYP11A1 enzymes PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosteronePROG progesterone HPROG 17120572-hydroxyprogesterone DIONE androstenedione DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICO corticosterone CORT cortisol ALDO aldosterone TESTO testosterone E1 estrone E2 17120573-estradiol OXYoxysterol CEH cholesterol ester hydrolase StAR steroidogenic acute regulatory protein CYP11A1 P450 side chain cleavage enzymeCYP17H 17120572-hydroxylase of CYP17 CYP17L C
17ndash20 lyase of CYP17 HSD3B2 3120573-hydroxysteroid dehydrogenase CYP21A2 21-hydroxylaseCYP11B1 11120573-hydroxylase CYP11B2 18-hydroxylase HSD17B3 17120573-hydroxysteroid dehydrogenase and CYP19A1 aromatase
119894 in compartment ℎ at time point 119895 119883simℎ119894119895
is the simulatedconcentration of molecule 119894 in compartment ℎ at time point119895 and 119883max
ℎ119894is the maximum concentration of molecule 119894 in
compartment ℎ over all time points Data points under thelower quantitation limit were excluded from the evaluationby the objective function
Effects of every static model parameter for parameteroptimization were calculated from differences of fitting theobjective function using sensitivity analysis
224 Quantitative Estimation of the Mechanism of Action ofAdrenal Toxicants Metabolic steroid profiling and differen-tial patterns of the adrenal steroid hormones by chemicalperturbation were reconstructed to optimize the relativeactivities of the steroidogenic enzymes The input datafor the quantitative mechanistic analysis of adrenal toxiccompounds was a fold change (ratio) of the measured 12
steroid concentrations induced by drug exposure for 72 hThe two-step optimization method of the real-coded geneticalgorithm (RCGA) was adopted as a global optimizationmethod in the quantitative mechanistic analysis of adrenaltoxic compoundsThe operations of the crossover and gener-ation alterationmodel in RCGAwere used for the real-codedensemble crossover (REX) and just generation gap (JGG)[52ndash55] As the initial parameters of RCGA maximum gen-eration population size selection size of parent individualspopulation size of child individuals and termination criteriawere 1000 100 6 25 and under 01 of NLSD respectivelyThe search space for the relative activities of the steroidogenicenzymes was from 1100 to 100 To evaluate the fitness ofeach individual the sum of squared residuals for fold changesof measured 12 steroid concentrations was used as theobjective function Nonlinear least squares optimization bythe Levenberg-Marquardt method was used as a local
Journal of Toxicology 7
search [49ndash51] As the estimated mechanisms of actionsof the adrenal toxic compounds the relative activities ofeight steroidogenic enzymes (CYP11A1 CYP17H CYP17LHSD3B2 CYP21A2 CYP11B1 CYP11B2 and HSD17B3)were optimized by the above-mentioned 2-step optimizationmethod Every optimization calculation was duplicated tocheck the numerical stability of the optimal parameters
225 Global Dynamic Sensitivity Analysis The propertyof every kinetic parameter in this computational modelof steroidogenesis in NCI-H295R cells was evaluated bydynamic sensitivity analysis The sensitivity (119878
119909119910) of kinetic
parameter 119909 for variable parameter 119910 was defined by the fol-lowing equation
119878119909119910(119905)=
Δ119910(119905)119910(119905)
Δ119909119909 (2)
where variable parameter119910was the concentration of a steroidhormone in the cytosolic space of NCI-H295R cells Theperturbation for kinetic parameters was +10 (Δ119909119909 = 01)
3 Results
31 Experimental Data on Adrenal Steroidogenesis
311 Adrenal Steroidogenesis of NCI-H295R Cells and theMass Balance All steroid hormones in the culture mediumwere significantly increased after 72 h of stimulation with50 nM ACTH 20120583M forskolin and 100 nM angiotensinII (Figure 2(a)) Mass balances in steroidogenesis of NCI-H295R cells under nontreatment and control (stimulated)conditions are shown in Figures 2(b) and 2(c) respectivelyUnder stimulation the dynamics of net mass in these exper-iments were unchanged and accumulated cholesterol wasconverted to adrenal steroids
312 Cytotoxicity of Adrenal Toxicants Viabilities of cellstreated with each compound were expressed as a relativevalue to the ATP level of the control Effects of AN SM TGFN AGT DDD SP MP KC MC and DZ on cell viabilitywere determined to be valid under 80 of the relative ATPlevel at 7 days after treatment AN SM TG and FN showedcytotoxicity at over 100 1 10 and 10120583M respectively AGTMP andDZ did not affect cell viability at up to 100 120583MDDDSP KC andMC induced less than 80 of cell viability at over100 50 100 and 25 120583M respectively
313 Differentially Steroid Profiling of Adrenal ToxicantsAfter NCI-H295R cells were exposed to each test compoundduring three days the concentrations of 12 steroid hor-mones in the culture mediumwere simultaneously measuredby LCMSMS All effects of the compounds on adrenalsteroidogenesis were evaluated at the concentration withoutany overt cytotoxicity The differential metabolic steroidprofiles of 11 adrenal toxic compounds were classified andvisualized by using hierarchical clustering analysis (Figure 3)
Four adrenal toxicants without steroidogenic inhibitionAN SM TG and FN did not change the medium concen-trations of all steroid hormones by more than 2-fold Above-mentioned 4 compounds at every condition and 7 adrenalsteroidogenic inhibitors at the low exposure concentrationwere gathered into a big cluster as nonchange group The7 steroidogenic inhibitors at the maximum exposure con-centration showed the characteristic steroid profiles eachbut 100 120583M DZ and 10 120583M SP were classified as a clusterAGT drastically decreased the medium concentrations ofPREG HPREG DHEA PROG DCORTICO CORTICOand ALDO at 100 120583M DDD dose-dependently decreased themedium concentrations of PROG DCORTICO CORTICOCORT and ALDO at gt10 120583M and decreased PREG HPREGDHEA PROG HPROG DIONE and DCORT at the maxi-mum exposure concentration of 25120583M SP increased PREGHPREG and DHEA and decreased PROG DIONE DCOR-TICODCORTCORTICOALDO andTESTOat 10120583MMPdose-dependently decreased CORTICO CORT and ALDOand decreased DHEA HPROG DIONE and TESTO at themaximum exposure concentration of 100 120583M KC drasticallydecreased the medium concentrations of PREG HPREGDHEAHPROGDIONEDCORTICODCORTCORTICOCORTO ALDO and TESTO at 10 120583M MC increased themedium concentrations of PROG and decreased DIONEDCORT CORT and TESTO at 10 120583M DZ increased PREGHPREG and DHEA and decreased DIONE DCORTICODCORT CORTICO CORT ALDO and TESTO at 100 120583M
32 The Mathematical Modeling
321 Optimization of the Mathematical Model of AdrenalSteroidogenesis in NCI-H295R Cells The mathematicalmodel of adrenal steroidogenesis in NCI-H295R cells wasoptimized for several kinetic parameters of cholesteroltransport intracellular localization the oxysterol pathwayand maximum velocity of steroidogenic enzymes to fit theexperimental time-course data All optimized kinetic para-meters are shown in Table S1 in a supplementary documentThe optimized mathematical model was reconstructedwith the experimental dynamic patterns of cholesteroland the 14 steroid hormones in the intracellular space andculture medium The fitness was 0621761 of NLSD valuesas the fitting objective function The simulation results andexperimental data are shown in Figure 4
Optimized kinetic parameters were calculated sensitivi-ties for the NLSD value as the fitting score and are shown inTable S1 in a supplementary document The highly sensitiveparameters for fitting the NLSD score were the extractednine kinetic parameters 119896Cholesterol Transport 119896
acc119891 119896loc119891 119896loc119887
119881CYP11A1max 119870CYP11A1
119898 119881CYP17HmaxA 119881HSD3B2
maxA and 119881CYP21A2maxA whichhad higher than 30 fitting sensitivity
33 The Validation Using the Adrenal Toxicants
331 Mechanistic Analysis of Adrenal Toxicants Effects ofadrenal toxic compounds on steroidogenic enzymes werequantitatively predicted from the change in the ratio ofthe measured medium concentrations of the 12 steroid
8 Journal of Toxicology
0001
001
01
1
10
PREG
HPR
EG
DH
EA
PRO
G
HPR
OG
AN
DRO
DCO
RTIC
O
DCO
RT
CORT
ICO
CORT
ALD
O
TEST
O E1 E2C
once
ntra
tion
in m
ediu
m (120583
M)
lowast
lowast
lowast
lowast
lowast
lowast
lowast lowast lowastlowast
lowast
lowast
lowast
lowast
(a)
0 8 24 48 72Time (hours)
0
40
80
120
160
200
240
Num
ber o
f mol
ecul
es (n
mol
)
(b)
0 8 24 48 72Time (hours)
0
40
80
120
160
200
240
Num
ber o
f mol
ecul
es (n
mol
)
(c)
Figure 2 Experimental data of metabolic profiling of adrenal steroid hormones and the mass balance Concentrations of steroid hormonessecreted from NCI-H295R cells in the culture medium at 72 h after stimulation were compared with the untreated condition and thestimulated condition by 50 nM ACTH 20 120583M forskolin and 100 nM angiotensin II (a) Net molecular amounts including cholesterol andsteroid hormones in the culture medium and intracellular space are plotted at five time points (0 8 24 48 and 72 h after stimulation)under the untreated condition (b) and the stimulated condition (c) In the bar graphs dark and light bars indicate the amount of cholesteroland adrenal steroids respectively All data are shown as the mean plusmn SD (119873 = 4) lowast119901 values corrected by the familywise error ratelt001 PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-hydroxyprogesteroneDIONE androstenedioneDCORTICO 11-deoxycorticosteroneDCORT 11-deoxycortisol CORTICO corticosteroneCORT cortisol ALDO aldosterone and TESTO testosterone
hormones at 72 h after drug exposure using themathematicalmodel of adrenal steroidogenesis in NCI-H295R cells Thereproducibility of the estimated results was confirmed byperforming the test twice The estimated effects of 11 adrenaltoxic compounds on eight steroidogenic enzymes are shownin Figure 5 The adrenal toxic compounds without steroido-genic inhibition such as vasculotoxic agents (AN SM TGand FN) were not estimated for the target steroidogenicenzymes under noncytotoxic conditions Every fitness valueswere under 005 of NLSD values used as the fitting objectivefunction (Figures 5(a)ndash5(d)) Other steroidogenic inhibitors(AGT DDD SPMP KCMC andDZ) are described in detailbelow
332 AGT The mechanism of action of AGT in adrenalsteroidogenesis was estimated by inhibition of CYP11A1HSD3B2 CYP21A2 and CYP11B1 at 100 120583M (estimatedinhibitions were 770 780 811 and 598 resp)(Figure 5(e)) AGT has been reported to inhibit CYP11A1
CYP21A2 CYP11B1 and CYP11B2 [6 27ndash30] Our resultswere mostly consistent with the previous reports In particu-lar CYP11A1 appeared to be inhibited strongly byAGT In ourstudy HSD3B2 inhibition of AGT was shown by mechanisticanalysis based on systems biology approaches as a novelmechanism of action of AGT Inhibition of AGT by CYP11B2was not estimated in our study However the concentrationof ALDO in the culture medium decreased to 38 ofthe normal stimulated condition Inhibition of AGT byCYP11B2 has been shown using sheep adrenal homogenatesas well as a human adrenal homogenate from a patientwith Cushingrsquos syndrome [30]The activity of 18-hydroxylaseinduced by CYP11B2 was determined as the conversion ofcorticosterone to 18-hydroxycorticosterone in the previousstudy The cause of the discrepancy regarding the effect ofAGT on CYP11B2 was suggested to be substrate inhibitionbecause the intracellular concentration of CORTICO wasincreased by over 10 times of that in the culture mediumto reach 50 120583M Another possibility was poor quantitative
Journal of Toxicology 9
log2
ratio
4
3
2
1
0
minus1
minus2
minus3
minus4
ALDO
CORTICO
CORT
TESTO
DIONE
HPROG
DCORTICO
DCORT
PROG
PREG
HPREG
DHEA
DZ(
100)
SP(1
0)SP
(1)
DZ(
10)
DD
D(0
1)
AN
(01
)A
N(1
)A
N(1
0)SM
(00
1)TG
(00
1)M
P(0
1)FN
(01
)TG
(1)
AGT(
1)AG
T(0
1)SM
(00
0001
)D
DD
(1)
SM(0
1)
AGT(
10)
MP(
1)FN
(1)
MC(
1)D
Z(1)
KC(0
1)
MC(
01)
DZ(
01)
KC( 1
)D
DD
(10)
MP(
10)
MC(
10)
MP(
100)
AGT(
100)
DD
D(2
5)KC
(10)
Figure 3 Hierarchical cluster analysis of differential metabolic profiles of 12 steroid hormones by exposure to adrenal toxicity compoundsAdrenal toxicants were classified by using the differential metabolic profiling of 12 steroid hormones Concentrations of 12 adrenalsteroids secreted from NCI-H295R cells were drastically changed by the exposure of adrenal steroidogenesis inhibitors The 12 adrenalsteroid hormones were quantitatively measured by LC-MSMS simultaneously Four adrenal vasculotoxic compounds acrylonitrile ANfumaronitrile FN salinomycin SM and thioguanine TG were used as negative control compounds for adrenal steroidogenesis inhibitorsSeven adrenal steroidogenesis inhibitors aminoglutethimide AGT op1015840-DDD DDD spironolactone SP metyrapone MP ketoconazoleKC miconazole MC and daidzein DZ showed a characteristic steroid profile each and were classified as each independent singletonat the maximum exposure condition Exposure concentrations of adrenal toxic compounds were described in brackets of the samplename and the units were prepared as 120583M A blue cluster was classified as a group of nonchange samples including negative controlcompounds and low exposure conditions of adrenal steroidogenesis inhibitors PREG pregnenolone HPREG 17120572-hydroxypregnenoloneDHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-hydroxyprogesterone DIONE androstenedione DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICO corticosterone CORT cortisol ALDO aldosterone and TESTO testosterone
reliability of the experimental data because the ALDOconcentration was under the lower limit of quantification at100 120583MAGT Hecker and colleagues reported that 3 120583MAGTdecreases PREG and PROG concentrations and increases theTESTO concentration [10] However AGT did not increasethe TESTO concentration in our study One possibility isthat the concentration of TESTO was already enhanced byabout 33-fold through stimulation with ACTH forskolinand angiotensin II
333 DDD The mechanism of action of DDD in adrenalsteroidogenesis was estimated by dose-dependent inhibitionof CYP11A1 HSD3B2 CYP21A2 and CYP11B1 (estimatedinhibitions at 25120583M were 870 869 769 and 849resp) (Figure 5(f)) DDD has been reported to inhibitCYP11A1 HSD3B2 CYP21A2 CYP11B1 and CYP11B2 [29 3132] Inhibition of DDD by CYP11B2 was not estimated in ourstudy However the concentration of ALDO in the culturemedium decreased to 3 of that in the normal stimulatedcondition Inhibition of DDD by CYP11B2 has been shownusing mitochondrial and microsomal fractions prepared bystandard centrifugation procedures from a bovine adrenal
cortex homogenate [32]The cause of the discrepancy regard-ing the inhibition ofDDDbyCYP11B2 could not be explainedby same effect in the case of AGT
334 SP Themechanism of action of SP in adrenal steroido-genesis was estimated by inhibition of HSD3B2 CYP21A2and HSD17B3 (estimated inhibitions at 10120583M were 702595 and 593 resp) (Figure 5(g)) SP has been reportedto inhibit CYP17H CYP17L CYP11B1 and CYP11B2 [6 33ndash35] The inhibitory effect of SP on the HSD3B2 enzyme wasa novel mechanism of action The main action of SP is asa mineralocorticoid receptor (MR) antagonist SP has alsobeen reported to exert some off-target effects by binding toandrogen glucocorticoid and progesterone receptors [56ndash58] SP has been shown to inhibit the production of ALDOand CORT from PREG induced by angiotensin II in H295Rcells but the specific MR antagonist eplerenone did not showthe inhibitory effects [59] Therefore HSD3B2 inhibition bySP is not mediated via MR and the action might be directinhibition of HSD3B2 enzymes or a part of known off-targeteffects mediated through other nuclear hormone receptorsRegarding CYP17H and CYP17L our results were consistent
10 Journal of Toxicology
CHOL
0
20
40
60
80
100C
once
ntra
tion
in m
ediu
m (120583
M)
10 20 30 40 50 60 70 800Time (hours)
(a)
PREG
DCORT
DCORTICO
CORTICO
CORT
Con
cent
ratio
n in
med
ium
(120583M
)
10
4
2
0
8
6
10 20 30 40 50 60 70 800Time (hours)
(b)
Con
cent
ratio
n in
med
ium
(120583M
)
0
020406
081
1214
0 10 20 30 40 50 60 70 80Time (hours)
HPREG
DIONE
PROGHPROG
E1
(c)
DHEAE2
ALDO
TESTO
0
005
01
015
02
025
Con
cent
ratio
n in
med
ium
(120583M
)
0 10 20 30Time (hours)
40 50 60 70 80
(d)
CHOL
0
5000
10000
15000
20000
25000
30000
Con
cent
ratio
n in
cell
(120583M
)
10 20 30 40 50 60 70 800Time (hours)
(e)
Con
cent
ratio
n in
cell
(120583M
)
0
50
100
150
200
250
0 10 20 30 40 50 60 70 80Time (hours)
PREG
DCORT
DCORTICO
CORTICOCORT
(f)
0 10 20 30 40 50 60 70 80Time (hours)
HPREG
DIONE
PROGHPROG
E1
Con
cent
ratio
n in
cell
(120583M
)
3025201510
35
05
(g)
005
115
2
253
35
Con
cent
ratio
n in
cell
(120583M
)
0 10 20 30 40 50 60 70 80Time (hours)
DHEA
E2ALDO
TESTO
(h)
Figure 4 Comparison of time-course profiles of cholesterol and adrenal steroids produced by NCI-H295R cells between experimentallymeasured and simulated data To intuitively confirm reconstruction of the measured experimental data in the developed simulation modelof NCI-H295R cells dynamics of cholesterol and adrenal steroids produced by NCI-H295R cells were plotted to overlay experimentaldata with the simulated results Graphs show the dynamics of medium concentrations of cholesterol (a) and adrenal steroids ((b)ndash(d))and intracellular concentrations of cholesterol (e) and adrenal steroids ((f)ndash(h)) Steroid hormones were categorized into three groups byconcentration levels Major steroids were PREG pregnenolone DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICOcorticosterone and CORT cortisol ((b) and (f)) Moderate steroids were HPREG 17120572-hydroxypregnenolone PROG progesterone HPROG17120572-hydroxyprogesterone DIONE androstenedione and E1 estrone ((c) and (g)) Minor steroids were DHEA dehydroepiandrosteroneALDO aldosterone TESTO testosterone and E2 17120573-estradiol ((d) and (h)) Experimental data are shown as symbols with dotted lines Alldata represent the mean plusmn SD (119873 = 4) Simulation data are shown as solid lines
Journal of Toxicology 11
0110
10
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
AN (120583M)
(a)
0110
FN (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(b)
00101
SM (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
100E minus 05
(c)
0110
TG (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B1
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(d)
0110
10100
AGT (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(e)
0110
1025
DDD (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(f)
1010
SP (120583M)
2750005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(g)
0110
10100
MP (120583M)
2110005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(h)
Figure 5 Continued
12 Journal of Toxicology
0110
10
KC (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(i)
0110
10
MC (120583M)
4430005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(j)
0110
10100
DZ (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(k)
Figure 5 Estimated mechanism of action of adrenal toxicants by using the mathematical model of adrenal steroidogenesis in NCI-H295Rcells Mechanisms of action of adrenal toxicity compounds were quantitatively estimated from experimental results of differential steroidprofiling using the mathematical model of adrenal steroidogenesis in NCI-H295R cells The drug action was defined as a scaling factorof enzymatic activity in the simulation model These scaling factors were optimized to fit experimental data by a hybrid optimizationmethod of the RCGA and nonlinear least squares Estimated drug actions by the exposure of vasculotoxic agents acrylonitrile AN (a)fumaronitrile FN (b) salinomycin SM (c) and thioguanine TG (d) and the steroidogenic inhibitors aminoglutethimide AGT (e) op1015840-DDD DDD (f) spironolactone SP (g) metyrapone MP (h) ketoconazole KC (i) miconazole MC (j) and daidzein DZ (k) are shownas a spider radar chart CYP11A1 P450 side chain cleavage enzyme CYP17H 17120572-hydroxylase of CYP17 CYP17L C
17ndash20 lyase of CYP17HSD3B2 3120573-hydroxysteroid dehydrogenase CYP21A2 21-hydroxylase CYP11B1 11120573-hydroxylase CYP11B2 18-hydroxylase and HSD17B317120573-hydroxysteroid dehydrogenase
with previous reports [33 34] 7120572-thiospironolactone whichis synthesized by deacetylation of SP inhibits CYP17H andCYP17L [34] The fact that there were no inhibitions ofCYP17HorCYP17L in our study suggests that SPmight not bedeacetylated to 7120572-thiospironolactone in NCI-H295R cellsRegarding CYP11B1 and CYP11B2 our results were unclearcompared with a previous study It has been shown that30 120583M SP inhibits CYP11B1 and CYP11B2 in human andbovine adrenal mitochondria [35]The cause of CYP11B1 andCYP11B2 inhibition by SP could not be determined in ourstudy which might be due to the lower maximum exposureconcentration of SP than that in the previous reportWe couldnot examine SP concentrations over 10 120583M because theseconcentrations were cytotoxic in NCI-H295R cells
335 MP The mechanism of action of MP in adrenalsteroidogenesis was estimated by dose-dependent inhibitionof CYP11B1 (estimated inhibitions at 1 10 and 100 120583M were571 827 and 982 resp) (Figure 5(h)) MP has beenreported to inhibit CYP11B1 as its major effect and CYP11A1and CYP11B2 as a weak effect [6 29 36ndash39] The results wereable to show that MP is a selective inhibitor of CYP11B1 inthe previous report However the estimated effect of MPat a high concentration 100 120583M as the maximum exposureconcentration was unclear According to the previous reportselectivity of MP for CYP11B1CYP11B2 is about five times[39] In addition 20 120583M MP has a slight inhibition effect onCYP11A1 in H295R cells [29]
336 KC The mechanism of action of KC in adrenalsteroidogenesis was estimated by inhibition of CYP11A1CYP17H CYP17L HSD3B2 CYP21A2 CYP11B1 and
CYP11B2 (estimated inhibitions at 10 120583Mwere 926 943518 830 882 974 and 798 resp) (Figure 5(i))KC has been reported to inhibit CYP11A1 CYP17H CYP17LHSD3B2 CYP21A2 and CYP11B1 [6 29 40ndash43] Our resultswere almost consistent with the previous reports KC inhibitsCYP11A1 CYP17H CYP21A2 and CYP11B1 in NCI-H295Rcells at 10 120583M [29] and CYP17H CYP17L CYP21A2 andCYP11B1 in human adrenal mitochondria and Leydig cellmicrosomes at 2ndash5120583M [60 61] However KC has shownonly weak inhibition of HSD3B2 and HSD17B3 in Leydigcells at the millimolar level [60 61] Regarding CYP11B2 andHSD17B3 we considered that these estimated inhibitions ofKC did have sufficient reliability in terms of quantitative pre-diction precision because ALDO and TESTO concentrationswere less than the lower limit of quantification at 10 120583MKC
337 MC Themechanism of action of MC in adrenal stero-idogenesis was estimated by inhibition of CYP17H CYP17LCYP11B1 and HSD17B3 (estimated inhibitions at 10120583Mwere691 530 764 and 571 resp) (Figure 5(j)) MC hasbeen reported to inhibit not only CYP17H and CYP17L butalsoCYP11A1 CYP21A2 andCYP11B1 [41 44 45]The resultsin the previous reports were able to estimate that MC isa CYP17 inhibitor However CYP11A1 inhibition by MCprobably instead of a reduction in StAR expression wasnot clearly detected in our study using NCI-H295R cellsbecause there were no decreases in the concentrations ofPREG and PROG in the culture medium Indirect inhibitionof CYP11A1 via the peripheral-type benzodiazepine receptorhas been reported in mouse adrenocortical Y-1 cells treatedwith MC in the absence of stimuli by measuring PREGproduction [44] On the other hand reductions of StAR
Journal of Toxicology 13
protein expression andor transport activity without affectingtotal steroid synthesis or CYP11A1 and HSD3B2 enzymeexpression or activities have been reported in (BU)
2cAMP-
stimulated MA-10 Leydig tumor cells treated with MC bymeasuring PROG production [45] Therefore the effect ofMC on the initial reaction in adrenal steroidogenesis fromcholesterol should be different according to the cell type andstimulation condition Inhibition of CYP21A2 and CYP11B1by MC has been reported as decreases in the consumptionof PROG and DCORTICO respectively [41] Inhibition ofCYP11B1 was estimated by the action of MC in this studybut that of CYP21A2 was not detected In the previousexperimental report inhibitory sites by MCmight have beenreflected by inhibition of CYP17H activity because CYP21A2activity was measured as a decrease in labeled PROG
338 DZ The mechanism of action of DZ in adrenalsteroidogenesis was estimated by inhibition of CYP11A1HSD3B2 CYP21A2 CYP11B1 and HSD17B3 (estimated inhi-bitions at 100 120583M were 586 941 965 872 and981 resp) (Figure 5(k)) DZ has been reported to inhibitHSD3B2 and CYP21A2 [46] The results of HSD3B2 andCYP21A2 were consistent with the previous report Howeverinhibitions have not been reported for CYP11A1 CYP11B1and HSD17B3These estimated effects of DZ on CYP11B1 andHSD17B3 were unclear because the concentrations of ALDOand TESTO were less than the lower limit of quantificationat 100 120583M In addition these enzymes act downstream of thestrong action points of DZ such as HSD3B2 and CYP21A2
34 The Simulations and the Systematic Model Analysis
341 Dynamic Sensitivity Analysis of Adrenal SteroidogenesisTo comprehensively understand the dynamics of adrenalsteroidogenesis dynamic sensitivities were calculated forsteroid concentrations secreted by NCI-H295R cells usingour constructed mathematical model of steroidogenesis Theresults of dynamic sensitivity analysis at 72 h of duration and6 h of interval time are presented as a heat-map in Figure 6
The top 10 parameters of the total area under the curveof dynamic sensitivity for cholesterol and the 14 steroids inculture medium were 119881HSD3B2
maxA 119881CYP11A1max 119881CYP21A2maxA 119896loc119891 119896loc119887
119870CYP11A1119898
119881CYP17HmaxA 119896Cholesterol Transport 119896acc119891 and 119881CYP21A2maxB in
order from the top Cholesterol uptake (119896Cholesterol Transport)StAR protein (119896loc
119891) and CYP11A1 (119881CYP11A1max ) which are
determining factors of the capacity for steroidogenesis pro-moted the production of mineralocorticoids (DCORTICOCORTICO and ALDO) and restrained the synthesis of glu-cocorticoids (DCORT and CORT) and sex steroids (DIONETESTO and E1) because of the accumulation of interme-diate molecules in steroidogenesis (PREG HPREG DHEAPROG and HPROG) only by self-activation The dynamicpatterns of the intermediate molecules in steroidogenesiswere mainly dependent on the activity of CYP17H andHSD3B2 with PREG as the substrate of these enzymes inwhich the dynamic sensitivities of 119881CYP17HmaxA for HPREG andHPROG and 119881HSD3B2
maxA for PROG HPROG and DCORTICO
reversed the direction of sensitivity at 49ndash66 h after stimu-lation The dynamic sensitivities of the maximum activitiesof HSD3B2 for PREG (119881HSD3B2
maxA ) and CYP21A2 for PROG(119881CYP21A2maxA ) were related to all steroids at 72 h Almost allmodel parameters had positive sensitivity for downstreamsteroids in the adrenal steroidogenic pathway and negativesensitivity for direct-binding steroids as substrates of thesteroidogenic enzyme The sensitivity of 119881max in all steroido-genic enzymes was relatively higher than 119870
119898for the same
steroid substrate
342 Simulation of the Metabolic Balance of AdrenalSteroidogenesis Pathway To clearly show the property of themetabolic shift between mineralocorticoid and glucocorti-coid biosynthesis we performed two-dimensional parameterscanning of the enzymatic activities of CYP17H and HSD3B2(Figure 7) NCI-H295R cells lost the ability to produce allsteroid hormones when enzymatic activities of CYP17H andHSD3B2 were changed by over 60 and 30 respectivelyActivation ofCYP17HandorHSD3B2 induced themetabolicshift that enhanced the glucocorticoid biosynthesis and devi-ated from the mineralocorticoid biosynthesis On the otherhand inhibition of CYP17H andor HSD3B2 induced themetabolic shift that enhanced the mineralocorticoid biosyn-thesis and deviated from the glucocorticoid biosynthesisMoreover the enzymatic activity of HSD3B2 regulated themetabolic balance of sex steroids and the precursors onadrenal steroidogenesis of NCI-H295R cells E1 TESTO andDIONE were produced by NCI-H295R cells when activatingthe enzymatic activity of HSD3B2 Conversely E2 andDHEAwere produced by NCI-H295R cells when suppressing theenzymatic activity of HSD3B2 The biosynthesis of down-stream steroids in adrenal steroidogenesis pathway such asmineralocorticoids and glucocorticoids was almost com-pletely terminated when the enzymatic activity of HSD3B2was decreased by over 80
4 Discussion
41 Importance of 3120573-HSDActivity in Adrenal SteroidogenesisOur systematic analysis using the mathematical model ofadrenal steroidogenesis in NCI-H295R cells revealed thatthe enzymatic activity of 3120573-HSD controls the dynamics ofadrenal steroidogenesis The activity of the StAR proteincontrols the net capacity of steroidogenesis in steroidogeniccells which is the transport of cholesterol from the outer toinner mitochondrial membranes Both the expression levelsof StAR protein and mRNA are rapidly elevated in responseto stimulation by tropic hormones such as ACTH [62 63]Another important factor in adrenal steroidogenesisis is thecholesterol side chain cleavage enzyme CYP11A1 the firstrate-limiting and hormonally regulated step in the synthesisof all steroid hormones which is conversion of cholesterolto pregnenolone in mitochondria [64] According to ourresults of global sensitivity analysis (Supplementary Table 1and Figure 6(d)) in addition to CYP11A1 and StAR proteins3120573-HSD was one of the key regulators in adrenal steroido-genesis of NCI-H295R cells And also this result suggeststhat a significant regulatory mechanism in steroidogenesis
14 Journal of Toxicology
60544830 4236 66 7218 246 120
Time (hours)
0
1
2
3C
once
ntra
tion
(a)
60544830 4236 66 7218 246 120
Time (hours)
minus05
0
05
minus1
1
Sens
itivi
ty
(b)
0
6
12
18
24
30
36
42
48
54
60
66
72
Sens
itivi
ty fo
r med
ium
conc
entr
atio
ns
of ch
oles
tero
l or s
tero
id h
orm
ones
Dur
atio
n tim
e (ho
urs)
Model parameters
+10
+05
00
minus05
minus10
(c)
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
Parameter number in the mathematical model of steroidogenesis in NCI-H295R cells
CHOL
PREG
HPREG
DHEA
PROG
HPROG
DIONE
DCORTICO
DCORT
CORTICO
CORT
ALDO
TESTO
E1
E2
Chol
este
rol o
r ste
roid
hor
mon
es
0
05
10
15
20
minus05
minus10
minus15
minus20
(d)
Figure 6 Heat-map of the global dynamic sensitivity analysis of adrenal steroid concentrations produced by NCI-H295R cells The globaldynamic sensitivity analysis is a powerful tool to comprehensively understand the dependencies of themodel parameters in themathematicalmodel of the biological complex system Dynamic sensitivities of model parameters in the mathematical model of adrenal steroidogenesisin NCI-H295R cells were calculated for all steroid concentrations in the culture medium every 6 h until 72 h after stimulation To clarify theview of heat-map of global dynamic sensitivity analysis imaginary data of the dynamics of steroid concentrations in the original model (blueline) and perturbed model for sensitivity analysis (red line) were prepared (a) Using this imaginary data the calculated dynamic sensitivities(b) and the visualized dynamic sensitivity as one block of the heat-map (c) were shown respectively By the same method that explainedusing imaginary data the large-scale data of the global dynamic sensitivity analysis on the mathematical model of adrenal steroidogenesisin NCI-H295R cells was comprehensively visualized as a big graph of heat-map (d) Parameter numbers in the horizontal axis are (1)119896Choresterol Transport (2) 119896CEH (3) 119896MTR
119891 (4) 119896MTR
119887 (5) 119896acc
119891 (6) 119896acc
119887 (7) 119896loc
119891 (8) 119896loc
119887 (9) 119896Oxysterol Synthesis (10)119870
CYP11A1119898
(11) 119881CYP11A1max (12)119870CYP17H119898A
(13) 119870CYP17H119898B (14) 119881CYP17HmaxA (15) 119881CYP17HmaxB (16) 119881CYP17HmaxB (17) 119870CYP17L
119898B (18) 119881CYP17LmaxA (19) 119881CYP17LmaxB (20) 119870HSD3B2119898A (21) 119870HSD3B2
119898B (22) 119870HSD3B2119898C (23)
119881HSD3B2maxA (24) 119881HSD3B2
maxB (25) 119881HSD3B2maxB (26) 119870CYP21A2
119898A (27) 119870CYP21A2119898B (28) 119881CYP21A2maxA (29) 119881CYP21A2maxB (30) 119870CYP11B1
119898A (31) 119870CYP11B1119898B (32) 119881CYP11B1maxA
(33) 119881CYP11B1maxB (34) 119896CYP11B2 (35) 119870HSD17B3119898A (36) 119870HSD17B3
119898B (37) 119881HSD17B3maxA (38) 119881HSD17B3
maxB (39) 119870CYP19A1119898A (40) 119870CYP19A1
119898B (41) 119881CYP19A1maxA and (42)119881
CYP19A1maxB PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-
hydroxyprogesteroneDIONE androstenedioneDCORTICO 11-deoxycorticosteroneDCORT 11-deoxycortisol CORTICO corticosteroneCORT cortisol ALDO aldosterone and TESTO testosterone
pathway is very reasonable StAR CYP11A1 and 3120573-HSD(isoforms 1 or 2 in humans) proteins generally respond to thesame hormones that stimulate steroid production throughcommon pathways such as cAMP signaling in adrenal glandsand testes [65 66] Moreover our data also support recentexperimental evidence from clinical and in vivo studiessuggesting that the enzymatic activity of 3120573-HSD plays
an important role in the regulation of mineralocorticoidsynthesis in adrenal steroidogenesis and contributes to hyper-tension caused by abnormal overproduction of aldosterone[67ndash70] Circadian clock-deficient Cry-null mice show salt-sensitive hypertension due to abnormally high synthesis ofaldosterone which is caused by constitutively high expressionof HSD3B6mRNA and protein in the adrenal cortex [67 68]
Journal of Toxicology 15
All steroid producing cells
Mineralocorticoid producing cells Glucocorticoid producing cells
Mineralocorticoid and glucocorticoid producing cells
Terminated steroidogenesis
Sex steroids and precursor sex steroidsDIONE TESTO E1DEHA E2
NCI-H295Rcells at controlcondition
Normalized activities of HSD3B2
Nor
mal
ized
activ
ities
of C
YP17
H
2
18
16
14
12
1
08
06
04
02
00 05 1 15 2
Figure 7 Metabolic categories of the steroidogenic cells determined by the balance of HSD3B2 and CYP17H activitiesThe two-dimensionalparameter scanning analysis by the perturbation of two focused parameters clarifies the interaction and the relationship between the twoparameters in the complex system Functional cellular categories of steroidogenic cells were defined by the levels of mineralocorticoid(ALDO) glucocorticoids (DCORTandCORT) and androgens (DHEAandDIONE) at 72 h after stimulation Enzymatic activities ofHSD3B2andCYP17were normalized by standard values of the simulationmodel inNCI-H295R cells Scanning ranges ofHSD3B2 andCYP17 activitieswere 0ndash200 each 10 Green regions are all 14 steroids produced by NCI-H295R cells Red blue and purple regions are mineralocorticoidproducing cells glucocorticoids-producing cells and both corticoid-producing cells respectively Yellow regions are steroidogenesis of NCI-H295R cells terminated upstream of the adrenal steroidogenesis pathway
Recent clinical observations have revealed predominantexpression of HSD3B2 mRNA and protein in tumor cellsof aldosterone-producing adenoma (APA) and HSD3B1mRNA significantly correlated with CYP11B2 mRNA levelsand plasma aldosterone concentrations in APA patients [6970] However the relationship is unclear and disputed ina small-scale clinical study indicating that genetic variationin HSD3B1 affects blood pressure and hypertension in APApatients [71] The results of our simulation study suggest that3120573-HSD protein (human genes are HSD3B1 and HSD3B2) isone of the determination factors for the dynamic propertyof adrenal steroidogenesis Our results support the clinicalevidence of Doi and colleagues [69] and we believe that theHSD3B1 enzyme has a promising potential as novel drugtarget for endocrine hypertension
42 Metabolic Shift of Adrenal Steroidogenesis and the Contri-butions of HSD3B2 and CYP17H The metabolic propertiesof adrenal steroidogenesis in NCI-H295R cells were revealedby dynamic sensitivity analysis using themathematicalmodel(Figure 6) Mineralocorticoids such as DCORTICO COR-TICO and ALDO and intermediate steroids upstream of
the adrenal steroidogenesis pathway such as PREG HPREGDHEA PROG and HPROG were accelerated by reactionsof cholesterol import (119896Cholesterol Transport) StAR protein (119896MTR
119891
and 119896loc119891) and CYP11A1 (119881CYP11A1max ) On the other hand glu-
cocorticoids such as DCORT and CORT and sex hormonessuch as DIONE TESTO and E1 were suppressed by thesemodel parametersTherefore enhancement of the net adrenalsteroidogenesis capacity which supplies PREG precursor tothe pathway causes a production shift from glucocorticoidsto mineralocorticoids by substrate inhibitions of CYP17HHSD3B2 and CYP21A2 caused by accumulation of initialintermediate steroids such as PREG and PROG Sensitivi-ties of CYP17H (119881CYP17HmaxA ) and HSD3B2 (119881HSD3B2
maxA ) for theproducts were dynamically changed and these parametersdetermined the metabolic balance of downstream steroidsin the adrenal steroidogenesis pathway According to theseresults of dynamic sensitivity analysis of StAR CYP11A1CYP17H and HSD3B2 we suggest that the enhancementof CYP17H and HSD3B2 activity during ACTH stimulationwas important to shift the steroidogenic output away fromALDO biosynthesis towards CORT biosynthesis as wellas adrenal androgen production This suggestion partially
16 Journal of Toxicology
supports a comparative animal study in which molecularand cellular variations in CYP17H activity dramatically affectacute cortisol production resulting in distinct physiologicaland behavioral responses [72]
Results of two-dimensional parameter scanning of theenzymatic activities of CYP17H and HSD3B2 quantitativelyshowed the detail of the metabolic relationship betweenmin-eralocorticoid and glucocorticoid biosynthesis (Figure 7)Particularly the results showed that the balance of theseenzymatic activities was very important for the typicalfunction of NCI-H295R cells namely the ability to produceall steroid hormones NCI-H295R cells lost this functionwhen enzymatic activities of CYP17H and HSD3B2 werechanged by over 60 and 30 respectively In addition theybecame mineralocorticoid (ALDO) secreting cells when theenzymatic activity of CYP17H or HSD3B2 was inhibited byover 50 or glucocorticoid (DCORT and CORT) secretingcells when these enzymes were activated by over 50 Inparticular this analysis also showed that HSD3B2 was a keyplayer in the adrenal steroidogenesis of NCI-H295R cellsbecause HSD3B2 inhibition by over 80 almost completelyinhibited the biosynthesis of downstream steroids The ratioof CYP17A1 to HSD3B2 mRNA expression levels has beenrelated to several endocrine diseases with a low level in APAs[73] and high level in cortisol-producing adenomas [74]Furthermore the expression levels or enzymatic activitiesof CYP17A1 and HSD3B1 have been related to androgenproduction in polycystic ovary syndrome [75 76] Theseclinical studies support our simulation results indicating thatthe balance of enzymatic activity of CYP17H and HSD3B2determines the shift in steroidogenic output to mineralocor-ticoids glucocorticoids or androgens
43 Methodologies of Quantitative Mechanistic Analysis forDrug Discovery According to our results obtained using themathematical model of steroidogenesis in NCI-H295R cellssuch as sensitivity analysis comprehensive analysis based onsystems biology is available to quantitatively estimate themechanism of action of steroidogenic disrupting compoundsfrom differential profiling of adrenal steroid hormonesbecause dynamic patterns of steroid hormones in adrenalsteroidogenesis pathway are highly complex Our proposedmethod of quantitative mechanistic analysis of steroidogenicinhibitors was able to predict known action sites in theadrenal steroidogenesis pathway at only one time point (72 hafter drug exposure) Moreover according to the resultsof sensitivity analysis (Figure 6) 119881max of all steroidogenicenzymes was more sensitive than the 119870
119898 because the
intracellular concentrations of steroid hormones were almostmaintained at sufficiently high levels compared with 119870
119898
values of steroidogenic enzymesThese results suggested thatestimation of the mechanism of action of drugs is moreeffective and detectable when using the influences of 119881maxas the searching parameters such as our proposed methodOur data showed that the proposed method based on asystems biology model is a very powerful tool for exploratoryscreening of steroidogenic disrupting compounds
RCGA as a solver of parameter estimation problemsin systems biology has been applied to biological network
identification of gene regulatory networks and metabolicpathways and optimization of biological processes usingexperimentally observed time-course data [77ndash82] In thisstudy RCGA was useful to estimate the mechanism ofaction of novel pharmaceutical drug candidates for adrenalsteroidogenesis as a new application of RCGA in systemsbiology We had two issues when applying RCGA to thequantitative mechanistic analysis of drug actions Theseissues were the vast calculation cost and multimodality ofquasi-optimum solutions in solving the optimization prob-lem because the mathematical model in systems biologyconsists of many equations and parameters A proposedoptimization strategy using RCGA based on REXJGG wasa highly stable and efficient calculation method for a betterquasi-optimum solution than the unimodal normal distri-bution crossover (UNDX)minimum generation gap (MGG)method that is well applied in the engineering field In addi-tion we expanded the RCGA optimization program basedon REXJGG to a hybrid method of GA and then applieda local search as recommended by Harada and Kobayashi[28 83] A final optimal solution was obtained with a goodconvergence property Because these problems are generalin systems biology studies we suggest that the proposedhybrid method based on REXJGG is a very useful tool forquantitative mechanistic analysis of novel pharmaceuticaldrugs not limited to steroidogenic disrupting compounds
5 Conclusions
Thenovelmathematicalmodel of adrenal steroidogenesis wasconstructed in this study including cholesterol transport anddistribution the C
21-steroid hormone pathway steroid trans-
port and cell proliferation which could reproduce adrenalsteroidogenesis in NCI-H295R cells According to the resultsof dynamic sensitivity analysis using the newmodel HSD3B2plays the most important role in the metabolic balance ofadrenal steroidogenesis in NCI-H295R cells Moreover toquantitatively estimate mechanisms of action of adrenal toxiccompounds we analyzed differential metabolic profiles of12 steroid hormones at 3 days after exposure to 11 adrenaltoxic compounds by using the new mathematical modeland a hybrid optimization method of the RCGA and a localsearch (nonlinear least squares) We could estimate whichsteroidogenic enzymes were affected by these compoundsusing the hybrid optimization method Vasculotoxic agentswere estimated to have no effect according to the resultsobtained by our method In terms of adrenal steroidogenicinhibitors the predicted action sites were approximatelymatched to the target enzymes as reported in the literatureThus our computer-aided method based on a systems biol-ogy approach may be useful to analyze the mechanism ofaction of endocrine-disrupting compounds and to assess thehuman adrenal toxicity of novel pharmaceutical drugs basedon steroid hormone profiling
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Journal of Toxicology 17
Authorsrsquo Contribution
Ryuta Saito Natsuko Terasaki and Naohisa Tsutsui designedthe study Naohisa Tsutsui and Naoya Masutomi supervisedthe experiments Natsuko Terasaki performed all in vitroexperiments and collected dataMakotoYamazaki performedall measurements of steroids Ryuta Saito conceived the insilico strategies analyzed data developed all programs andthe mathematical model and performed all simulation anal-yses with Masahiro Okamoto Ryuta Saito Natsuko TerasakiMakoto Yamazaki NaoyaMasutomi andMasahiroOkamotowrote the paper
Acknowledgments
The authors would like to thank Dr Michael S Breenand Dr Miyuki Breen for helpful discussions concerningprevious mathematical models of adrenal steroidogenesisThis work was partially supported by the Grant-in-Aid forScientific Research on Innovative Areas ldquoSynthetic Biologyrdquo(No 23119001 (MO)) from Ministry of Education CultureSports Science and Technology in Japan
References
[1] P W Harvey D J Everett and C J Springall ldquoAdrenal toxi-cology a strategy for assessment of functional toxicity to theadrenal cortex and steroidogenesisrdquo Journal of Applied Toxi-cology vol 27 no 2 pp 103ndash115 2007
[2] W E Ribelin ldquoThe effects of drugs and chemicals upon thestructure of the adrenal glandrdquo Fundamental and Applied Toxi-cology vol 4 no 1 pp 105ndash119 1984
[3] H D Colby and P A Longhurst ldquoToxicology of the adrenalglandrdquo in Endocrine Toxicology C K Atterwill and J D FlackEds pp 243ndash281 CambridgeUniversity Press Cambridge UK1992
[4] H D Colby ldquoIn vitro assessment of adrenocortical toxicityrdquoJournal of Pharmacological and Toxicological Methods vol 32no 1 pp 1ndash6 1994
[5] T J Rosol J T Yarrington J Latendresse and C C CapenldquoAdrenal gland structure function and mechanisms of toxic-ityrdquo Toxicologic Pathology vol 29 no 1 pp 41ndash48 2001
[6] P W Harvey and D J Everett ldquoThe adrenal cortex and ster-oidogenesis as cellular and molecular targets for toxicity crit-ical omissions from regulatory endocrine disrupter screeningstrategies for human healthrdquo Journal of Applied Toxicology vol23 no 2 pp 81ndash87 2003
[7] A F Gazdar H K Oie C H Shackleton et al ldquoEstablishmentand characterization of a human adrenocortical carcinoma cellline that expresses multiple pathways of steroid biosynthesisrdquoCancer Research vol 50 no 17 pp 5488ndash5496 1990
[8] W E Rainey I M Bird C Sawetawan et al ldquoRegulation ofhuman adrenal carcinoma cell (NCI-H295) production of C19steroidsrdquo The Journal of Clinical Endocrinology amp Metabolismvol 77 no 3 pp 731ndash737 1993
[9] B Staels D W Hum andW L Miller ldquoRegulation of steroido-genesis in NCI-H295 cells a cellular model of the human fetaladrenalrdquo Molecular Endocrinology vol 7 no 3 pp 423ndash4331993
[10] M Hecker J L Newsted M B Murphy et al ldquoHuman adreno-carcinoma (H295R) cells for rapid in vitro determination of
effects on steroidogenesis hormone productionrdquo Toxicologyand Applied Pharmacology vol 217 no 1 pp 114ndash124 2006
[11] A Oskarsson E Ulleras K E Plant J P Hinson and P SGoldfarb ldquoSteroidogenic gene expression in H295R cells andthe human adrenal gland adrenotoxic effects of lindane invitrordquo Journal of Applied Toxicology vol 26 no 6 pp 484ndash4922006
[12] UMuller-Vieira M Angotti and RW Hartman ldquoThe adreno-cortical tumor cell line NCI-H295R as an in vitro screeningsystem for the evaluation of CYP11B2 (aldosterone synthase)and CYP11B1 (steroid-11120573-hydroxylase) inhibitorsrdquoThe Journalof Steroid Biochemistry and Molecular Biology vol 96 no 3-4pp 259ndash270 2005
[13] K Imagawa S Okayama M Takaoka et al ldquoInhibitory effectof efonidipine on aldosterone synthesis and secretion in humanadrenocarcinoma (H295R) cellsrdquo Journal of CardiovascularPharmacology vol 47 no 1 pp 133ndash138 2006
[14] M Voets UMuller-Vieira S Marchais-Oberwinkler and RWHartmann ldquoSynthesis of amidinohydrazones and evaluation oftheir inhibitory effect towards aldosterone synthase (CYP11B2)and the formation of selected steroidsrdquo Archiv der Pharmazievol 337 no 7 pp 411ndash416 2004
[15] F K Nielsen C H Hansen J A Fey et al ldquoH295R cells as amodel for steroidogenic disruption a broader perspective usingsimultaneous chemical analysis of 7 key steroid hormonesrdquoToxicology in Vitro vol 26 no 2 pp 343ndash350 2012
[16] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cortisolproduction and on activities of steroidogenic enzymes inhuman adrenocortical H295R cellsrdquo Journal of Steroid Biochem-istry and Molecular Biology vol 80 no 3 pp 355ndash363 2002
[17] R F Canton J T Sanderson S Nijmeijer A Bergman R JLetcher and M van den Berg ldquoIn vitro effects of brominatedflame retardants and metabolites on CYP17 catalytic activity anovel mechanism of actionrdquo Toxicology and Applied Pharma-cology vol 216 no 2 pp 274ndash281 2006
[18] K Hilscherova P D Jones T Gracia et al ldquoAssessment ofthe effects of chemicals on the expression of ten steroidogenicgenes in the H295R cell line using real-time PCRrdquo ToxicologicalSciences vol 81 no 1 pp 78ndash89 2004
[19] T Gracia K Hilscherova P D Jones et al ldquoThe H295R systemfor evaluation of endocrine-disrupting effectsrdquo Ecotoxicologyand Environmental Safety vol 65 no 3 pp 293ndash305 2006
[20] X Zhang J L NewstedM Hecker E B Higley P D Jones andJ P Giesy ldquoClassification of chemicals based on concentration-dependent toxicological data using ToxClustrdquo EnvironmentalScience and Technology vol 43 no 10 pp 3926ndash3932 2009
[21] M S Breen M Breen N Terasaki M Yamazaki and R BConolly ldquoComputational model of steroidogenesis in humanH295R cells to predict biochemical response to endocrine-active chemicalsmodel development formetyraponerdquoEnviron-mental Health Perspectives vol 118 no 2 pp 265ndash272 2010
[22] M Breen M S Breen N Terasaki M Yamazaki A LLloyd and R B Conolly ldquoMechanistic computational model ofsteroidogenesis in H295R cells role of oxysterols and cell pro-liferation to improve predictability of biochemical response toendocrine active chemical-metyraponerdquo Toxicological Sciencesvol 123 no 1 pp 80ndash93 2011
[23] F B Kraemer ldquoAdrenal cholesterol utilizationrdquo Molecular andCellular Endocrinology vol 265-266 pp 42ndash45 2007
[24] W L Miller ldquoSteroidogenic acute regulatory protein (StAR)a novel mitochondrial cholesterol transporterrdquo Biochimica et
18 Journal of Toxicology
Biophysica Acta (BBA)mdashMolecular and Cell Biology of Lipidsvol 1771 no 6 pp 663ndash676 2007
[25] S Szabo and I T Lippe ldquoAdrenal gland chemically inducedstructural and functional changes in the cortexrdquo ToxicologicPathology vol 17 no 2 pp 317ndash329 1989
[26] C Chen-Pan I-J Pan Y Yamamoto T Sakogawa J Yamadaand YHayashi ldquoPrompt recovery of damaged adrenalmedullaeinduced by salinomycinrdquoToxicologic Pathology vol 27 no 5 pp563ndash572 1999
[27] R J Santen and R I Misbin ldquoAminoglutethimide review ofpharmacology and clinical userdquo Pharmacotherapy vol 1 no 2pp 95ndash120 1981
[28] M J Carella N V Dimitrov V V Gossain L Srivastava andD R Rovner ldquoAdrenal effects of low-dose aminoglutethimidewhen used alone in postmenopausal women with advancedbreast cancerrdquoMetabolism vol 43 no 6 pp 723ndash727 1994
[29] M K Johansson J T Sanderson and B-O Lund ldquoEffects of3-MeSO2-DDE and some CYP inhibitors on glucocorticoidsteroidogenesis in the H295R human adrenocortical carcinomacell linerdquo Toxicology in Vitro vol 16 no 2 pp 113ndash121 2002
[30] Y Touitou A Bogdan J C Legrand and P Desgrez ldquoAmino-glutethimide and glutethimide effects on 18-hydroxycorti-costerone biosynthesis by human and sheep adrenals in vitrordquoActa Endocrinologica vol 80 no 3 pp 517ndash526 1975
[31] M M Hart and J A Straw ldquoStudies on the site of action ofop1015840-DDD in the dog adrenal cortex 1 Inhibition of ACTH-mediated pregnenolone synthesisrdquo Steroids vol 17 no 1ndash5 pp559ndash574 1971
[32] M Ojima M Saitoh N Itoh Y Kusano S Fukuchi and HNaganuma ldquoThe effects of op1015840-DDD on adrenal steroidogen-esis and hepatic steroid metabolismrdquo Nippon Naibunpi Gakkaizasshi vol 61 no 3 pp 168ndash178 1985
[33] L F Canosa and N R Ceballos ldquoEffects of different steroid-biosynthesis inhibitors on the testicular steroidogenesis of thetoad Bufo arenarumrdquo Journal of Comparative Physiology B vol171 no 6 pp 519ndash526 2001
[34] D C Kossor S Kominami S Takemori and H D ColbyldquoRole of the steroid 17 alpha-hydroxylase in spironolactone-mediated destruction of adrenal cytochrome P-450rdquoMolecularPharmacology vol 40 no 2 pp 321ndash325 1991
[35] S C Cheng K Suzuki W Sadee and B W Harding ldquoEffectsof spironolactone canrenone and canrenoate K on cytochromeP450 and 11120573 and 18 hydroxylation in bovine and humanadrenal corticalmitochondriardquo Endocrinology vol 99 no 4 pp1097ndash1106 1976
[36] K Yanagibashi M Haniu J E Shively W H Shen and P HallldquoThe synthesis of aldosterone by the adrenal cortex Two zones(fasciculata and glomerulosa) possess one enzyme for 11120573- 18-hydroxylation and aldehyde synthesisrdquoThe Journal of BiologicalChemistry vol 261 no 8 pp 3556ndash3562 1986
[37] K Morishita H Okumura N Ito and N Takahashi ldquoPrimaryculture system of adrenocortical cells from dogs to evaluatedirect effects of chemicals on steroidogenesisrdquo Toxicology vol165 no 2-3 pp 171ndash178 2001
[38] S W J Lamberts E G Bons H A Bruining and F H de JongldquoDifferential effects of the imidazole derivatives etomidateketoconazole and miconazole and of metyrapone on the secre-tion of cortisol and its precursors by human adrenocorticalcellsrdquo Journal of Pharmacology and Experimental Therapeuticsvol 240 no 1 pp 259ndash264 1987
[39] L Yin S Lucas F Maurer U Kazmaier Q Hu and RW Hartmann ldquoNovel imidazol-1-ylmethyl substituted 1256-tetrahydropyrrolo[321-ij]quinolin-4-ones as potent and selec-tive CYP11B1 inhibitors for the treatment of Cushingrsquos syn-dromerdquo Journal ofMedicinal Chemistry vol 55 no 14 pp 6629ndash6633 2012
[40] M DiMattina N Maronian H Ashby D L Loriaux andB D Albertson ldquoKetoconazole inhibits multiple steroidogenicenzymes involved in androgen biosynthesis in the humanovaryrdquo Fertility and Sterility vol 49 no 1 pp 62ndash65 1988
[41] M Ayub and M J Levell ldquoInhibition of human adrenal steroi-dogenic enzymes in vitro by imidazole drugs including keto-conazolerdquo Journal of Steroid Biochemistry vol 32 no 4 pp 515ndash524 1989
[42] D Engelhardt M M Weber T Miksch F Abedinpour andC Jaspers ldquoThe influence of ketoconazole on human adrenalsteroidogenesis incubation studies with tissue slicesrdquo ClinicalEndocrinology vol 35 no 2 pp 163ndash168 1991
[43] S Bhasin S Sikka T Fielder et al ldquoHormonal effects of keto-conazole in vivo in the male rat mechanism of actionrdquo Endo-crinology vol 118 no 3 pp 1229ndash1232 1986
[44] D M Zisterer and D C Williams ldquoCalmidazolium and otherimidazole compounds affect steroidogenesis in Y1 cells lack ofinvolvement of the peripheral-type benzodiazepine receptorrdquoJournal of Steroid Biochemistry and Molecular Biology vol 60no 3-4 pp 189ndash195 1997
[45] L P Walsh C N Kuratko and D M Stocco ldquoEconazoleand miconazole inhibit steroidogenesis and disrupt steroido-genic acute regulatory (StAR) protein expression post-trans-criptionallyrdquo Journal of Steroid Biochemistry and Molecular Bio-logy vol 75 no 4-5 pp 229ndash236 2000
[46] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cor-tisol production and on activities of steroidogenic enzymesin human adrenocortical H295R cellsrdquo The Journal of SteroidBiochemistry and Molecular Biology vol 80 no 3 pp 355ndash3632002
[47] D P Dempsher D S Gann and R D Phair ldquoA mechanisticmodel of ACTH-stimulated cortisol secretionrdquo American Jour-nal of Physiology vol 246 no 4 part 2 pp R587ndashR596 1984
[48] N Sarai S Matsuoka and A Noma ldquosimBio a Java package forthe development of detailed cell modelsrdquo Progress in Biophysicsand Molecular Biology vol 90 no 1ndash3 pp 360ndash377 2006
[49] K Levenberg ldquoA method for the solution of certain non-linearproblems in least squaresrdquo Quarterly of Applied Mathematicsvol 2 pp 164ndash168 1944
[50] D W Marquardt ldquoAn algorithm for least-squares estimation ofnonlinear parametersrdquo SIAM Journal on Applied Mathematicsvol 11 no 2 pp 431ndash441 1963
[51] P E Gill and W Murray ldquoAlgorithms for the solution of thenonlinear least-squares problemrdquo SIAM Journal on NumericalAnalysis vol 15 no 5 pp 977ndash992 1978
[52] Y Akimoto R Hasada J Sakuma I Ono and S KobayashildquoGeneration alternation model for real-coded GA using multi-parent proposal and evaluation of just generation gap (JGG)rdquoin Proceedings of the 19th SICE Symposium on DecentralizedAutonomous Systems pp 341ndash346 Tokyo Japan January 2007
[53] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoProposal and evaluation of adaptive real-coded crossoverAREXrdquo Transactions of the Japanese Society for Artificial Intel-ligence vol 24 no 6 pp 446ndash458 2009
Journal of Toxicology 19
[54] S Kobayashi ldquoThe frontiers of real-coded genetic algorithmsrdquoJournal of Japanese Society for Artificial Intelligence vol 24 no1 pp 147ndash162 2009
[55] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoAnalysis of the behavior of MGG and JGG as a selection modelfor real-coded genetic algorithmsrdquo Transactions of the JapaneseSociety for Artificial Intelligence vol 25 no 2 pp 281ndash289 2010
[56] A Struthers H Krum and G H Williams ldquoA comparisonof the aldosterone-blocking agents eplerenone and spironolac-tonerdquo Clinical Cardiology vol 31 no 4 pp 153ndash158 2008
[57] K Yamasaki M Sawaki S Noda et al ldquoComparison of theHershberger assay and androgen receptor binding assay oftwelve chemicalsrdquo Toxicology vol 195 no 2-3 pp 177ndash1862004
[58] C Eil and S K Edelson ldquoThe use of human skin fibroblasts toobtain potency estimates of drug binding to androgen recept-orsrdquo The Journal of Clinical Endocrinology amp Metabolism vol59 no 1 pp 51ndash55 1984
[59] P Ye T Yamashita D M Pollock H Sasano and W ERainey ldquoContrasting effects of eplerenone and spironolac-tone on adrenal cell steroidogenesisrdquo Hormone and MetabolicResearch vol 41 no 1 pp 35ndash39 2009
[60] B D Albertson K L Frederick N C Maronian et al ldquoTheeffect of ketoconazole on steroidogenesis I Leydig cell enzymeactivity in vitrordquo Research Communications in Chemical Patho-logy and Pharmacology vol 61 no 1 pp 17ndash26 1988
[61] B D Albertson N C Maronian K L Frederick et al ldquoTheeffect of ketoconazole on steroidogenesis II Adrenocorticalenzyme activity in vitrordquoResearch Communications in ChemicalPathology and Pharmacology vol 61 no 1 pp 27ndash34 1988
[62] DM Stocco andB J Clark ldquoRegulation of the acute productionof steroids in steroidogenic cellsrdquo Endocrine Reviews vol 17 no3 pp 221ndash244 1996
[63] L K Christenson and J F Strauss III ldquoSteroidogenic acute regu-latory protein (StAR) and the intramitochondrial translocationof cholesterolrdquo Biochimica et Biophysica Acta (BBA)mdashMolecularand Cell Biology of Lipids vol 1529 no 1ndash3 pp 175ndash187 2000
[64] E R Simpson ldquoCholesterol side-chain cleavage cytochromeP450 and the control of steroidogenesisrdquo Molecular and Cel-lular Endocrinology vol 13 no 3 pp 213ndash227 1979
[65] H A LaVoie and S R King ldquoTranscriptional regulation of ster-oidogenic genes STARD1 CYP11A1 and HSD3Brdquo ExperimentalBiology and Medicine vol 234 no 8 pp 880ndash907 2009
[66] F Raucci A DrsquoAniello and M M Di Fiore ldquoStimulation ofandrogen production by d-aspartate through the enhancementof StAR P450scc and 3120573-HSD mRNA levels in vivo rat testisand in culture of immature rat Leydig cellsrdquo Steroids vol 84pp 103ndash110 2014
[67] M Doi Y Takahashi R Komatsu et al ldquoSalt-sensitive hyper-tension in circadian clock-deficient Cry-null mice involves dys-regulated adrenal Hsd3b6rdquo Nature Medicine vol 16 no 1 pp67ndash74 2010
[68] H Okamura M Doi Y Yamaguchi and J-M Fustin ldquoHyper-tension due to loss of clock novel insight from the molecu-lar analysis of Cry1Cry2-deleted micerdquo Current HypertensionReports vol 13 no 2 pp 103ndash108 2011
[69] M Doi F Satoh T Maekawa et al ldquoIsoform-specific mono-clonal antibodies against 3120573-hydroxysteroid dehydrogenaseisomerase family provide markers for subclassification ofhuman primary aldosteronismrdquo The Journal of Clinical Endo-crinology amp Metabolism vol 99 no 2 pp E257ndashE262 2014
[70] S Konosu-Fukaya Y Nakamura F Satoh et al ldquo3120573-hydroxy-steroid dehydrogenase isoforms in human aldosterone-produc-ing adenomardquo Molecular and Cellular Endocrinology vol 408pp 205ndash212 2015
[71] G C Verwoert J Hofland N Amin et al ldquoExpression and genevariation studies deny association of humanHSD3B1 gene withaldosterone production or blood pressurerdquoAmerican Journal ofHypertension vol 28 no 1 pp 113ndash120 2015
[72] D Hough K Storbeck SW P Cloete A C Swart and P SwartldquoRelative contribution of P450c17 towards the acute cortisolresponse lessons from sheep and goatsrdquoMolecular and CellularEndocrinology vol 408 pp 107ndash113 2015
[73] I Sakuma S Suematsu Y Matsuzawa et al ldquoCharacterizationof steroidogenic enzyme expression in aldosterone-producingadenoma a comparison with various human adrenal tumorsrdquoEndocrine Journal vol 60 no 3 pp 329ndash336 2013
[74] A Tong A Jia S Yan Y Zhang Y Xie and G Liu ldquoEctopiccortisol-producing adrenocortical adenoma in the renal hilumhistopathological features and steroidogenic enzyme profilerdquoInternational Journal of Clinical and Experimental Pathologyvol 7 no 7 pp 4415ndash4421 2014
[75] A Hirsch D Hahn P Kempna et al ldquoMetformin inhibitshuman androgen production by regulating steroidogenicenzymes HSD3B2 and CYP17A1 and complex I activity of therespiratory chainrdquo Endocrinology vol 153 no 9 pp 4354ndash43662012
[76] B Xu L Gao Y Cui et al ldquoSET protein up-regulated test-osterone production in the cultured preantral folliclesrdquo Repro-ductive Biology and Endocrinology vol 11 article 9 2013
[77] S Kikuchi D Tominaga M Arita M Takahashi and MTomita ldquoDynamic modeling of genetic networks using geneticalgorithm and S-systemrdquo Bioinformatics vol 19 no 5 pp 643ndash650 2003
[78] N Shikata YMaki M Nakatsui et al ldquoDetermining importantregulatory relations of amino acids from dynamic networkanalysis of plasma amino acidsrdquo Amino Acids vol 38 no 1 pp179ndash187 2010
[79] M Nakatsui T Ueda Y Maki I Ono and M OkamotoldquoMethod for inferring and extracting reliable genetic interac-tions from time-series profile of gene expressionrdquoMathematicalBiosciences vol 215 no 1 pp 105ndash114 2008
[80] A Komori Y Maki M Nakatsui I Ono and M OkamotoldquoEfficient numerical optimization algorithm based on new real-coded genetic algorithm AREX + JGG and application to theinverse problem in systems biologyrdquo Applied Mathematics vol3 no 10 pp 1463ndash1470 2012
[81] J A Roubos G van Straten and A J B van Boxtel ldquoAn evolu-tionary strategy for fed-batch bioreactor optimization conceptsand performancerdquo Journal of Biotechnology vol 67 no 2-3 pp173ndash187 1999
[82] D Sarkar and J M Modak ldquoGenetic algorithms with filters foroptimal control problems in fed-batch bioreactorsrdquo Bioprocessand Biosystems Engineering vol 26 no 5 pp 295ndash306 2004
[83] K Harada K Ikeda J Sakuma I Ono and S KobayashildquoHybridization of genetic algorithm with local search in mul-tiobjective function optimization recommendation of GA thenLSrdquoTransactions of the Japanese Society for Artificial Intelligencevol 21 no 6 pp 482ndash492 2006
Submit your manuscripts athttpwwwhindawicom
PainResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
ToxinsJournal of
VaccinesJournal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AntibioticsInternational Journal of
ToxicologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
StrokeResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Drug DeliveryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Pharmacological Sciences
Tropical MedicineJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AddictionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Emergency Medicine InternationalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Autoimmune Diseases
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anesthesiology Research and Practice
ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Pharmaceutics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
MEDIATORSINFLAMMATION
of
Journal of Toxicology 7
search [49ndash51] As the estimated mechanisms of actionsof the adrenal toxic compounds the relative activities ofeight steroidogenic enzymes (CYP11A1 CYP17H CYP17LHSD3B2 CYP21A2 CYP11B1 CYP11B2 and HSD17B3)were optimized by the above-mentioned 2-step optimizationmethod Every optimization calculation was duplicated tocheck the numerical stability of the optimal parameters
225 Global Dynamic Sensitivity Analysis The propertyof every kinetic parameter in this computational modelof steroidogenesis in NCI-H295R cells was evaluated bydynamic sensitivity analysis The sensitivity (119878
119909119910) of kinetic
parameter 119909 for variable parameter 119910 was defined by the fol-lowing equation
119878119909119910(119905)=
Δ119910(119905)119910(119905)
Δ119909119909 (2)
where variable parameter119910was the concentration of a steroidhormone in the cytosolic space of NCI-H295R cells Theperturbation for kinetic parameters was +10 (Δ119909119909 = 01)
3 Results
31 Experimental Data on Adrenal Steroidogenesis
311 Adrenal Steroidogenesis of NCI-H295R Cells and theMass Balance All steroid hormones in the culture mediumwere significantly increased after 72 h of stimulation with50 nM ACTH 20120583M forskolin and 100 nM angiotensinII (Figure 2(a)) Mass balances in steroidogenesis of NCI-H295R cells under nontreatment and control (stimulated)conditions are shown in Figures 2(b) and 2(c) respectivelyUnder stimulation the dynamics of net mass in these exper-iments were unchanged and accumulated cholesterol wasconverted to adrenal steroids
312 Cytotoxicity of Adrenal Toxicants Viabilities of cellstreated with each compound were expressed as a relativevalue to the ATP level of the control Effects of AN SM TGFN AGT DDD SP MP KC MC and DZ on cell viabilitywere determined to be valid under 80 of the relative ATPlevel at 7 days after treatment AN SM TG and FN showedcytotoxicity at over 100 1 10 and 10120583M respectively AGTMP andDZ did not affect cell viability at up to 100 120583MDDDSP KC andMC induced less than 80 of cell viability at over100 50 100 and 25 120583M respectively
313 Differentially Steroid Profiling of Adrenal ToxicantsAfter NCI-H295R cells were exposed to each test compoundduring three days the concentrations of 12 steroid hor-mones in the culture mediumwere simultaneously measuredby LCMSMS All effects of the compounds on adrenalsteroidogenesis were evaluated at the concentration withoutany overt cytotoxicity The differential metabolic steroidprofiles of 11 adrenal toxic compounds were classified andvisualized by using hierarchical clustering analysis (Figure 3)
Four adrenal toxicants without steroidogenic inhibitionAN SM TG and FN did not change the medium concen-trations of all steroid hormones by more than 2-fold Above-mentioned 4 compounds at every condition and 7 adrenalsteroidogenic inhibitors at the low exposure concentrationwere gathered into a big cluster as nonchange group The7 steroidogenic inhibitors at the maximum exposure con-centration showed the characteristic steroid profiles eachbut 100 120583M DZ and 10 120583M SP were classified as a clusterAGT drastically decreased the medium concentrations ofPREG HPREG DHEA PROG DCORTICO CORTICOand ALDO at 100 120583M DDD dose-dependently decreased themedium concentrations of PROG DCORTICO CORTICOCORT and ALDO at gt10 120583M and decreased PREG HPREGDHEA PROG HPROG DIONE and DCORT at the maxi-mum exposure concentration of 25120583M SP increased PREGHPREG and DHEA and decreased PROG DIONE DCOR-TICODCORTCORTICOALDO andTESTOat 10120583MMPdose-dependently decreased CORTICO CORT and ALDOand decreased DHEA HPROG DIONE and TESTO at themaximum exposure concentration of 100 120583M KC drasticallydecreased the medium concentrations of PREG HPREGDHEAHPROGDIONEDCORTICODCORTCORTICOCORTO ALDO and TESTO at 10 120583M MC increased themedium concentrations of PROG and decreased DIONEDCORT CORT and TESTO at 10 120583M DZ increased PREGHPREG and DHEA and decreased DIONE DCORTICODCORT CORTICO CORT ALDO and TESTO at 100 120583M
32 The Mathematical Modeling
321 Optimization of the Mathematical Model of AdrenalSteroidogenesis in NCI-H295R Cells The mathematicalmodel of adrenal steroidogenesis in NCI-H295R cells wasoptimized for several kinetic parameters of cholesteroltransport intracellular localization the oxysterol pathwayand maximum velocity of steroidogenic enzymes to fit theexperimental time-course data All optimized kinetic para-meters are shown in Table S1 in a supplementary documentThe optimized mathematical model was reconstructedwith the experimental dynamic patterns of cholesteroland the 14 steroid hormones in the intracellular space andculture medium The fitness was 0621761 of NLSD valuesas the fitting objective function The simulation results andexperimental data are shown in Figure 4
Optimized kinetic parameters were calculated sensitivi-ties for the NLSD value as the fitting score and are shown inTable S1 in a supplementary document The highly sensitiveparameters for fitting the NLSD score were the extractednine kinetic parameters 119896Cholesterol Transport 119896
acc119891 119896loc119891 119896loc119887
119881CYP11A1max 119870CYP11A1
119898 119881CYP17HmaxA 119881HSD3B2
maxA and 119881CYP21A2maxA whichhad higher than 30 fitting sensitivity
33 The Validation Using the Adrenal Toxicants
331 Mechanistic Analysis of Adrenal Toxicants Effects ofadrenal toxic compounds on steroidogenic enzymes werequantitatively predicted from the change in the ratio ofthe measured medium concentrations of the 12 steroid
8 Journal of Toxicology
0001
001
01
1
10
PREG
HPR
EG
DH
EA
PRO
G
HPR
OG
AN
DRO
DCO
RTIC
O
DCO
RT
CORT
ICO
CORT
ALD
O
TEST
O E1 E2C
once
ntra
tion
in m
ediu
m (120583
M)
lowast
lowast
lowast
lowast
lowast
lowast
lowast lowast lowastlowast
lowast
lowast
lowast
lowast
(a)
0 8 24 48 72Time (hours)
0
40
80
120
160
200
240
Num
ber o
f mol
ecul
es (n
mol
)
(b)
0 8 24 48 72Time (hours)
0
40
80
120
160
200
240
Num
ber o
f mol
ecul
es (n
mol
)
(c)
Figure 2 Experimental data of metabolic profiling of adrenal steroid hormones and the mass balance Concentrations of steroid hormonessecreted from NCI-H295R cells in the culture medium at 72 h after stimulation were compared with the untreated condition and thestimulated condition by 50 nM ACTH 20 120583M forskolin and 100 nM angiotensin II (a) Net molecular amounts including cholesterol andsteroid hormones in the culture medium and intracellular space are plotted at five time points (0 8 24 48 and 72 h after stimulation)under the untreated condition (b) and the stimulated condition (c) In the bar graphs dark and light bars indicate the amount of cholesteroland adrenal steroids respectively All data are shown as the mean plusmn SD (119873 = 4) lowast119901 values corrected by the familywise error ratelt001 PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-hydroxyprogesteroneDIONE androstenedioneDCORTICO 11-deoxycorticosteroneDCORT 11-deoxycortisol CORTICO corticosteroneCORT cortisol ALDO aldosterone and TESTO testosterone
hormones at 72 h after drug exposure using themathematicalmodel of adrenal steroidogenesis in NCI-H295R cells Thereproducibility of the estimated results was confirmed byperforming the test twice The estimated effects of 11 adrenaltoxic compounds on eight steroidogenic enzymes are shownin Figure 5 The adrenal toxic compounds without steroido-genic inhibition such as vasculotoxic agents (AN SM TGand FN) were not estimated for the target steroidogenicenzymes under noncytotoxic conditions Every fitness valueswere under 005 of NLSD values used as the fitting objectivefunction (Figures 5(a)ndash5(d)) Other steroidogenic inhibitors(AGT DDD SPMP KCMC andDZ) are described in detailbelow
332 AGT The mechanism of action of AGT in adrenalsteroidogenesis was estimated by inhibition of CYP11A1HSD3B2 CYP21A2 and CYP11B1 at 100 120583M (estimatedinhibitions were 770 780 811 and 598 resp)(Figure 5(e)) AGT has been reported to inhibit CYP11A1
CYP21A2 CYP11B1 and CYP11B2 [6 27ndash30] Our resultswere mostly consistent with the previous reports In particu-lar CYP11A1 appeared to be inhibited strongly byAGT In ourstudy HSD3B2 inhibition of AGT was shown by mechanisticanalysis based on systems biology approaches as a novelmechanism of action of AGT Inhibition of AGT by CYP11B2was not estimated in our study However the concentrationof ALDO in the culture medium decreased to 38 ofthe normal stimulated condition Inhibition of AGT byCYP11B2 has been shown using sheep adrenal homogenatesas well as a human adrenal homogenate from a patientwith Cushingrsquos syndrome [30]The activity of 18-hydroxylaseinduced by CYP11B2 was determined as the conversion ofcorticosterone to 18-hydroxycorticosterone in the previousstudy The cause of the discrepancy regarding the effect ofAGT on CYP11B2 was suggested to be substrate inhibitionbecause the intracellular concentration of CORTICO wasincreased by over 10 times of that in the culture mediumto reach 50 120583M Another possibility was poor quantitative
Journal of Toxicology 9
log2
ratio
4
3
2
1
0
minus1
minus2
minus3
minus4
ALDO
CORTICO
CORT
TESTO
DIONE
HPROG
DCORTICO
DCORT
PROG
PREG
HPREG
DHEA
DZ(
100)
SP(1
0)SP
(1)
DZ(
10)
DD
D(0
1)
AN
(01
)A
N(1
)A
N(1
0)SM
(00
1)TG
(00
1)M
P(0
1)FN
(01
)TG
(1)
AGT(
1)AG
T(0
1)SM
(00
0001
)D
DD
(1)
SM(0
1)
AGT(
10)
MP(
1)FN
(1)
MC(
1)D
Z(1)
KC(0
1)
MC(
01)
DZ(
01)
KC( 1
)D
DD
(10)
MP(
10)
MC(
10)
MP(
100)
AGT(
100)
DD
D(2
5)KC
(10)
Figure 3 Hierarchical cluster analysis of differential metabolic profiles of 12 steroid hormones by exposure to adrenal toxicity compoundsAdrenal toxicants were classified by using the differential metabolic profiling of 12 steroid hormones Concentrations of 12 adrenalsteroids secreted from NCI-H295R cells were drastically changed by the exposure of adrenal steroidogenesis inhibitors The 12 adrenalsteroid hormones were quantitatively measured by LC-MSMS simultaneously Four adrenal vasculotoxic compounds acrylonitrile ANfumaronitrile FN salinomycin SM and thioguanine TG were used as negative control compounds for adrenal steroidogenesis inhibitorsSeven adrenal steroidogenesis inhibitors aminoglutethimide AGT op1015840-DDD DDD spironolactone SP metyrapone MP ketoconazoleKC miconazole MC and daidzein DZ showed a characteristic steroid profile each and were classified as each independent singletonat the maximum exposure condition Exposure concentrations of adrenal toxic compounds were described in brackets of the samplename and the units were prepared as 120583M A blue cluster was classified as a group of nonchange samples including negative controlcompounds and low exposure conditions of adrenal steroidogenesis inhibitors PREG pregnenolone HPREG 17120572-hydroxypregnenoloneDHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-hydroxyprogesterone DIONE androstenedione DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICO corticosterone CORT cortisol ALDO aldosterone and TESTO testosterone
reliability of the experimental data because the ALDOconcentration was under the lower limit of quantification at100 120583MAGT Hecker and colleagues reported that 3 120583MAGTdecreases PREG and PROG concentrations and increases theTESTO concentration [10] However AGT did not increasethe TESTO concentration in our study One possibility isthat the concentration of TESTO was already enhanced byabout 33-fold through stimulation with ACTH forskolinand angiotensin II
333 DDD The mechanism of action of DDD in adrenalsteroidogenesis was estimated by dose-dependent inhibitionof CYP11A1 HSD3B2 CYP21A2 and CYP11B1 (estimatedinhibitions at 25120583M were 870 869 769 and 849resp) (Figure 5(f)) DDD has been reported to inhibitCYP11A1 HSD3B2 CYP21A2 CYP11B1 and CYP11B2 [29 3132] Inhibition of DDD by CYP11B2 was not estimated in ourstudy However the concentration of ALDO in the culturemedium decreased to 3 of that in the normal stimulatedcondition Inhibition of DDD by CYP11B2 has been shownusing mitochondrial and microsomal fractions prepared bystandard centrifugation procedures from a bovine adrenal
cortex homogenate [32]The cause of the discrepancy regard-ing the inhibition ofDDDbyCYP11B2 could not be explainedby same effect in the case of AGT
334 SP Themechanism of action of SP in adrenal steroido-genesis was estimated by inhibition of HSD3B2 CYP21A2and HSD17B3 (estimated inhibitions at 10120583M were 702595 and 593 resp) (Figure 5(g)) SP has been reportedto inhibit CYP17H CYP17L CYP11B1 and CYP11B2 [6 33ndash35] The inhibitory effect of SP on the HSD3B2 enzyme wasa novel mechanism of action The main action of SP is asa mineralocorticoid receptor (MR) antagonist SP has alsobeen reported to exert some off-target effects by binding toandrogen glucocorticoid and progesterone receptors [56ndash58] SP has been shown to inhibit the production of ALDOand CORT from PREG induced by angiotensin II in H295Rcells but the specific MR antagonist eplerenone did not showthe inhibitory effects [59] Therefore HSD3B2 inhibition bySP is not mediated via MR and the action might be directinhibition of HSD3B2 enzymes or a part of known off-targeteffects mediated through other nuclear hormone receptorsRegarding CYP17H and CYP17L our results were consistent
10 Journal of Toxicology
CHOL
0
20
40
60
80
100C
once
ntra
tion
in m
ediu
m (120583
M)
10 20 30 40 50 60 70 800Time (hours)
(a)
PREG
DCORT
DCORTICO
CORTICO
CORT
Con
cent
ratio
n in
med
ium
(120583M
)
10
4
2
0
8
6
10 20 30 40 50 60 70 800Time (hours)
(b)
Con
cent
ratio
n in
med
ium
(120583M
)
0
020406
081
1214
0 10 20 30 40 50 60 70 80Time (hours)
HPREG
DIONE
PROGHPROG
E1
(c)
DHEAE2
ALDO
TESTO
0
005
01
015
02
025
Con
cent
ratio
n in
med
ium
(120583M
)
0 10 20 30Time (hours)
40 50 60 70 80
(d)
CHOL
0
5000
10000
15000
20000
25000
30000
Con
cent
ratio
n in
cell
(120583M
)
10 20 30 40 50 60 70 800Time (hours)
(e)
Con
cent
ratio
n in
cell
(120583M
)
0
50
100
150
200
250
0 10 20 30 40 50 60 70 80Time (hours)
PREG
DCORT
DCORTICO
CORTICOCORT
(f)
0 10 20 30 40 50 60 70 80Time (hours)
HPREG
DIONE
PROGHPROG
E1
Con
cent
ratio
n in
cell
(120583M
)
3025201510
35
05
(g)
005
115
2
253
35
Con
cent
ratio
n in
cell
(120583M
)
0 10 20 30 40 50 60 70 80Time (hours)
DHEA
E2ALDO
TESTO
(h)
Figure 4 Comparison of time-course profiles of cholesterol and adrenal steroids produced by NCI-H295R cells between experimentallymeasured and simulated data To intuitively confirm reconstruction of the measured experimental data in the developed simulation modelof NCI-H295R cells dynamics of cholesterol and adrenal steroids produced by NCI-H295R cells were plotted to overlay experimentaldata with the simulated results Graphs show the dynamics of medium concentrations of cholesterol (a) and adrenal steroids ((b)ndash(d))and intracellular concentrations of cholesterol (e) and adrenal steroids ((f)ndash(h)) Steroid hormones were categorized into three groups byconcentration levels Major steroids were PREG pregnenolone DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICOcorticosterone and CORT cortisol ((b) and (f)) Moderate steroids were HPREG 17120572-hydroxypregnenolone PROG progesterone HPROG17120572-hydroxyprogesterone DIONE androstenedione and E1 estrone ((c) and (g)) Minor steroids were DHEA dehydroepiandrosteroneALDO aldosterone TESTO testosterone and E2 17120573-estradiol ((d) and (h)) Experimental data are shown as symbols with dotted lines Alldata represent the mean plusmn SD (119873 = 4) Simulation data are shown as solid lines
Journal of Toxicology 11
0110
10
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
AN (120583M)
(a)
0110
FN (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(b)
00101
SM (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
100E minus 05
(c)
0110
TG (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B1
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(d)
0110
10100
AGT (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(e)
0110
1025
DDD (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(f)
1010
SP (120583M)
2750005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(g)
0110
10100
MP (120583M)
2110005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(h)
Figure 5 Continued
12 Journal of Toxicology
0110
10
KC (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(i)
0110
10
MC (120583M)
4430005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(j)
0110
10100
DZ (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(k)
Figure 5 Estimated mechanism of action of adrenal toxicants by using the mathematical model of adrenal steroidogenesis in NCI-H295Rcells Mechanisms of action of adrenal toxicity compounds were quantitatively estimated from experimental results of differential steroidprofiling using the mathematical model of adrenal steroidogenesis in NCI-H295R cells The drug action was defined as a scaling factorof enzymatic activity in the simulation model These scaling factors were optimized to fit experimental data by a hybrid optimizationmethod of the RCGA and nonlinear least squares Estimated drug actions by the exposure of vasculotoxic agents acrylonitrile AN (a)fumaronitrile FN (b) salinomycin SM (c) and thioguanine TG (d) and the steroidogenic inhibitors aminoglutethimide AGT (e) op1015840-DDD DDD (f) spironolactone SP (g) metyrapone MP (h) ketoconazole KC (i) miconazole MC (j) and daidzein DZ (k) are shownas a spider radar chart CYP11A1 P450 side chain cleavage enzyme CYP17H 17120572-hydroxylase of CYP17 CYP17L C
17ndash20 lyase of CYP17HSD3B2 3120573-hydroxysteroid dehydrogenase CYP21A2 21-hydroxylase CYP11B1 11120573-hydroxylase CYP11B2 18-hydroxylase and HSD17B317120573-hydroxysteroid dehydrogenase
with previous reports [33 34] 7120572-thiospironolactone whichis synthesized by deacetylation of SP inhibits CYP17H andCYP17L [34] The fact that there were no inhibitions ofCYP17HorCYP17L in our study suggests that SPmight not bedeacetylated to 7120572-thiospironolactone in NCI-H295R cellsRegarding CYP11B1 and CYP11B2 our results were unclearcompared with a previous study It has been shown that30 120583M SP inhibits CYP11B1 and CYP11B2 in human andbovine adrenal mitochondria [35]The cause of CYP11B1 andCYP11B2 inhibition by SP could not be determined in ourstudy which might be due to the lower maximum exposureconcentration of SP than that in the previous reportWe couldnot examine SP concentrations over 10 120583M because theseconcentrations were cytotoxic in NCI-H295R cells
335 MP The mechanism of action of MP in adrenalsteroidogenesis was estimated by dose-dependent inhibitionof CYP11B1 (estimated inhibitions at 1 10 and 100 120583M were571 827 and 982 resp) (Figure 5(h)) MP has beenreported to inhibit CYP11B1 as its major effect and CYP11A1and CYP11B2 as a weak effect [6 29 36ndash39] The results wereable to show that MP is a selective inhibitor of CYP11B1 inthe previous report However the estimated effect of MPat a high concentration 100 120583M as the maximum exposureconcentration was unclear According to the previous reportselectivity of MP for CYP11B1CYP11B2 is about five times[39] In addition 20 120583M MP has a slight inhibition effect onCYP11A1 in H295R cells [29]
336 KC The mechanism of action of KC in adrenalsteroidogenesis was estimated by inhibition of CYP11A1CYP17H CYP17L HSD3B2 CYP21A2 CYP11B1 and
CYP11B2 (estimated inhibitions at 10 120583Mwere 926 943518 830 882 974 and 798 resp) (Figure 5(i))KC has been reported to inhibit CYP11A1 CYP17H CYP17LHSD3B2 CYP21A2 and CYP11B1 [6 29 40ndash43] Our resultswere almost consistent with the previous reports KC inhibitsCYP11A1 CYP17H CYP21A2 and CYP11B1 in NCI-H295Rcells at 10 120583M [29] and CYP17H CYP17L CYP21A2 andCYP11B1 in human adrenal mitochondria and Leydig cellmicrosomes at 2ndash5120583M [60 61] However KC has shownonly weak inhibition of HSD3B2 and HSD17B3 in Leydigcells at the millimolar level [60 61] Regarding CYP11B2 andHSD17B3 we considered that these estimated inhibitions ofKC did have sufficient reliability in terms of quantitative pre-diction precision because ALDO and TESTO concentrationswere less than the lower limit of quantification at 10 120583MKC
337 MC Themechanism of action of MC in adrenal stero-idogenesis was estimated by inhibition of CYP17H CYP17LCYP11B1 and HSD17B3 (estimated inhibitions at 10120583Mwere691 530 764 and 571 resp) (Figure 5(j)) MC hasbeen reported to inhibit not only CYP17H and CYP17L butalsoCYP11A1 CYP21A2 andCYP11B1 [41 44 45]The resultsin the previous reports were able to estimate that MC isa CYP17 inhibitor However CYP11A1 inhibition by MCprobably instead of a reduction in StAR expression wasnot clearly detected in our study using NCI-H295R cellsbecause there were no decreases in the concentrations ofPREG and PROG in the culture medium Indirect inhibitionof CYP11A1 via the peripheral-type benzodiazepine receptorhas been reported in mouse adrenocortical Y-1 cells treatedwith MC in the absence of stimuli by measuring PREGproduction [44] On the other hand reductions of StAR
Journal of Toxicology 13
protein expression andor transport activity without affectingtotal steroid synthesis or CYP11A1 and HSD3B2 enzymeexpression or activities have been reported in (BU)
2cAMP-
stimulated MA-10 Leydig tumor cells treated with MC bymeasuring PROG production [45] Therefore the effect ofMC on the initial reaction in adrenal steroidogenesis fromcholesterol should be different according to the cell type andstimulation condition Inhibition of CYP21A2 and CYP11B1by MC has been reported as decreases in the consumptionof PROG and DCORTICO respectively [41] Inhibition ofCYP11B1 was estimated by the action of MC in this studybut that of CYP21A2 was not detected In the previousexperimental report inhibitory sites by MCmight have beenreflected by inhibition of CYP17H activity because CYP21A2activity was measured as a decrease in labeled PROG
338 DZ The mechanism of action of DZ in adrenalsteroidogenesis was estimated by inhibition of CYP11A1HSD3B2 CYP21A2 CYP11B1 and HSD17B3 (estimated inhi-bitions at 100 120583M were 586 941 965 872 and981 resp) (Figure 5(k)) DZ has been reported to inhibitHSD3B2 and CYP21A2 [46] The results of HSD3B2 andCYP21A2 were consistent with the previous report Howeverinhibitions have not been reported for CYP11A1 CYP11B1and HSD17B3These estimated effects of DZ on CYP11B1 andHSD17B3 were unclear because the concentrations of ALDOand TESTO were less than the lower limit of quantificationat 100 120583M In addition these enzymes act downstream of thestrong action points of DZ such as HSD3B2 and CYP21A2
34 The Simulations and the Systematic Model Analysis
341 Dynamic Sensitivity Analysis of Adrenal SteroidogenesisTo comprehensively understand the dynamics of adrenalsteroidogenesis dynamic sensitivities were calculated forsteroid concentrations secreted by NCI-H295R cells usingour constructed mathematical model of steroidogenesis Theresults of dynamic sensitivity analysis at 72 h of duration and6 h of interval time are presented as a heat-map in Figure 6
The top 10 parameters of the total area under the curveof dynamic sensitivity for cholesterol and the 14 steroids inculture medium were 119881HSD3B2
maxA 119881CYP11A1max 119881CYP21A2maxA 119896loc119891 119896loc119887
119870CYP11A1119898
119881CYP17HmaxA 119896Cholesterol Transport 119896acc119891 and 119881CYP21A2maxB in
order from the top Cholesterol uptake (119896Cholesterol Transport)StAR protein (119896loc
119891) and CYP11A1 (119881CYP11A1max ) which are
determining factors of the capacity for steroidogenesis pro-moted the production of mineralocorticoids (DCORTICOCORTICO and ALDO) and restrained the synthesis of glu-cocorticoids (DCORT and CORT) and sex steroids (DIONETESTO and E1) because of the accumulation of interme-diate molecules in steroidogenesis (PREG HPREG DHEAPROG and HPROG) only by self-activation The dynamicpatterns of the intermediate molecules in steroidogenesiswere mainly dependent on the activity of CYP17H andHSD3B2 with PREG as the substrate of these enzymes inwhich the dynamic sensitivities of 119881CYP17HmaxA for HPREG andHPROG and 119881HSD3B2
maxA for PROG HPROG and DCORTICO
reversed the direction of sensitivity at 49ndash66 h after stimu-lation The dynamic sensitivities of the maximum activitiesof HSD3B2 for PREG (119881HSD3B2
maxA ) and CYP21A2 for PROG(119881CYP21A2maxA ) were related to all steroids at 72 h Almost allmodel parameters had positive sensitivity for downstreamsteroids in the adrenal steroidogenic pathway and negativesensitivity for direct-binding steroids as substrates of thesteroidogenic enzyme The sensitivity of 119881max in all steroido-genic enzymes was relatively higher than 119870
119898for the same
steroid substrate
342 Simulation of the Metabolic Balance of AdrenalSteroidogenesis Pathway To clearly show the property of themetabolic shift between mineralocorticoid and glucocorti-coid biosynthesis we performed two-dimensional parameterscanning of the enzymatic activities of CYP17H and HSD3B2(Figure 7) NCI-H295R cells lost the ability to produce allsteroid hormones when enzymatic activities of CYP17H andHSD3B2 were changed by over 60 and 30 respectivelyActivation ofCYP17HandorHSD3B2 induced themetabolicshift that enhanced the glucocorticoid biosynthesis and devi-ated from the mineralocorticoid biosynthesis On the otherhand inhibition of CYP17H andor HSD3B2 induced themetabolic shift that enhanced the mineralocorticoid biosyn-thesis and deviated from the glucocorticoid biosynthesisMoreover the enzymatic activity of HSD3B2 regulated themetabolic balance of sex steroids and the precursors onadrenal steroidogenesis of NCI-H295R cells E1 TESTO andDIONE were produced by NCI-H295R cells when activatingthe enzymatic activity of HSD3B2 Conversely E2 andDHEAwere produced by NCI-H295R cells when suppressing theenzymatic activity of HSD3B2 The biosynthesis of down-stream steroids in adrenal steroidogenesis pathway such asmineralocorticoids and glucocorticoids was almost com-pletely terminated when the enzymatic activity of HSD3B2was decreased by over 80
4 Discussion
41 Importance of 3120573-HSDActivity in Adrenal SteroidogenesisOur systematic analysis using the mathematical model ofadrenal steroidogenesis in NCI-H295R cells revealed thatthe enzymatic activity of 3120573-HSD controls the dynamics ofadrenal steroidogenesis The activity of the StAR proteincontrols the net capacity of steroidogenesis in steroidogeniccells which is the transport of cholesterol from the outer toinner mitochondrial membranes Both the expression levelsof StAR protein and mRNA are rapidly elevated in responseto stimulation by tropic hormones such as ACTH [62 63]Another important factor in adrenal steroidogenesisis is thecholesterol side chain cleavage enzyme CYP11A1 the firstrate-limiting and hormonally regulated step in the synthesisof all steroid hormones which is conversion of cholesterolto pregnenolone in mitochondria [64] According to ourresults of global sensitivity analysis (Supplementary Table 1and Figure 6(d)) in addition to CYP11A1 and StAR proteins3120573-HSD was one of the key regulators in adrenal steroido-genesis of NCI-H295R cells And also this result suggeststhat a significant regulatory mechanism in steroidogenesis
14 Journal of Toxicology
60544830 4236 66 7218 246 120
Time (hours)
0
1
2
3C
once
ntra
tion
(a)
60544830 4236 66 7218 246 120
Time (hours)
minus05
0
05
minus1
1
Sens
itivi
ty
(b)
0
6
12
18
24
30
36
42
48
54
60
66
72
Sens
itivi
ty fo
r med
ium
conc
entr
atio
ns
of ch
oles
tero
l or s
tero
id h
orm
ones
Dur
atio
n tim
e (ho
urs)
Model parameters
+10
+05
00
minus05
minus10
(c)
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
Parameter number in the mathematical model of steroidogenesis in NCI-H295R cells
CHOL
PREG
HPREG
DHEA
PROG
HPROG
DIONE
DCORTICO
DCORT
CORTICO
CORT
ALDO
TESTO
E1
E2
Chol
este
rol o
r ste
roid
hor
mon
es
0
05
10
15
20
minus05
minus10
minus15
minus20
(d)
Figure 6 Heat-map of the global dynamic sensitivity analysis of adrenal steroid concentrations produced by NCI-H295R cells The globaldynamic sensitivity analysis is a powerful tool to comprehensively understand the dependencies of themodel parameters in themathematicalmodel of the biological complex system Dynamic sensitivities of model parameters in the mathematical model of adrenal steroidogenesisin NCI-H295R cells were calculated for all steroid concentrations in the culture medium every 6 h until 72 h after stimulation To clarify theview of heat-map of global dynamic sensitivity analysis imaginary data of the dynamics of steroid concentrations in the original model (blueline) and perturbed model for sensitivity analysis (red line) were prepared (a) Using this imaginary data the calculated dynamic sensitivities(b) and the visualized dynamic sensitivity as one block of the heat-map (c) were shown respectively By the same method that explainedusing imaginary data the large-scale data of the global dynamic sensitivity analysis on the mathematical model of adrenal steroidogenesisin NCI-H295R cells was comprehensively visualized as a big graph of heat-map (d) Parameter numbers in the horizontal axis are (1)119896Choresterol Transport (2) 119896CEH (3) 119896MTR
119891 (4) 119896MTR
119887 (5) 119896acc
119891 (6) 119896acc
119887 (7) 119896loc
119891 (8) 119896loc
119887 (9) 119896Oxysterol Synthesis (10)119870
CYP11A1119898
(11) 119881CYP11A1max (12)119870CYP17H119898A
(13) 119870CYP17H119898B (14) 119881CYP17HmaxA (15) 119881CYP17HmaxB (16) 119881CYP17HmaxB (17) 119870CYP17L
119898B (18) 119881CYP17LmaxA (19) 119881CYP17LmaxB (20) 119870HSD3B2119898A (21) 119870HSD3B2
119898B (22) 119870HSD3B2119898C (23)
119881HSD3B2maxA (24) 119881HSD3B2
maxB (25) 119881HSD3B2maxB (26) 119870CYP21A2
119898A (27) 119870CYP21A2119898B (28) 119881CYP21A2maxA (29) 119881CYP21A2maxB (30) 119870CYP11B1
119898A (31) 119870CYP11B1119898B (32) 119881CYP11B1maxA
(33) 119881CYP11B1maxB (34) 119896CYP11B2 (35) 119870HSD17B3119898A (36) 119870HSD17B3
119898B (37) 119881HSD17B3maxA (38) 119881HSD17B3
maxB (39) 119870CYP19A1119898A (40) 119870CYP19A1
119898B (41) 119881CYP19A1maxA and (42)119881
CYP19A1maxB PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-
hydroxyprogesteroneDIONE androstenedioneDCORTICO 11-deoxycorticosteroneDCORT 11-deoxycortisol CORTICO corticosteroneCORT cortisol ALDO aldosterone and TESTO testosterone
pathway is very reasonable StAR CYP11A1 and 3120573-HSD(isoforms 1 or 2 in humans) proteins generally respond to thesame hormones that stimulate steroid production throughcommon pathways such as cAMP signaling in adrenal glandsand testes [65 66] Moreover our data also support recentexperimental evidence from clinical and in vivo studiessuggesting that the enzymatic activity of 3120573-HSD plays
an important role in the regulation of mineralocorticoidsynthesis in adrenal steroidogenesis and contributes to hyper-tension caused by abnormal overproduction of aldosterone[67ndash70] Circadian clock-deficient Cry-null mice show salt-sensitive hypertension due to abnormally high synthesis ofaldosterone which is caused by constitutively high expressionof HSD3B6mRNA and protein in the adrenal cortex [67 68]
Journal of Toxicology 15
All steroid producing cells
Mineralocorticoid producing cells Glucocorticoid producing cells
Mineralocorticoid and glucocorticoid producing cells
Terminated steroidogenesis
Sex steroids and precursor sex steroidsDIONE TESTO E1DEHA E2
NCI-H295Rcells at controlcondition
Normalized activities of HSD3B2
Nor
mal
ized
activ
ities
of C
YP17
H
2
18
16
14
12
1
08
06
04
02
00 05 1 15 2
Figure 7 Metabolic categories of the steroidogenic cells determined by the balance of HSD3B2 and CYP17H activitiesThe two-dimensionalparameter scanning analysis by the perturbation of two focused parameters clarifies the interaction and the relationship between the twoparameters in the complex system Functional cellular categories of steroidogenic cells were defined by the levels of mineralocorticoid(ALDO) glucocorticoids (DCORTandCORT) and androgens (DHEAandDIONE) at 72 h after stimulation Enzymatic activities ofHSD3B2andCYP17were normalized by standard values of the simulationmodel inNCI-H295R cells Scanning ranges ofHSD3B2 andCYP17 activitieswere 0ndash200 each 10 Green regions are all 14 steroids produced by NCI-H295R cells Red blue and purple regions are mineralocorticoidproducing cells glucocorticoids-producing cells and both corticoid-producing cells respectively Yellow regions are steroidogenesis of NCI-H295R cells terminated upstream of the adrenal steroidogenesis pathway
Recent clinical observations have revealed predominantexpression of HSD3B2 mRNA and protein in tumor cellsof aldosterone-producing adenoma (APA) and HSD3B1mRNA significantly correlated with CYP11B2 mRNA levelsand plasma aldosterone concentrations in APA patients [6970] However the relationship is unclear and disputed ina small-scale clinical study indicating that genetic variationin HSD3B1 affects blood pressure and hypertension in APApatients [71] The results of our simulation study suggest that3120573-HSD protein (human genes are HSD3B1 and HSD3B2) isone of the determination factors for the dynamic propertyof adrenal steroidogenesis Our results support the clinicalevidence of Doi and colleagues [69] and we believe that theHSD3B1 enzyme has a promising potential as novel drugtarget for endocrine hypertension
42 Metabolic Shift of Adrenal Steroidogenesis and the Contri-butions of HSD3B2 and CYP17H The metabolic propertiesof adrenal steroidogenesis in NCI-H295R cells were revealedby dynamic sensitivity analysis using themathematicalmodel(Figure 6) Mineralocorticoids such as DCORTICO COR-TICO and ALDO and intermediate steroids upstream of
the adrenal steroidogenesis pathway such as PREG HPREGDHEA PROG and HPROG were accelerated by reactionsof cholesterol import (119896Cholesterol Transport) StAR protein (119896MTR
119891
and 119896loc119891) and CYP11A1 (119881CYP11A1max ) On the other hand glu-
cocorticoids such as DCORT and CORT and sex hormonessuch as DIONE TESTO and E1 were suppressed by thesemodel parametersTherefore enhancement of the net adrenalsteroidogenesis capacity which supplies PREG precursor tothe pathway causes a production shift from glucocorticoidsto mineralocorticoids by substrate inhibitions of CYP17HHSD3B2 and CYP21A2 caused by accumulation of initialintermediate steroids such as PREG and PROG Sensitivi-ties of CYP17H (119881CYP17HmaxA ) and HSD3B2 (119881HSD3B2
maxA ) for theproducts were dynamically changed and these parametersdetermined the metabolic balance of downstream steroidsin the adrenal steroidogenesis pathway According to theseresults of dynamic sensitivity analysis of StAR CYP11A1CYP17H and HSD3B2 we suggest that the enhancementof CYP17H and HSD3B2 activity during ACTH stimulationwas important to shift the steroidogenic output away fromALDO biosynthesis towards CORT biosynthesis as wellas adrenal androgen production This suggestion partially
16 Journal of Toxicology
supports a comparative animal study in which molecularand cellular variations in CYP17H activity dramatically affectacute cortisol production resulting in distinct physiologicaland behavioral responses [72]
Results of two-dimensional parameter scanning of theenzymatic activities of CYP17H and HSD3B2 quantitativelyshowed the detail of the metabolic relationship betweenmin-eralocorticoid and glucocorticoid biosynthesis (Figure 7)Particularly the results showed that the balance of theseenzymatic activities was very important for the typicalfunction of NCI-H295R cells namely the ability to produceall steroid hormones NCI-H295R cells lost this functionwhen enzymatic activities of CYP17H and HSD3B2 werechanged by over 60 and 30 respectively In addition theybecame mineralocorticoid (ALDO) secreting cells when theenzymatic activity of CYP17H or HSD3B2 was inhibited byover 50 or glucocorticoid (DCORT and CORT) secretingcells when these enzymes were activated by over 50 Inparticular this analysis also showed that HSD3B2 was a keyplayer in the adrenal steroidogenesis of NCI-H295R cellsbecause HSD3B2 inhibition by over 80 almost completelyinhibited the biosynthesis of downstream steroids The ratioof CYP17A1 to HSD3B2 mRNA expression levels has beenrelated to several endocrine diseases with a low level in APAs[73] and high level in cortisol-producing adenomas [74]Furthermore the expression levels or enzymatic activitiesof CYP17A1 and HSD3B1 have been related to androgenproduction in polycystic ovary syndrome [75 76] Theseclinical studies support our simulation results indicating thatthe balance of enzymatic activity of CYP17H and HSD3B2determines the shift in steroidogenic output to mineralocor-ticoids glucocorticoids or androgens
43 Methodologies of Quantitative Mechanistic Analysis forDrug Discovery According to our results obtained using themathematical model of steroidogenesis in NCI-H295R cellssuch as sensitivity analysis comprehensive analysis based onsystems biology is available to quantitatively estimate themechanism of action of steroidogenic disrupting compoundsfrom differential profiling of adrenal steroid hormonesbecause dynamic patterns of steroid hormones in adrenalsteroidogenesis pathway are highly complex Our proposedmethod of quantitative mechanistic analysis of steroidogenicinhibitors was able to predict known action sites in theadrenal steroidogenesis pathway at only one time point (72 hafter drug exposure) Moreover according to the resultsof sensitivity analysis (Figure 6) 119881max of all steroidogenicenzymes was more sensitive than the 119870
119898 because the
intracellular concentrations of steroid hormones were almostmaintained at sufficiently high levels compared with 119870
119898
values of steroidogenic enzymesThese results suggested thatestimation of the mechanism of action of drugs is moreeffective and detectable when using the influences of 119881maxas the searching parameters such as our proposed methodOur data showed that the proposed method based on asystems biology model is a very powerful tool for exploratoryscreening of steroidogenic disrupting compounds
RCGA as a solver of parameter estimation problemsin systems biology has been applied to biological network
identification of gene regulatory networks and metabolicpathways and optimization of biological processes usingexperimentally observed time-course data [77ndash82] In thisstudy RCGA was useful to estimate the mechanism ofaction of novel pharmaceutical drug candidates for adrenalsteroidogenesis as a new application of RCGA in systemsbiology We had two issues when applying RCGA to thequantitative mechanistic analysis of drug actions Theseissues were the vast calculation cost and multimodality ofquasi-optimum solutions in solving the optimization prob-lem because the mathematical model in systems biologyconsists of many equations and parameters A proposedoptimization strategy using RCGA based on REXJGG wasa highly stable and efficient calculation method for a betterquasi-optimum solution than the unimodal normal distri-bution crossover (UNDX)minimum generation gap (MGG)method that is well applied in the engineering field In addi-tion we expanded the RCGA optimization program basedon REXJGG to a hybrid method of GA and then applieda local search as recommended by Harada and Kobayashi[28 83] A final optimal solution was obtained with a goodconvergence property Because these problems are generalin systems biology studies we suggest that the proposedhybrid method based on REXJGG is a very useful tool forquantitative mechanistic analysis of novel pharmaceuticaldrugs not limited to steroidogenic disrupting compounds
5 Conclusions
Thenovelmathematicalmodel of adrenal steroidogenesis wasconstructed in this study including cholesterol transport anddistribution the C
21-steroid hormone pathway steroid trans-
port and cell proliferation which could reproduce adrenalsteroidogenesis in NCI-H295R cells According to the resultsof dynamic sensitivity analysis using the newmodel HSD3B2plays the most important role in the metabolic balance ofadrenal steroidogenesis in NCI-H295R cells Moreover toquantitatively estimate mechanisms of action of adrenal toxiccompounds we analyzed differential metabolic profiles of12 steroid hormones at 3 days after exposure to 11 adrenaltoxic compounds by using the new mathematical modeland a hybrid optimization method of the RCGA and a localsearch (nonlinear least squares) We could estimate whichsteroidogenic enzymes were affected by these compoundsusing the hybrid optimization method Vasculotoxic agentswere estimated to have no effect according to the resultsobtained by our method In terms of adrenal steroidogenicinhibitors the predicted action sites were approximatelymatched to the target enzymes as reported in the literatureThus our computer-aided method based on a systems biol-ogy approach may be useful to analyze the mechanism ofaction of endocrine-disrupting compounds and to assess thehuman adrenal toxicity of novel pharmaceutical drugs basedon steroid hormone profiling
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Journal of Toxicology 17
Authorsrsquo Contribution
Ryuta Saito Natsuko Terasaki and Naohisa Tsutsui designedthe study Naohisa Tsutsui and Naoya Masutomi supervisedthe experiments Natsuko Terasaki performed all in vitroexperiments and collected dataMakotoYamazaki performedall measurements of steroids Ryuta Saito conceived the insilico strategies analyzed data developed all programs andthe mathematical model and performed all simulation anal-yses with Masahiro Okamoto Ryuta Saito Natsuko TerasakiMakoto Yamazaki NaoyaMasutomi andMasahiroOkamotowrote the paper
Acknowledgments
The authors would like to thank Dr Michael S Breenand Dr Miyuki Breen for helpful discussions concerningprevious mathematical models of adrenal steroidogenesisThis work was partially supported by the Grant-in-Aid forScientific Research on Innovative Areas ldquoSynthetic Biologyrdquo(No 23119001 (MO)) from Ministry of Education CultureSports Science and Technology in Japan
References
[1] P W Harvey D J Everett and C J Springall ldquoAdrenal toxi-cology a strategy for assessment of functional toxicity to theadrenal cortex and steroidogenesisrdquo Journal of Applied Toxi-cology vol 27 no 2 pp 103ndash115 2007
[2] W E Ribelin ldquoThe effects of drugs and chemicals upon thestructure of the adrenal glandrdquo Fundamental and Applied Toxi-cology vol 4 no 1 pp 105ndash119 1984
[3] H D Colby and P A Longhurst ldquoToxicology of the adrenalglandrdquo in Endocrine Toxicology C K Atterwill and J D FlackEds pp 243ndash281 CambridgeUniversity Press Cambridge UK1992
[4] H D Colby ldquoIn vitro assessment of adrenocortical toxicityrdquoJournal of Pharmacological and Toxicological Methods vol 32no 1 pp 1ndash6 1994
[5] T J Rosol J T Yarrington J Latendresse and C C CapenldquoAdrenal gland structure function and mechanisms of toxic-ityrdquo Toxicologic Pathology vol 29 no 1 pp 41ndash48 2001
[6] P W Harvey and D J Everett ldquoThe adrenal cortex and ster-oidogenesis as cellular and molecular targets for toxicity crit-ical omissions from regulatory endocrine disrupter screeningstrategies for human healthrdquo Journal of Applied Toxicology vol23 no 2 pp 81ndash87 2003
[7] A F Gazdar H K Oie C H Shackleton et al ldquoEstablishmentand characterization of a human adrenocortical carcinoma cellline that expresses multiple pathways of steroid biosynthesisrdquoCancer Research vol 50 no 17 pp 5488ndash5496 1990
[8] W E Rainey I M Bird C Sawetawan et al ldquoRegulation ofhuman adrenal carcinoma cell (NCI-H295) production of C19steroidsrdquo The Journal of Clinical Endocrinology amp Metabolismvol 77 no 3 pp 731ndash737 1993
[9] B Staels D W Hum andW L Miller ldquoRegulation of steroido-genesis in NCI-H295 cells a cellular model of the human fetaladrenalrdquo Molecular Endocrinology vol 7 no 3 pp 423ndash4331993
[10] M Hecker J L Newsted M B Murphy et al ldquoHuman adreno-carcinoma (H295R) cells for rapid in vitro determination of
effects on steroidogenesis hormone productionrdquo Toxicologyand Applied Pharmacology vol 217 no 1 pp 114ndash124 2006
[11] A Oskarsson E Ulleras K E Plant J P Hinson and P SGoldfarb ldquoSteroidogenic gene expression in H295R cells andthe human adrenal gland adrenotoxic effects of lindane invitrordquo Journal of Applied Toxicology vol 26 no 6 pp 484ndash4922006
[12] UMuller-Vieira M Angotti and RW Hartman ldquoThe adreno-cortical tumor cell line NCI-H295R as an in vitro screeningsystem for the evaluation of CYP11B2 (aldosterone synthase)and CYP11B1 (steroid-11120573-hydroxylase) inhibitorsrdquoThe Journalof Steroid Biochemistry and Molecular Biology vol 96 no 3-4pp 259ndash270 2005
[13] K Imagawa S Okayama M Takaoka et al ldquoInhibitory effectof efonidipine on aldosterone synthesis and secretion in humanadrenocarcinoma (H295R) cellsrdquo Journal of CardiovascularPharmacology vol 47 no 1 pp 133ndash138 2006
[14] M Voets UMuller-Vieira S Marchais-Oberwinkler and RWHartmann ldquoSynthesis of amidinohydrazones and evaluation oftheir inhibitory effect towards aldosterone synthase (CYP11B2)and the formation of selected steroidsrdquo Archiv der Pharmazievol 337 no 7 pp 411ndash416 2004
[15] F K Nielsen C H Hansen J A Fey et al ldquoH295R cells as amodel for steroidogenic disruption a broader perspective usingsimultaneous chemical analysis of 7 key steroid hormonesrdquoToxicology in Vitro vol 26 no 2 pp 343ndash350 2012
[16] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cortisolproduction and on activities of steroidogenic enzymes inhuman adrenocortical H295R cellsrdquo Journal of Steroid Biochem-istry and Molecular Biology vol 80 no 3 pp 355ndash363 2002
[17] R F Canton J T Sanderson S Nijmeijer A Bergman R JLetcher and M van den Berg ldquoIn vitro effects of brominatedflame retardants and metabolites on CYP17 catalytic activity anovel mechanism of actionrdquo Toxicology and Applied Pharma-cology vol 216 no 2 pp 274ndash281 2006
[18] K Hilscherova P D Jones T Gracia et al ldquoAssessment ofthe effects of chemicals on the expression of ten steroidogenicgenes in the H295R cell line using real-time PCRrdquo ToxicologicalSciences vol 81 no 1 pp 78ndash89 2004
[19] T Gracia K Hilscherova P D Jones et al ldquoThe H295R systemfor evaluation of endocrine-disrupting effectsrdquo Ecotoxicologyand Environmental Safety vol 65 no 3 pp 293ndash305 2006
[20] X Zhang J L NewstedM Hecker E B Higley P D Jones andJ P Giesy ldquoClassification of chemicals based on concentration-dependent toxicological data using ToxClustrdquo EnvironmentalScience and Technology vol 43 no 10 pp 3926ndash3932 2009
[21] M S Breen M Breen N Terasaki M Yamazaki and R BConolly ldquoComputational model of steroidogenesis in humanH295R cells to predict biochemical response to endocrine-active chemicalsmodel development formetyraponerdquoEnviron-mental Health Perspectives vol 118 no 2 pp 265ndash272 2010
[22] M Breen M S Breen N Terasaki M Yamazaki A LLloyd and R B Conolly ldquoMechanistic computational model ofsteroidogenesis in H295R cells role of oxysterols and cell pro-liferation to improve predictability of biochemical response toendocrine active chemical-metyraponerdquo Toxicological Sciencesvol 123 no 1 pp 80ndash93 2011
[23] F B Kraemer ldquoAdrenal cholesterol utilizationrdquo Molecular andCellular Endocrinology vol 265-266 pp 42ndash45 2007
[24] W L Miller ldquoSteroidogenic acute regulatory protein (StAR)a novel mitochondrial cholesterol transporterrdquo Biochimica et
18 Journal of Toxicology
Biophysica Acta (BBA)mdashMolecular and Cell Biology of Lipidsvol 1771 no 6 pp 663ndash676 2007
[25] S Szabo and I T Lippe ldquoAdrenal gland chemically inducedstructural and functional changes in the cortexrdquo ToxicologicPathology vol 17 no 2 pp 317ndash329 1989
[26] C Chen-Pan I-J Pan Y Yamamoto T Sakogawa J Yamadaand YHayashi ldquoPrompt recovery of damaged adrenalmedullaeinduced by salinomycinrdquoToxicologic Pathology vol 27 no 5 pp563ndash572 1999
[27] R J Santen and R I Misbin ldquoAminoglutethimide review ofpharmacology and clinical userdquo Pharmacotherapy vol 1 no 2pp 95ndash120 1981
[28] M J Carella N V Dimitrov V V Gossain L Srivastava andD R Rovner ldquoAdrenal effects of low-dose aminoglutethimidewhen used alone in postmenopausal women with advancedbreast cancerrdquoMetabolism vol 43 no 6 pp 723ndash727 1994
[29] M K Johansson J T Sanderson and B-O Lund ldquoEffects of3-MeSO2-DDE and some CYP inhibitors on glucocorticoidsteroidogenesis in the H295R human adrenocortical carcinomacell linerdquo Toxicology in Vitro vol 16 no 2 pp 113ndash121 2002
[30] Y Touitou A Bogdan J C Legrand and P Desgrez ldquoAmino-glutethimide and glutethimide effects on 18-hydroxycorti-costerone biosynthesis by human and sheep adrenals in vitrordquoActa Endocrinologica vol 80 no 3 pp 517ndash526 1975
[31] M M Hart and J A Straw ldquoStudies on the site of action ofop1015840-DDD in the dog adrenal cortex 1 Inhibition of ACTH-mediated pregnenolone synthesisrdquo Steroids vol 17 no 1ndash5 pp559ndash574 1971
[32] M Ojima M Saitoh N Itoh Y Kusano S Fukuchi and HNaganuma ldquoThe effects of op1015840-DDD on adrenal steroidogen-esis and hepatic steroid metabolismrdquo Nippon Naibunpi Gakkaizasshi vol 61 no 3 pp 168ndash178 1985
[33] L F Canosa and N R Ceballos ldquoEffects of different steroid-biosynthesis inhibitors on the testicular steroidogenesis of thetoad Bufo arenarumrdquo Journal of Comparative Physiology B vol171 no 6 pp 519ndash526 2001
[34] D C Kossor S Kominami S Takemori and H D ColbyldquoRole of the steroid 17 alpha-hydroxylase in spironolactone-mediated destruction of adrenal cytochrome P-450rdquoMolecularPharmacology vol 40 no 2 pp 321ndash325 1991
[35] S C Cheng K Suzuki W Sadee and B W Harding ldquoEffectsof spironolactone canrenone and canrenoate K on cytochromeP450 and 11120573 and 18 hydroxylation in bovine and humanadrenal corticalmitochondriardquo Endocrinology vol 99 no 4 pp1097ndash1106 1976
[36] K Yanagibashi M Haniu J E Shively W H Shen and P HallldquoThe synthesis of aldosterone by the adrenal cortex Two zones(fasciculata and glomerulosa) possess one enzyme for 11120573- 18-hydroxylation and aldehyde synthesisrdquoThe Journal of BiologicalChemistry vol 261 no 8 pp 3556ndash3562 1986
[37] K Morishita H Okumura N Ito and N Takahashi ldquoPrimaryculture system of adrenocortical cells from dogs to evaluatedirect effects of chemicals on steroidogenesisrdquo Toxicology vol165 no 2-3 pp 171ndash178 2001
[38] S W J Lamberts E G Bons H A Bruining and F H de JongldquoDifferential effects of the imidazole derivatives etomidateketoconazole and miconazole and of metyrapone on the secre-tion of cortisol and its precursors by human adrenocorticalcellsrdquo Journal of Pharmacology and Experimental Therapeuticsvol 240 no 1 pp 259ndash264 1987
[39] L Yin S Lucas F Maurer U Kazmaier Q Hu and RW Hartmann ldquoNovel imidazol-1-ylmethyl substituted 1256-tetrahydropyrrolo[321-ij]quinolin-4-ones as potent and selec-tive CYP11B1 inhibitors for the treatment of Cushingrsquos syn-dromerdquo Journal ofMedicinal Chemistry vol 55 no 14 pp 6629ndash6633 2012
[40] M DiMattina N Maronian H Ashby D L Loriaux andB D Albertson ldquoKetoconazole inhibits multiple steroidogenicenzymes involved in androgen biosynthesis in the humanovaryrdquo Fertility and Sterility vol 49 no 1 pp 62ndash65 1988
[41] M Ayub and M J Levell ldquoInhibition of human adrenal steroi-dogenic enzymes in vitro by imidazole drugs including keto-conazolerdquo Journal of Steroid Biochemistry vol 32 no 4 pp 515ndash524 1989
[42] D Engelhardt M M Weber T Miksch F Abedinpour andC Jaspers ldquoThe influence of ketoconazole on human adrenalsteroidogenesis incubation studies with tissue slicesrdquo ClinicalEndocrinology vol 35 no 2 pp 163ndash168 1991
[43] S Bhasin S Sikka T Fielder et al ldquoHormonal effects of keto-conazole in vivo in the male rat mechanism of actionrdquo Endo-crinology vol 118 no 3 pp 1229ndash1232 1986
[44] D M Zisterer and D C Williams ldquoCalmidazolium and otherimidazole compounds affect steroidogenesis in Y1 cells lack ofinvolvement of the peripheral-type benzodiazepine receptorrdquoJournal of Steroid Biochemistry and Molecular Biology vol 60no 3-4 pp 189ndash195 1997
[45] L P Walsh C N Kuratko and D M Stocco ldquoEconazoleand miconazole inhibit steroidogenesis and disrupt steroido-genic acute regulatory (StAR) protein expression post-trans-criptionallyrdquo Journal of Steroid Biochemistry and Molecular Bio-logy vol 75 no 4-5 pp 229ndash236 2000
[46] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cor-tisol production and on activities of steroidogenic enzymesin human adrenocortical H295R cellsrdquo The Journal of SteroidBiochemistry and Molecular Biology vol 80 no 3 pp 355ndash3632002
[47] D P Dempsher D S Gann and R D Phair ldquoA mechanisticmodel of ACTH-stimulated cortisol secretionrdquo American Jour-nal of Physiology vol 246 no 4 part 2 pp R587ndashR596 1984
[48] N Sarai S Matsuoka and A Noma ldquosimBio a Java package forthe development of detailed cell modelsrdquo Progress in Biophysicsand Molecular Biology vol 90 no 1ndash3 pp 360ndash377 2006
[49] K Levenberg ldquoA method for the solution of certain non-linearproblems in least squaresrdquo Quarterly of Applied Mathematicsvol 2 pp 164ndash168 1944
[50] D W Marquardt ldquoAn algorithm for least-squares estimation ofnonlinear parametersrdquo SIAM Journal on Applied Mathematicsvol 11 no 2 pp 431ndash441 1963
[51] P E Gill and W Murray ldquoAlgorithms for the solution of thenonlinear least-squares problemrdquo SIAM Journal on NumericalAnalysis vol 15 no 5 pp 977ndash992 1978
[52] Y Akimoto R Hasada J Sakuma I Ono and S KobayashildquoGeneration alternation model for real-coded GA using multi-parent proposal and evaluation of just generation gap (JGG)rdquoin Proceedings of the 19th SICE Symposium on DecentralizedAutonomous Systems pp 341ndash346 Tokyo Japan January 2007
[53] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoProposal and evaluation of adaptive real-coded crossoverAREXrdquo Transactions of the Japanese Society for Artificial Intel-ligence vol 24 no 6 pp 446ndash458 2009
Journal of Toxicology 19
[54] S Kobayashi ldquoThe frontiers of real-coded genetic algorithmsrdquoJournal of Japanese Society for Artificial Intelligence vol 24 no1 pp 147ndash162 2009
[55] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoAnalysis of the behavior of MGG and JGG as a selection modelfor real-coded genetic algorithmsrdquo Transactions of the JapaneseSociety for Artificial Intelligence vol 25 no 2 pp 281ndash289 2010
[56] A Struthers H Krum and G H Williams ldquoA comparisonof the aldosterone-blocking agents eplerenone and spironolac-tonerdquo Clinical Cardiology vol 31 no 4 pp 153ndash158 2008
[57] K Yamasaki M Sawaki S Noda et al ldquoComparison of theHershberger assay and androgen receptor binding assay oftwelve chemicalsrdquo Toxicology vol 195 no 2-3 pp 177ndash1862004
[58] C Eil and S K Edelson ldquoThe use of human skin fibroblasts toobtain potency estimates of drug binding to androgen recept-orsrdquo The Journal of Clinical Endocrinology amp Metabolism vol59 no 1 pp 51ndash55 1984
[59] P Ye T Yamashita D M Pollock H Sasano and W ERainey ldquoContrasting effects of eplerenone and spironolac-tone on adrenal cell steroidogenesisrdquo Hormone and MetabolicResearch vol 41 no 1 pp 35ndash39 2009
[60] B D Albertson K L Frederick N C Maronian et al ldquoTheeffect of ketoconazole on steroidogenesis I Leydig cell enzymeactivity in vitrordquo Research Communications in Chemical Patho-logy and Pharmacology vol 61 no 1 pp 17ndash26 1988
[61] B D Albertson N C Maronian K L Frederick et al ldquoTheeffect of ketoconazole on steroidogenesis II Adrenocorticalenzyme activity in vitrordquoResearch Communications in ChemicalPathology and Pharmacology vol 61 no 1 pp 27ndash34 1988
[62] DM Stocco andB J Clark ldquoRegulation of the acute productionof steroids in steroidogenic cellsrdquo Endocrine Reviews vol 17 no3 pp 221ndash244 1996
[63] L K Christenson and J F Strauss III ldquoSteroidogenic acute regu-latory protein (StAR) and the intramitochondrial translocationof cholesterolrdquo Biochimica et Biophysica Acta (BBA)mdashMolecularand Cell Biology of Lipids vol 1529 no 1ndash3 pp 175ndash187 2000
[64] E R Simpson ldquoCholesterol side-chain cleavage cytochromeP450 and the control of steroidogenesisrdquo Molecular and Cel-lular Endocrinology vol 13 no 3 pp 213ndash227 1979
[65] H A LaVoie and S R King ldquoTranscriptional regulation of ster-oidogenic genes STARD1 CYP11A1 and HSD3Brdquo ExperimentalBiology and Medicine vol 234 no 8 pp 880ndash907 2009
[66] F Raucci A DrsquoAniello and M M Di Fiore ldquoStimulation ofandrogen production by d-aspartate through the enhancementof StAR P450scc and 3120573-HSD mRNA levels in vivo rat testisand in culture of immature rat Leydig cellsrdquo Steroids vol 84pp 103ndash110 2014
[67] M Doi Y Takahashi R Komatsu et al ldquoSalt-sensitive hyper-tension in circadian clock-deficient Cry-null mice involves dys-regulated adrenal Hsd3b6rdquo Nature Medicine vol 16 no 1 pp67ndash74 2010
[68] H Okamura M Doi Y Yamaguchi and J-M Fustin ldquoHyper-tension due to loss of clock novel insight from the molecu-lar analysis of Cry1Cry2-deleted micerdquo Current HypertensionReports vol 13 no 2 pp 103ndash108 2011
[69] M Doi F Satoh T Maekawa et al ldquoIsoform-specific mono-clonal antibodies against 3120573-hydroxysteroid dehydrogenaseisomerase family provide markers for subclassification ofhuman primary aldosteronismrdquo The Journal of Clinical Endo-crinology amp Metabolism vol 99 no 2 pp E257ndashE262 2014
[70] S Konosu-Fukaya Y Nakamura F Satoh et al ldquo3120573-hydroxy-steroid dehydrogenase isoforms in human aldosterone-produc-ing adenomardquo Molecular and Cellular Endocrinology vol 408pp 205ndash212 2015
[71] G C Verwoert J Hofland N Amin et al ldquoExpression and genevariation studies deny association of humanHSD3B1 gene withaldosterone production or blood pressurerdquoAmerican Journal ofHypertension vol 28 no 1 pp 113ndash120 2015
[72] D Hough K Storbeck SW P Cloete A C Swart and P SwartldquoRelative contribution of P450c17 towards the acute cortisolresponse lessons from sheep and goatsrdquoMolecular and CellularEndocrinology vol 408 pp 107ndash113 2015
[73] I Sakuma S Suematsu Y Matsuzawa et al ldquoCharacterizationof steroidogenic enzyme expression in aldosterone-producingadenoma a comparison with various human adrenal tumorsrdquoEndocrine Journal vol 60 no 3 pp 329ndash336 2013
[74] A Tong A Jia S Yan Y Zhang Y Xie and G Liu ldquoEctopiccortisol-producing adrenocortical adenoma in the renal hilumhistopathological features and steroidogenic enzyme profilerdquoInternational Journal of Clinical and Experimental Pathologyvol 7 no 7 pp 4415ndash4421 2014
[75] A Hirsch D Hahn P Kempna et al ldquoMetformin inhibitshuman androgen production by regulating steroidogenicenzymes HSD3B2 and CYP17A1 and complex I activity of therespiratory chainrdquo Endocrinology vol 153 no 9 pp 4354ndash43662012
[76] B Xu L Gao Y Cui et al ldquoSET protein up-regulated test-osterone production in the cultured preantral folliclesrdquo Repro-ductive Biology and Endocrinology vol 11 article 9 2013
[77] S Kikuchi D Tominaga M Arita M Takahashi and MTomita ldquoDynamic modeling of genetic networks using geneticalgorithm and S-systemrdquo Bioinformatics vol 19 no 5 pp 643ndash650 2003
[78] N Shikata YMaki M Nakatsui et al ldquoDetermining importantregulatory relations of amino acids from dynamic networkanalysis of plasma amino acidsrdquo Amino Acids vol 38 no 1 pp179ndash187 2010
[79] M Nakatsui T Ueda Y Maki I Ono and M OkamotoldquoMethod for inferring and extracting reliable genetic interac-tions from time-series profile of gene expressionrdquoMathematicalBiosciences vol 215 no 1 pp 105ndash114 2008
[80] A Komori Y Maki M Nakatsui I Ono and M OkamotoldquoEfficient numerical optimization algorithm based on new real-coded genetic algorithm AREX + JGG and application to theinverse problem in systems biologyrdquo Applied Mathematics vol3 no 10 pp 1463ndash1470 2012
[81] J A Roubos G van Straten and A J B van Boxtel ldquoAn evolu-tionary strategy for fed-batch bioreactor optimization conceptsand performancerdquo Journal of Biotechnology vol 67 no 2-3 pp173ndash187 1999
[82] D Sarkar and J M Modak ldquoGenetic algorithms with filters foroptimal control problems in fed-batch bioreactorsrdquo Bioprocessand Biosystems Engineering vol 26 no 5 pp 295ndash306 2004
[83] K Harada K Ikeda J Sakuma I Ono and S KobayashildquoHybridization of genetic algorithm with local search in mul-tiobjective function optimization recommendation of GA thenLSrdquoTransactions of the Japanese Society for Artificial Intelligencevol 21 no 6 pp 482ndash492 2006
Submit your manuscripts athttpwwwhindawicom
PainResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
ToxinsJournal of
VaccinesJournal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AntibioticsInternational Journal of
ToxicologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
StrokeResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Drug DeliveryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Pharmacological Sciences
Tropical MedicineJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AddictionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Emergency Medicine InternationalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Autoimmune Diseases
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anesthesiology Research and Practice
ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Pharmaceutics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
MEDIATORSINFLAMMATION
of
8 Journal of Toxicology
0001
001
01
1
10
PREG
HPR
EG
DH
EA
PRO
G
HPR
OG
AN
DRO
DCO
RTIC
O
DCO
RT
CORT
ICO
CORT
ALD
O
TEST
O E1 E2C
once
ntra
tion
in m
ediu
m (120583
M)
lowast
lowast
lowast
lowast
lowast
lowast
lowast lowast lowastlowast
lowast
lowast
lowast
lowast
(a)
0 8 24 48 72Time (hours)
0
40
80
120
160
200
240
Num
ber o
f mol
ecul
es (n
mol
)
(b)
0 8 24 48 72Time (hours)
0
40
80
120
160
200
240
Num
ber o
f mol
ecul
es (n
mol
)
(c)
Figure 2 Experimental data of metabolic profiling of adrenal steroid hormones and the mass balance Concentrations of steroid hormonessecreted from NCI-H295R cells in the culture medium at 72 h after stimulation were compared with the untreated condition and thestimulated condition by 50 nM ACTH 20 120583M forskolin and 100 nM angiotensin II (a) Net molecular amounts including cholesterol andsteroid hormones in the culture medium and intracellular space are plotted at five time points (0 8 24 48 and 72 h after stimulation)under the untreated condition (b) and the stimulated condition (c) In the bar graphs dark and light bars indicate the amount of cholesteroland adrenal steroids respectively All data are shown as the mean plusmn SD (119873 = 4) lowast119901 values corrected by the familywise error ratelt001 PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-hydroxyprogesteroneDIONE androstenedioneDCORTICO 11-deoxycorticosteroneDCORT 11-deoxycortisol CORTICO corticosteroneCORT cortisol ALDO aldosterone and TESTO testosterone
hormones at 72 h after drug exposure using themathematicalmodel of adrenal steroidogenesis in NCI-H295R cells Thereproducibility of the estimated results was confirmed byperforming the test twice The estimated effects of 11 adrenaltoxic compounds on eight steroidogenic enzymes are shownin Figure 5 The adrenal toxic compounds without steroido-genic inhibition such as vasculotoxic agents (AN SM TGand FN) were not estimated for the target steroidogenicenzymes under noncytotoxic conditions Every fitness valueswere under 005 of NLSD values used as the fitting objectivefunction (Figures 5(a)ndash5(d)) Other steroidogenic inhibitors(AGT DDD SPMP KCMC andDZ) are described in detailbelow
332 AGT The mechanism of action of AGT in adrenalsteroidogenesis was estimated by inhibition of CYP11A1HSD3B2 CYP21A2 and CYP11B1 at 100 120583M (estimatedinhibitions were 770 780 811 and 598 resp)(Figure 5(e)) AGT has been reported to inhibit CYP11A1
CYP21A2 CYP11B1 and CYP11B2 [6 27ndash30] Our resultswere mostly consistent with the previous reports In particu-lar CYP11A1 appeared to be inhibited strongly byAGT In ourstudy HSD3B2 inhibition of AGT was shown by mechanisticanalysis based on systems biology approaches as a novelmechanism of action of AGT Inhibition of AGT by CYP11B2was not estimated in our study However the concentrationof ALDO in the culture medium decreased to 38 ofthe normal stimulated condition Inhibition of AGT byCYP11B2 has been shown using sheep adrenal homogenatesas well as a human adrenal homogenate from a patientwith Cushingrsquos syndrome [30]The activity of 18-hydroxylaseinduced by CYP11B2 was determined as the conversion ofcorticosterone to 18-hydroxycorticosterone in the previousstudy The cause of the discrepancy regarding the effect ofAGT on CYP11B2 was suggested to be substrate inhibitionbecause the intracellular concentration of CORTICO wasincreased by over 10 times of that in the culture mediumto reach 50 120583M Another possibility was poor quantitative
Journal of Toxicology 9
log2
ratio
4
3
2
1
0
minus1
minus2
minus3
minus4
ALDO
CORTICO
CORT
TESTO
DIONE
HPROG
DCORTICO
DCORT
PROG
PREG
HPREG
DHEA
DZ(
100)
SP(1
0)SP
(1)
DZ(
10)
DD
D(0
1)
AN
(01
)A
N(1
)A
N(1
0)SM
(00
1)TG
(00
1)M
P(0
1)FN
(01
)TG
(1)
AGT(
1)AG
T(0
1)SM
(00
0001
)D
DD
(1)
SM(0
1)
AGT(
10)
MP(
1)FN
(1)
MC(
1)D
Z(1)
KC(0
1)
MC(
01)
DZ(
01)
KC( 1
)D
DD
(10)
MP(
10)
MC(
10)
MP(
100)
AGT(
100)
DD
D(2
5)KC
(10)
Figure 3 Hierarchical cluster analysis of differential metabolic profiles of 12 steroid hormones by exposure to adrenal toxicity compoundsAdrenal toxicants were classified by using the differential metabolic profiling of 12 steroid hormones Concentrations of 12 adrenalsteroids secreted from NCI-H295R cells were drastically changed by the exposure of adrenal steroidogenesis inhibitors The 12 adrenalsteroid hormones were quantitatively measured by LC-MSMS simultaneously Four adrenal vasculotoxic compounds acrylonitrile ANfumaronitrile FN salinomycin SM and thioguanine TG were used as negative control compounds for adrenal steroidogenesis inhibitorsSeven adrenal steroidogenesis inhibitors aminoglutethimide AGT op1015840-DDD DDD spironolactone SP metyrapone MP ketoconazoleKC miconazole MC and daidzein DZ showed a characteristic steroid profile each and were classified as each independent singletonat the maximum exposure condition Exposure concentrations of adrenal toxic compounds were described in brackets of the samplename and the units were prepared as 120583M A blue cluster was classified as a group of nonchange samples including negative controlcompounds and low exposure conditions of adrenal steroidogenesis inhibitors PREG pregnenolone HPREG 17120572-hydroxypregnenoloneDHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-hydroxyprogesterone DIONE androstenedione DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICO corticosterone CORT cortisol ALDO aldosterone and TESTO testosterone
reliability of the experimental data because the ALDOconcentration was under the lower limit of quantification at100 120583MAGT Hecker and colleagues reported that 3 120583MAGTdecreases PREG and PROG concentrations and increases theTESTO concentration [10] However AGT did not increasethe TESTO concentration in our study One possibility isthat the concentration of TESTO was already enhanced byabout 33-fold through stimulation with ACTH forskolinand angiotensin II
333 DDD The mechanism of action of DDD in adrenalsteroidogenesis was estimated by dose-dependent inhibitionof CYP11A1 HSD3B2 CYP21A2 and CYP11B1 (estimatedinhibitions at 25120583M were 870 869 769 and 849resp) (Figure 5(f)) DDD has been reported to inhibitCYP11A1 HSD3B2 CYP21A2 CYP11B1 and CYP11B2 [29 3132] Inhibition of DDD by CYP11B2 was not estimated in ourstudy However the concentration of ALDO in the culturemedium decreased to 3 of that in the normal stimulatedcondition Inhibition of DDD by CYP11B2 has been shownusing mitochondrial and microsomal fractions prepared bystandard centrifugation procedures from a bovine adrenal
cortex homogenate [32]The cause of the discrepancy regard-ing the inhibition ofDDDbyCYP11B2 could not be explainedby same effect in the case of AGT
334 SP Themechanism of action of SP in adrenal steroido-genesis was estimated by inhibition of HSD3B2 CYP21A2and HSD17B3 (estimated inhibitions at 10120583M were 702595 and 593 resp) (Figure 5(g)) SP has been reportedto inhibit CYP17H CYP17L CYP11B1 and CYP11B2 [6 33ndash35] The inhibitory effect of SP on the HSD3B2 enzyme wasa novel mechanism of action The main action of SP is asa mineralocorticoid receptor (MR) antagonist SP has alsobeen reported to exert some off-target effects by binding toandrogen glucocorticoid and progesterone receptors [56ndash58] SP has been shown to inhibit the production of ALDOand CORT from PREG induced by angiotensin II in H295Rcells but the specific MR antagonist eplerenone did not showthe inhibitory effects [59] Therefore HSD3B2 inhibition bySP is not mediated via MR and the action might be directinhibition of HSD3B2 enzymes or a part of known off-targeteffects mediated through other nuclear hormone receptorsRegarding CYP17H and CYP17L our results were consistent
10 Journal of Toxicology
CHOL
0
20
40
60
80
100C
once
ntra
tion
in m
ediu
m (120583
M)
10 20 30 40 50 60 70 800Time (hours)
(a)
PREG
DCORT
DCORTICO
CORTICO
CORT
Con
cent
ratio
n in
med
ium
(120583M
)
10
4
2
0
8
6
10 20 30 40 50 60 70 800Time (hours)
(b)
Con
cent
ratio
n in
med
ium
(120583M
)
0
020406
081
1214
0 10 20 30 40 50 60 70 80Time (hours)
HPREG
DIONE
PROGHPROG
E1
(c)
DHEAE2
ALDO
TESTO
0
005
01
015
02
025
Con
cent
ratio
n in
med
ium
(120583M
)
0 10 20 30Time (hours)
40 50 60 70 80
(d)
CHOL
0
5000
10000
15000
20000
25000
30000
Con
cent
ratio
n in
cell
(120583M
)
10 20 30 40 50 60 70 800Time (hours)
(e)
Con
cent
ratio
n in
cell
(120583M
)
0
50
100
150
200
250
0 10 20 30 40 50 60 70 80Time (hours)
PREG
DCORT
DCORTICO
CORTICOCORT
(f)
0 10 20 30 40 50 60 70 80Time (hours)
HPREG
DIONE
PROGHPROG
E1
Con
cent
ratio
n in
cell
(120583M
)
3025201510
35
05
(g)
005
115
2
253
35
Con
cent
ratio
n in
cell
(120583M
)
0 10 20 30 40 50 60 70 80Time (hours)
DHEA
E2ALDO
TESTO
(h)
Figure 4 Comparison of time-course profiles of cholesterol and adrenal steroids produced by NCI-H295R cells between experimentallymeasured and simulated data To intuitively confirm reconstruction of the measured experimental data in the developed simulation modelof NCI-H295R cells dynamics of cholesterol and adrenal steroids produced by NCI-H295R cells were plotted to overlay experimentaldata with the simulated results Graphs show the dynamics of medium concentrations of cholesterol (a) and adrenal steroids ((b)ndash(d))and intracellular concentrations of cholesterol (e) and adrenal steroids ((f)ndash(h)) Steroid hormones were categorized into three groups byconcentration levels Major steroids were PREG pregnenolone DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICOcorticosterone and CORT cortisol ((b) and (f)) Moderate steroids were HPREG 17120572-hydroxypregnenolone PROG progesterone HPROG17120572-hydroxyprogesterone DIONE androstenedione and E1 estrone ((c) and (g)) Minor steroids were DHEA dehydroepiandrosteroneALDO aldosterone TESTO testosterone and E2 17120573-estradiol ((d) and (h)) Experimental data are shown as symbols with dotted lines Alldata represent the mean plusmn SD (119873 = 4) Simulation data are shown as solid lines
Journal of Toxicology 11
0110
10
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
AN (120583M)
(a)
0110
FN (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(b)
00101
SM (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
100E minus 05
(c)
0110
TG (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B1
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(d)
0110
10100
AGT (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(e)
0110
1025
DDD (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(f)
1010
SP (120583M)
2750005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(g)
0110
10100
MP (120583M)
2110005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(h)
Figure 5 Continued
12 Journal of Toxicology
0110
10
KC (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(i)
0110
10
MC (120583M)
4430005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(j)
0110
10100
DZ (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(k)
Figure 5 Estimated mechanism of action of adrenal toxicants by using the mathematical model of adrenal steroidogenesis in NCI-H295Rcells Mechanisms of action of adrenal toxicity compounds were quantitatively estimated from experimental results of differential steroidprofiling using the mathematical model of adrenal steroidogenesis in NCI-H295R cells The drug action was defined as a scaling factorof enzymatic activity in the simulation model These scaling factors were optimized to fit experimental data by a hybrid optimizationmethod of the RCGA and nonlinear least squares Estimated drug actions by the exposure of vasculotoxic agents acrylonitrile AN (a)fumaronitrile FN (b) salinomycin SM (c) and thioguanine TG (d) and the steroidogenic inhibitors aminoglutethimide AGT (e) op1015840-DDD DDD (f) spironolactone SP (g) metyrapone MP (h) ketoconazole KC (i) miconazole MC (j) and daidzein DZ (k) are shownas a spider radar chart CYP11A1 P450 side chain cleavage enzyme CYP17H 17120572-hydroxylase of CYP17 CYP17L C
17ndash20 lyase of CYP17HSD3B2 3120573-hydroxysteroid dehydrogenase CYP21A2 21-hydroxylase CYP11B1 11120573-hydroxylase CYP11B2 18-hydroxylase and HSD17B317120573-hydroxysteroid dehydrogenase
with previous reports [33 34] 7120572-thiospironolactone whichis synthesized by deacetylation of SP inhibits CYP17H andCYP17L [34] The fact that there were no inhibitions ofCYP17HorCYP17L in our study suggests that SPmight not bedeacetylated to 7120572-thiospironolactone in NCI-H295R cellsRegarding CYP11B1 and CYP11B2 our results were unclearcompared with a previous study It has been shown that30 120583M SP inhibits CYP11B1 and CYP11B2 in human andbovine adrenal mitochondria [35]The cause of CYP11B1 andCYP11B2 inhibition by SP could not be determined in ourstudy which might be due to the lower maximum exposureconcentration of SP than that in the previous reportWe couldnot examine SP concentrations over 10 120583M because theseconcentrations were cytotoxic in NCI-H295R cells
335 MP The mechanism of action of MP in adrenalsteroidogenesis was estimated by dose-dependent inhibitionof CYP11B1 (estimated inhibitions at 1 10 and 100 120583M were571 827 and 982 resp) (Figure 5(h)) MP has beenreported to inhibit CYP11B1 as its major effect and CYP11A1and CYP11B2 as a weak effect [6 29 36ndash39] The results wereable to show that MP is a selective inhibitor of CYP11B1 inthe previous report However the estimated effect of MPat a high concentration 100 120583M as the maximum exposureconcentration was unclear According to the previous reportselectivity of MP for CYP11B1CYP11B2 is about five times[39] In addition 20 120583M MP has a slight inhibition effect onCYP11A1 in H295R cells [29]
336 KC The mechanism of action of KC in adrenalsteroidogenesis was estimated by inhibition of CYP11A1CYP17H CYP17L HSD3B2 CYP21A2 CYP11B1 and
CYP11B2 (estimated inhibitions at 10 120583Mwere 926 943518 830 882 974 and 798 resp) (Figure 5(i))KC has been reported to inhibit CYP11A1 CYP17H CYP17LHSD3B2 CYP21A2 and CYP11B1 [6 29 40ndash43] Our resultswere almost consistent with the previous reports KC inhibitsCYP11A1 CYP17H CYP21A2 and CYP11B1 in NCI-H295Rcells at 10 120583M [29] and CYP17H CYP17L CYP21A2 andCYP11B1 in human adrenal mitochondria and Leydig cellmicrosomes at 2ndash5120583M [60 61] However KC has shownonly weak inhibition of HSD3B2 and HSD17B3 in Leydigcells at the millimolar level [60 61] Regarding CYP11B2 andHSD17B3 we considered that these estimated inhibitions ofKC did have sufficient reliability in terms of quantitative pre-diction precision because ALDO and TESTO concentrationswere less than the lower limit of quantification at 10 120583MKC
337 MC Themechanism of action of MC in adrenal stero-idogenesis was estimated by inhibition of CYP17H CYP17LCYP11B1 and HSD17B3 (estimated inhibitions at 10120583Mwere691 530 764 and 571 resp) (Figure 5(j)) MC hasbeen reported to inhibit not only CYP17H and CYP17L butalsoCYP11A1 CYP21A2 andCYP11B1 [41 44 45]The resultsin the previous reports were able to estimate that MC isa CYP17 inhibitor However CYP11A1 inhibition by MCprobably instead of a reduction in StAR expression wasnot clearly detected in our study using NCI-H295R cellsbecause there were no decreases in the concentrations ofPREG and PROG in the culture medium Indirect inhibitionof CYP11A1 via the peripheral-type benzodiazepine receptorhas been reported in mouse adrenocortical Y-1 cells treatedwith MC in the absence of stimuli by measuring PREGproduction [44] On the other hand reductions of StAR
Journal of Toxicology 13
protein expression andor transport activity without affectingtotal steroid synthesis or CYP11A1 and HSD3B2 enzymeexpression or activities have been reported in (BU)
2cAMP-
stimulated MA-10 Leydig tumor cells treated with MC bymeasuring PROG production [45] Therefore the effect ofMC on the initial reaction in adrenal steroidogenesis fromcholesterol should be different according to the cell type andstimulation condition Inhibition of CYP21A2 and CYP11B1by MC has been reported as decreases in the consumptionof PROG and DCORTICO respectively [41] Inhibition ofCYP11B1 was estimated by the action of MC in this studybut that of CYP21A2 was not detected In the previousexperimental report inhibitory sites by MCmight have beenreflected by inhibition of CYP17H activity because CYP21A2activity was measured as a decrease in labeled PROG
338 DZ The mechanism of action of DZ in adrenalsteroidogenesis was estimated by inhibition of CYP11A1HSD3B2 CYP21A2 CYP11B1 and HSD17B3 (estimated inhi-bitions at 100 120583M were 586 941 965 872 and981 resp) (Figure 5(k)) DZ has been reported to inhibitHSD3B2 and CYP21A2 [46] The results of HSD3B2 andCYP21A2 were consistent with the previous report Howeverinhibitions have not been reported for CYP11A1 CYP11B1and HSD17B3These estimated effects of DZ on CYP11B1 andHSD17B3 were unclear because the concentrations of ALDOand TESTO were less than the lower limit of quantificationat 100 120583M In addition these enzymes act downstream of thestrong action points of DZ such as HSD3B2 and CYP21A2
34 The Simulations and the Systematic Model Analysis
341 Dynamic Sensitivity Analysis of Adrenal SteroidogenesisTo comprehensively understand the dynamics of adrenalsteroidogenesis dynamic sensitivities were calculated forsteroid concentrations secreted by NCI-H295R cells usingour constructed mathematical model of steroidogenesis Theresults of dynamic sensitivity analysis at 72 h of duration and6 h of interval time are presented as a heat-map in Figure 6
The top 10 parameters of the total area under the curveof dynamic sensitivity for cholesterol and the 14 steroids inculture medium were 119881HSD3B2
maxA 119881CYP11A1max 119881CYP21A2maxA 119896loc119891 119896loc119887
119870CYP11A1119898
119881CYP17HmaxA 119896Cholesterol Transport 119896acc119891 and 119881CYP21A2maxB in
order from the top Cholesterol uptake (119896Cholesterol Transport)StAR protein (119896loc
119891) and CYP11A1 (119881CYP11A1max ) which are
determining factors of the capacity for steroidogenesis pro-moted the production of mineralocorticoids (DCORTICOCORTICO and ALDO) and restrained the synthesis of glu-cocorticoids (DCORT and CORT) and sex steroids (DIONETESTO and E1) because of the accumulation of interme-diate molecules in steroidogenesis (PREG HPREG DHEAPROG and HPROG) only by self-activation The dynamicpatterns of the intermediate molecules in steroidogenesiswere mainly dependent on the activity of CYP17H andHSD3B2 with PREG as the substrate of these enzymes inwhich the dynamic sensitivities of 119881CYP17HmaxA for HPREG andHPROG and 119881HSD3B2
maxA for PROG HPROG and DCORTICO
reversed the direction of sensitivity at 49ndash66 h after stimu-lation The dynamic sensitivities of the maximum activitiesof HSD3B2 for PREG (119881HSD3B2
maxA ) and CYP21A2 for PROG(119881CYP21A2maxA ) were related to all steroids at 72 h Almost allmodel parameters had positive sensitivity for downstreamsteroids in the adrenal steroidogenic pathway and negativesensitivity for direct-binding steroids as substrates of thesteroidogenic enzyme The sensitivity of 119881max in all steroido-genic enzymes was relatively higher than 119870
119898for the same
steroid substrate
342 Simulation of the Metabolic Balance of AdrenalSteroidogenesis Pathway To clearly show the property of themetabolic shift between mineralocorticoid and glucocorti-coid biosynthesis we performed two-dimensional parameterscanning of the enzymatic activities of CYP17H and HSD3B2(Figure 7) NCI-H295R cells lost the ability to produce allsteroid hormones when enzymatic activities of CYP17H andHSD3B2 were changed by over 60 and 30 respectivelyActivation ofCYP17HandorHSD3B2 induced themetabolicshift that enhanced the glucocorticoid biosynthesis and devi-ated from the mineralocorticoid biosynthesis On the otherhand inhibition of CYP17H andor HSD3B2 induced themetabolic shift that enhanced the mineralocorticoid biosyn-thesis and deviated from the glucocorticoid biosynthesisMoreover the enzymatic activity of HSD3B2 regulated themetabolic balance of sex steroids and the precursors onadrenal steroidogenesis of NCI-H295R cells E1 TESTO andDIONE were produced by NCI-H295R cells when activatingthe enzymatic activity of HSD3B2 Conversely E2 andDHEAwere produced by NCI-H295R cells when suppressing theenzymatic activity of HSD3B2 The biosynthesis of down-stream steroids in adrenal steroidogenesis pathway such asmineralocorticoids and glucocorticoids was almost com-pletely terminated when the enzymatic activity of HSD3B2was decreased by over 80
4 Discussion
41 Importance of 3120573-HSDActivity in Adrenal SteroidogenesisOur systematic analysis using the mathematical model ofadrenal steroidogenesis in NCI-H295R cells revealed thatthe enzymatic activity of 3120573-HSD controls the dynamics ofadrenal steroidogenesis The activity of the StAR proteincontrols the net capacity of steroidogenesis in steroidogeniccells which is the transport of cholesterol from the outer toinner mitochondrial membranes Both the expression levelsof StAR protein and mRNA are rapidly elevated in responseto stimulation by tropic hormones such as ACTH [62 63]Another important factor in adrenal steroidogenesisis is thecholesterol side chain cleavage enzyme CYP11A1 the firstrate-limiting and hormonally regulated step in the synthesisof all steroid hormones which is conversion of cholesterolto pregnenolone in mitochondria [64] According to ourresults of global sensitivity analysis (Supplementary Table 1and Figure 6(d)) in addition to CYP11A1 and StAR proteins3120573-HSD was one of the key regulators in adrenal steroido-genesis of NCI-H295R cells And also this result suggeststhat a significant regulatory mechanism in steroidogenesis
14 Journal of Toxicology
60544830 4236 66 7218 246 120
Time (hours)
0
1
2
3C
once
ntra
tion
(a)
60544830 4236 66 7218 246 120
Time (hours)
minus05
0
05
minus1
1
Sens
itivi
ty
(b)
0
6
12
18
24
30
36
42
48
54
60
66
72
Sens
itivi
ty fo
r med
ium
conc
entr
atio
ns
of ch
oles
tero
l or s
tero
id h
orm
ones
Dur
atio
n tim
e (ho
urs)
Model parameters
+10
+05
00
minus05
minus10
(c)
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
Parameter number in the mathematical model of steroidogenesis in NCI-H295R cells
CHOL
PREG
HPREG
DHEA
PROG
HPROG
DIONE
DCORTICO
DCORT
CORTICO
CORT
ALDO
TESTO
E1
E2
Chol
este
rol o
r ste
roid
hor
mon
es
0
05
10
15
20
minus05
minus10
minus15
minus20
(d)
Figure 6 Heat-map of the global dynamic sensitivity analysis of adrenal steroid concentrations produced by NCI-H295R cells The globaldynamic sensitivity analysis is a powerful tool to comprehensively understand the dependencies of themodel parameters in themathematicalmodel of the biological complex system Dynamic sensitivities of model parameters in the mathematical model of adrenal steroidogenesisin NCI-H295R cells were calculated for all steroid concentrations in the culture medium every 6 h until 72 h after stimulation To clarify theview of heat-map of global dynamic sensitivity analysis imaginary data of the dynamics of steroid concentrations in the original model (blueline) and perturbed model for sensitivity analysis (red line) were prepared (a) Using this imaginary data the calculated dynamic sensitivities(b) and the visualized dynamic sensitivity as one block of the heat-map (c) were shown respectively By the same method that explainedusing imaginary data the large-scale data of the global dynamic sensitivity analysis on the mathematical model of adrenal steroidogenesisin NCI-H295R cells was comprehensively visualized as a big graph of heat-map (d) Parameter numbers in the horizontal axis are (1)119896Choresterol Transport (2) 119896CEH (3) 119896MTR
119891 (4) 119896MTR
119887 (5) 119896acc
119891 (6) 119896acc
119887 (7) 119896loc
119891 (8) 119896loc
119887 (9) 119896Oxysterol Synthesis (10)119870
CYP11A1119898
(11) 119881CYP11A1max (12)119870CYP17H119898A
(13) 119870CYP17H119898B (14) 119881CYP17HmaxA (15) 119881CYP17HmaxB (16) 119881CYP17HmaxB (17) 119870CYP17L
119898B (18) 119881CYP17LmaxA (19) 119881CYP17LmaxB (20) 119870HSD3B2119898A (21) 119870HSD3B2
119898B (22) 119870HSD3B2119898C (23)
119881HSD3B2maxA (24) 119881HSD3B2
maxB (25) 119881HSD3B2maxB (26) 119870CYP21A2
119898A (27) 119870CYP21A2119898B (28) 119881CYP21A2maxA (29) 119881CYP21A2maxB (30) 119870CYP11B1
119898A (31) 119870CYP11B1119898B (32) 119881CYP11B1maxA
(33) 119881CYP11B1maxB (34) 119896CYP11B2 (35) 119870HSD17B3119898A (36) 119870HSD17B3
119898B (37) 119881HSD17B3maxA (38) 119881HSD17B3
maxB (39) 119870CYP19A1119898A (40) 119870CYP19A1
119898B (41) 119881CYP19A1maxA and (42)119881
CYP19A1maxB PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-
hydroxyprogesteroneDIONE androstenedioneDCORTICO 11-deoxycorticosteroneDCORT 11-deoxycortisol CORTICO corticosteroneCORT cortisol ALDO aldosterone and TESTO testosterone
pathway is very reasonable StAR CYP11A1 and 3120573-HSD(isoforms 1 or 2 in humans) proteins generally respond to thesame hormones that stimulate steroid production throughcommon pathways such as cAMP signaling in adrenal glandsand testes [65 66] Moreover our data also support recentexperimental evidence from clinical and in vivo studiessuggesting that the enzymatic activity of 3120573-HSD plays
an important role in the regulation of mineralocorticoidsynthesis in adrenal steroidogenesis and contributes to hyper-tension caused by abnormal overproduction of aldosterone[67ndash70] Circadian clock-deficient Cry-null mice show salt-sensitive hypertension due to abnormally high synthesis ofaldosterone which is caused by constitutively high expressionof HSD3B6mRNA and protein in the adrenal cortex [67 68]
Journal of Toxicology 15
All steroid producing cells
Mineralocorticoid producing cells Glucocorticoid producing cells
Mineralocorticoid and glucocorticoid producing cells
Terminated steroidogenesis
Sex steroids and precursor sex steroidsDIONE TESTO E1DEHA E2
NCI-H295Rcells at controlcondition
Normalized activities of HSD3B2
Nor
mal
ized
activ
ities
of C
YP17
H
2
18
16
14
12
1
08
06
04
02
00 05 1 15 2
Figure 7 Metabolic categories of the steroidogenic cells determined by the balance of HSD3B2 and CYP17H activitiesThe two-dimensionalparameter scanning analysis by the perturbation of two focused parameters clarifies the interaction and the relationship between the twoparameters in the complex system Functional cellular categories of steroidogenic cells were defined by the levels of mineralocorticoid(ALDO) glucocorticoids (DCORTandCORT) and androgens (DHEAandDIONE) at 72 h after stimulation Enzymatic activities ofHSD3B2andCYP17were normalized by standard values of the simulationmodel inNCI-H295R cells Scanning ranges ofHSD3B2 andCYP17 activitieswere 0ndash200 each 10 Green regions are all 14 steroids produced by NCI-H295R cells Red blue and purple regions are mineralocorticoidproducing cells glucocorticoids-producing cells and both corticoid-producing cells respectively Yellow regions are steroidogenesis of NCI-H295R cells terminated upstream of the adrenal steroidogenesis pathway
Recent clinical observations have revealed predominantexpression of HSD3B2 mRNA and protein in tumor cellsof aldosterone-producing adenoma (APA) and HSD3B1mRNA significantly correlated with CYP11B2 mRNA levelsand plasma aldosterone concentrations in APA patients [6970] However the relationship is unclear and disputed ina small-scale clinical study indicating that genetic variationin HSD3B1 affects blood pressure and hypertension in APApatients [71] The results of our simulation study suggest that3120573-HSD protein (human genes are HSD3B1 and HSD3B2) isone of the determination factors for the dynamic propertyof adrenal steroidogenesis Our results support the clinicalevidence of Doi and colleagues [69] and we believe that theHSD3B1 enzyme has a promising potential as novel drugtarget for endocrine hypertension
42 Metabolic Shift of Adrenal Steroidogenesis and the Contri-butions of HSD3B2 and CYP17H The metabolic propertiesof adrenal steroidogenesis in NCI-H295R cells were revealedby dynamic sensitivity analysis using themathematicalmodel(Figure 6) Mineralocorticoids such as DCORTICO COR-TICO and ALDO and intermediate steroids upstream of
the adrenal steroidogenesis pathway such as PREG HPREGDHEA PROG and HPROG were accelerated by reactionsof cholesterol import (119896Cholesterol Transport) StAR protein (119896MTR
119891
and 119896loc119891) and CYP11A1 (119881CYP11A1max ) On the other hand glu-
cocorticoids such as DCORT and CORT and sex hormonessuch as DIONE TESTO and E1 were suppressed by thesemodel parametersTherefore enhancement of the net adrenalsteroidogenesis capacity which supplies PREG precursor tothe pathway causes a production shift from glucocorticoidsto mineralocorticoids by substrate inhibitions of CYP17HHSD3B2 and CYP21A2 caused by accumulation of initialintermediate steroids such as PREG and PROG Sensitivi-ties of CYP17H (119881CYP17HmaxA ) and HSD3B2 (119881HSD3B2
maxA ) for theproducts were dynamically changed and these parametersdetermined the metabolic balance of downstream steroidsin the adrenal steroidogenesis pathway According to theseresults of dynamic sensitivity analysis of StAR CYP11A1CYP17H and HSD3B2 we suggest that the enhancementof CYP17H and HSD3B2 activity during ACTH stimulationwas important to shift the steroidogenic output away fromALDO biosynthesis towards CORT biosynthesis as wellas adrenal androgen production This suggestion partially
16 Journal of Toxicology
supports a comparative animal study in which molecularand cellular variations in CYP17H activity dramatically affectacute cortisol production resulting in distinct physiologicaland behavioral responses [72]
Results of two-dimensional parameter scanning of theenzymatic activities of CYP17H and HSD3B2 quantitativelyshowed the detail of the metabolic relationship betweenmin-eralocorticoid and glucocorticoid biosynthesis (Figure 7)Particularly the results showed that the balance of theseenzymatic activities was very important for the typicalfunction of NCI-H295R cells namely the ability to produceall steroid hormones NCI-H295R cells lost this functionwhen enzymatic activities of CYP17H and HSD3B2 werechanged by over 60 and 30 respectively In addition theybecame mineralocorticoid (ALDO) secreting cells when theenzymatic activity of CYP17H or HSD3B2 was inhibited byover 50 or glucocorticoid (DCORT and CORT) secretingcells when these enzymes were activated by over 50 Inparticular this analysis also showed that HSD3B2 was a keyplayer in the adrenal steroidogenesis of NCI-H295R cellsbecause HSD3B2 inhibition by over 80 almost completelyinhibited the biosynthesis of downstream steroids The ratioof CYP17A1 to HSD3B2 mRNA expression levels has beenrelated to several endocrine diseases with a low level in APAs[73] and high level in cortisol-producing adenomas [74]Furthermore the expression levels or enzymatic activitiesof CYP17A1 and HSD3B1 have been related to androgenproduction in polycystic ovary syndrome [75 76] Theseclinical studies support our simulation results indicating thatthe balance of enzymatic activity of CYP17H and HSD3B2determines the shift in steroidogenic output to mineralocor-ticoids glucocorticoids or androgens
43 Methodologies of Quantitative Mechanistic Analysis forDrug Discovery According to our results obtained using themathematical model of steroidogenesis in NCI-H295R cellssuch as sensitivity analysis comprehensive analysis based onsystems biology is available to quantitatively estimate themechanism of action of steroidogenic disrupting compoundsfrom differential profiling of adrenal steroid hormonesbecause dynamic patterns of steroid hormones in adrenalsteroidogenesis pathway are highly complex Our proposedmethod of quantitative mechanistic analysis of steroidogenicinhibitors was able to predict known action sites in theadrenal steroidogenesis pathway at only one time point (72 hafter drug exposure) Moreover according to the resultsof sensitivity analysis (Figure 6) 119881max of all steroidogenicenzymes was more sensitive than the 119870
119898 because the
intracellular concentrations of steroid hormones were almostmaintained at sufficiently high levels compared with 119870
119898
values of steroidogenic enzymesThese results suggested thatestimation of the mechanism of action of drugs is moreeffective and detectable when using the influences of 119881maxas the searching parameters such as our proposed methodOur data showed that the proposed method based on asystems biology model is a very powerful tool for exploratoryscreening of steroidogenic disrupting compounds
RCGA as a solver of parameter estimation problemsin systems biology has been applied to biological network
identification of gene regulatory networks and metabolicpathways and optimization of biological processes usingexperimentally observed time-course data [77ndash82] In thisstudy RCGA was useful to estimate the mechanism ofaction of novel pharmaceutical drug candidates for adrenalsteroidogenesis as a new application of RCGA in systemsbiology We had two issues when applying RCGA to thequantitative mechanistic analysis of drug actions Theseissues were the vast calculation cost and multimodality ofquasi-optimum solutions in solving the optimization prob-lem because the mathematical model in systems biologyconsists of many equations and parameters A proposedoptimization strategy using RCGA based on REXJGG wasa highly stable and efficient calculation method for a betterquasi-optimum solution than the unimodal normal distri-bution crossover (UNDX)minimum generation gap (MGG)method that is well applied in the engineering field In addi-tion we expanded the RCGA optimization program basedon REXJGG to a hybrid method of GA and then applieda local search as recommended by Harada and Kobayashi[28 83] A final optimal solution was obtained with a goodconvergence property Because these problems are generalin systems biology studies we suggest that the proposedhybrid method based on REXJGG is a very useful tool forquantitative mechanistic analysis of novel pharmaceuticaldrugs not limited to steroidogenic disrupting compounds
5 Conclusions
Thenovelmathematicalmodel of adrenal steroidogenesis wasconstructed in this study including cholesterol transport anddistribution the C
21-steroid hormone pathway steroid trans-
port and cell proliferation which could reproduce adrenalsteroidogenesis in NCI-H295R cells According to the resultsof dynamic sensitivity analysis using the newmodel HSD3B2plays the most important role in the metabolic balance ofadrenal steroidogenesis in NCI-H295R cells Moreover toquantitatively estimate mechanisms of action of adrenal toxiccompounds we analyzed differential metabolic profiles of12 steroid hormones at 3 days after exposure to 11 adrenaltoxic compounds by using the new mathematical modeland a hybrid optimization method of the RCGA and a localsearch (nonlinear least squares) We could estimate whichsteroidogenic enzymes were affected by these compoundsusing the hybrid optimization method Vasculotoxic agentswere estimated to have no effect according to the resultsobtained by our method In terms of adrenal steroidogenicinhibitors the predicted action sites were approximatelymatched to the target enzymes as reported in the literatureThus our computer-aided method based on a systems biol-ogy approach may be useful to analyze the mechanism ofaction of endocrine-disrupting compounds and to assess thehuman adrenal toxicity of novel pharmaceutical drugs basedon steroid hormone profiling
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Journal of Toxicology 17
Authorsrsquo Contribution
Ryuta Saito Natsuko Terasaki and Naohisa Tsutsui designedthe study Naohisa Tsutsui and Naoya Masutomi supervisedthe experiments Natsuko Terasaki performed all in vitroexperiments and collected dataMakotoYamazaki performedall measurements of steroids Ryuta Saito conceived the insilico strategies analyzed data developed all programs andthe mathematical model and performed all simulation anal-yses with Masahiro Okamoto Ryuta Saito Natsuko TerasakiMakoto Yamazaki NaoyaMasutomi andMasahiroOkamotowrote the paper
Acknowledgments
The authors would like to thank Dr Michael S Breenand Dr Miyuki Breen for helpful discussions concerningprevious mathematical models of adrenal steroidogenesisThis work was partially supported by the Grant-in-Aid forScientific Research on Innovative Areas ldquoSynthetic Biologyrdquo(No 23119001 (MO)) from Ministry of Education CultureSports Science and Technology in Japan
References
[1] P W Harvey D J Everett and C J Springall ldquoAdrenal toxi-cology a strategy for assessment of functional toxicity to theadrenal cortex and steroidogenesisrdquo Journal of Applied Toxi-cology vol 27 no 2 pp 103ndash115 2007
[2] W E Ribelin ldquoThe effects of drugs and chemicals upon thestructure of the adrenal glandrdquo Fundamental and Applied Toxi-cology vol 4 no 1 pp 105ndash119 1984
[3] H D Colby and P A Longhurst ldquoToxicology of the adrenalglandrdquo in Endocrine Toxicology C K Atterwill and J D FlackEds pp 243ndash281 CambridgeUniversity Press Cambridge UK1992
[4] H D Colby ldquoIn vitro assessment of adrenocortical toxicityrdquoJournal of Pharmacological and Toxicological Methods vol 32no 1 pp 1ndash6 1994
[5] T J Rosol J T Yarrington J Latendresse and C C CapenldquoAdrenal gland structure function and mechanisms of toxic-ityrdquo Toxicologic Pathology vol 29 no 1 pp 41ndash48 2001
[6] P W Harvey and D J Everett ldquoThe adrenal cortex and ster-oidogenesis as cellular and molecular targets for toxicity crit-ical omissions from regulatory endocrine disrupter screeningstrategies for human healthrdquo Journal of Applied Toxicology vol23 no 2 pp 81ndash87 2003
[7] A F Gazdar H K Oie C H Shackleton et al ldquoEstablishmentand characterization of a human adrenocortical carcinoma cellline that expresses multiple pathways of steroid biosynthesisrdquoCancer Research vol 50 no 17 pp 5488ndash5496 1990
[8] W E Rainey I M Bird C Sawetawan et al ldquoRegulation ofhuman adrenal carcinoma cell (NCI-H295) production of C19steroidsrdquo The Journal of Clinical Endocrinology amp Metabolismvol 77 no 3 pp 731ndash737 1993
[9] B Staels D W Hum andW L Miller ldquoRegulation of steroido-genesis in NCI-H295 cells a cellular model of the human fetaladrenalrdquo Molecular Endocrinology vol 7 no 3 pp 423ndash4331993
[10] M Hecker J L Newsted M B Murphy et al ldquoHuman adreno-carcinoma (H295R) cells for rapid in vitro determination of
effects on steroidogenesis hormone productionrdquo Toxicologyand Applied Pharmacology vol 217 no 1 pp 114ndash124 2006
[11] A Oskarsson E Ulleras K E Plant J P Hinson and P SGoldfarb ldquoSteroidogenic gene expression in H295R cells andthe human adrenal gland adrenotoxic effects of lindane invitrordquo Journal of Applied Toxicology vol 26 no 6 pp 484ndash4922006
[12] UMuller-Vieira M Angotti and RW Hartman ldquoThe adreno-cortical tumor cell line NCI-H295R as an in vitro screeningsystem for the evaluation of CYP11B2 (aldosterone synthase)and CYP11B1 (steroid-11120573-hydroxylase) inhibitorsrdquoThe Journalof Steroid Biochemistry and Molecular Biology vol 96 no 3-4pp 259ndash270 2005
[13] K Imagawa S Okayama M Takaoka et al ldquoInhibitory effectof efonidipine on aldosterone synthesis and secretion in humanadrenocarcinoma (H295R) cellsrdquo Journal of CardiovascularPharmacology vol 47 no 1 pp 133ndash138 2006
[14] M Voets UMuller-Vieira S Marchais-Oberwinkler and RWHartmann ldquoSynthesis of amidinohydrazones and evaluation oftheir inhibitory effect towards aldosterone synthase (CYP11B2)and the formation of selected steroidsrdquo Archiv der Pharmazievol 337 no 7 pp 411ndash416 2004
[15] F K Nielsen C H Hansen J A Fey et al ldquoH295R cells as amodel for steroidogenic disruption a broader perspective usingsimultaneous chemical analysis of 7 key steroid hormonesrdquoToxicology in Vitro vol 26 no 2 pp 343ndash350 2012
[16] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cortisolproduction and on activities of steroidogenic enzymes inhuman adrenocortical H295R cellsrdquo Journal of Steroid Biochem-istry and Molecular Biology vol 80 no 3 pp 355ndash363 2002
[17] R F Canton J T Sanderson S Nijmeijer A Bergman R JLetcher and M van den Berg ldquoIn vitro effects of brominatedflame retardants and metabolites on CYP17 catalytic activity anovel mechanism of actionrdquo Toxicology and Applied Pharma-cology vol 216 no 2 pp 274ndash281 2006
[18] K Hilscherova P D Jones T Gracia et al ldquoAssessment ofthe effects of chemicals on the expression of ten steroidogenicgenes in the H295R cell line using real-time PCRrdquo ToxicologicalSciences vol 81 no 1 pp 78ndash89 2004
[19] T Gracia K Hilscherova P D Jones et al ldquoThe H295R systemfor evaluation of endocrine-disrupting effectsrdquo Ecotoxicologyand Environmental Safety vol 65 no 3 pp 293ndash305 2006
[20] X Zhang J L NewstedM Hecker E B Higley P D Jones andJ P Giesy ldquoClassification of chemicals based on concentration-dependent toxicological data using ToxClustrdquo EnvironmentalScience and Technology vol 43 no 10 pp 3926ndash3932 2009
[21] M S Breen M Breen N Terasaki M Yamazaki and R BConolly ldquoComputational model of steroidogenesis in humanH295R cells to predict biochemical response to endocrine-active chemicalsmodel development formetyraponerdquoEnviron-mental Health Perspectives vol 118 no 2 pp 265ndash272 2010
[22] M Breen M S Breen N Terasaki M Yamazaki A LLloyd and R B Conolly ldquoMechanistic computational model ofsteroidogenesis in H295R cells role of oxysterols and cell pro-liferation to improve predictability of biochemical response toendocrine active chemical-metyraponerdquo Toxicological Sciencesvol 123 no 1 pp 80ndash93 2011
[23] F B Kraemer ldquoAdrenal cholesterol utilizationrdquo Molecular andCellular Endocrinology vol 265-266 pp 42ndash45 2007
[24] W L Miller ldquoSteroidogenic acute regulatory protein (StAR)a novel mitochondrial cholesterol transporterrdquo Biochimica et
18 Journal of Toxicology
Biophysica Acta (BBA)mdashMolecular and Cell Biology of Lipidsvol 1771 no 6 pp 663ndash676 2007
[25] S Szabo and I T Lippe ldquoAdrenal gland chemically inducedstructural and functional changes in the cortexrdquo ToxicologicPathology vol 17 no 2 pp 317ndash329 1989
[26] C Chen-Pan I-J Pan Y Yamamoto T Sakogawa J Yamadaand YHayashi ldquoPrompt recovery of damaged adrenalmedullaeinduced by salinomycinrdquoToxicologic Pathology vol 27 no 5 pp563ndash572 1999
[27] R J Santen and R I Misbin ldquoAminoglutethimide review ofpharmacology and clinical userdquo Pharmacotherapy vol 1 no 2pp 95ndash120 1981
[28] M J Carella N V Dimitrov V V Gossain L Srivastava andD R Rovner ldquoAdrenal effects of low-dose aminoglutethimidewhen used alone in postmenopausal women with advancedbreast cancerrdquoMetabolism vol 43 no 6 pp 723ndash727 1994
[29] M K Johansson J T Sanderson and B-O Lund ldquoEffects of3-MeSO2-DDE and some CYP inhibitors on glucocorticoidsteroidogenesis in the H295R human adrenocortical carcinomacell linerdquo Toxicology in Vitro vol 16 no 2 pp 113ndash121 2002
[30] Y Touitou A Bogdan J C Legrand and P Desgrez ldquoAmino-glutethimide and glutethimide effects on 18-hydroxycorti-costerone biosynthesis by human and sheep adrenals in vitrordquoActa Endocrinologica vol 80 no 3 pp 517ndash526 1975
[31] M M Hart and J A Straw ldquoStudies on the site of action ofop1015840-DDD in the dog adrenal cortex 1 Inhibition of ACTH-mediated pregnenolone synthesisrdquo Steroids vol 17 no 1ndash5 pp559ndash574 1971
[32] M Ojima M Saitoh N Itoh Y Kusano S Fukuchi and HNaganuma ldquoThe effects of op1015840-DDD on adrenal steroidogen-esis and hepatic steroid metabolismrdquo Nippon Naibunpi Gakkaizasshi vol 61 no 3 pp 168ndash178 1985
[33] L F Canosa and N R Ceballos ldquoEffects of different steroid-biosynthesis inhibitors on the testicular steroidogenesis of thetoad Bufo arenarumrdquo Journal of Comparative Physiology B vol171 no 6 pp 519ndash526 2001
[34] D C Kossor S Kominami S Takemori and H D ColbyldquoRole of the steroid 17 alpha-hydroxylase in spironolactone-mediated destruction of adrenal cytochrome P-450rdquoMolecularPharmacology vol 40 no 2 pp 321ndash325 1991
[35] S C Cheng K Suzuki W Sadee and B W Harding ldquoEffectsof spironolactone canrenone and canrenoate K on cytochromeP450 and 11120573 and 18 hydroxylation in bovine and humanadrenal corticalmitochondriardquo Endocrinology vol 99 no 4 pp1097ndash1106 1976
[36] K Yanagibashi M Haniu J E Shively W H Shen and P HallldquoThe synthesis of aldosterone by the adrenal cortex Two zones(fasciculata and glomerulosa) possess one enzyme for 11120573- 18-hydroxylation and aldehyde synthesisrdquoThe Journal of BiologicalChemistry vol 261 no 8 pp 3556ndash3562 1986
[37] K Morishita H Okumura N Ito and N Takahashi ldquoPrimaryculture system of adrenocortical cells from dogs to evaluatedirect effects of chemicals on steroidogenesisrdquo Toxicology vol165 no 2-3 pp 171ndash178 2001
[38] S W J Lamberts E G Bons H A Bruining and F H de JongldquoDifferential effects of the imidazole derivatives etomidateketoconazole and miconazole and of metyrapone on the secre-tion of cortisol and its precursors by human adrenocorticalcellsrdquo Journal of Pharmacology and Experimental Therapeuticsvol 240 no 1 pp 259ndash264 1987
[39] L Yin S Lucas F Maurer U Kazmaier Q Hu and RW Hartmann ldquoNovel imidazol-1-ylmethyl substituted 1256-tetrahydropyrrolo[321-ij]quinolin-4-ones as potent and selec-tive CYP11B1 inhibitors for the treatment of Cushingrsquos syn-dromerdquo Journal ofMedicinal Chemistry vol 55 no 14 pp 6629ndash6633 2012
[40] M DiMattina N Maronian H Ashby D L Loriaux andB D Albertson ldquoKetoconazole inhibits multiple steroidogenicenzymes involved in androgen biosynthesis in the humanovaryrdquo Fertility and Sterility vol 49 no 1 pp 62ndash65 1988
[41] M Ayub and M J Levell ldquoInhibition of human adrenal steroi-dogenic enzymes in vitro by imidazole drugs including keto-conazolerdquo Journal of Steroid Biochemistry vol 32 no 4 pp 515ndash524 1989
[42] D Engelhardt M M Weber T Miksch F Abedinpour andC Jaspers ldquoThe influence of ketoconazole on human adrenalsteroidogenesis incubation studies with tissue slicesrdquo ClinicalEndocrinology vol 35 no 2 pp 163ndash168 1991
[43] S Bhasin S Sikka T Fielder et al ldquoHormonal effects of keto-conazole in vivo in the male rat mechanism of actionrdquo Endo-crinology vol 118 no 3 pp 1229ndash1232 1986
[44] D M Zisterer and D C Williams ldquoCalmidazolium and otherimidazole compounds affect steroidogenesis in Y1 cells lack ofinvolvement of the peripheral-type benzodiazepine receptorrdquoJournal of Steroid Biochemistry and Molecular Biology vol 60no 3-4 pp 189ndash195 1997
[45] L P Walsh C N Kuratko and D M Stocco ldquoEconazoleand miconazole inhibit steroidogenesis and disrupt steroido-genic acute regulatory (StAR) protein expression post-trans-criptionallyrdquo Journal of Steroid Biochemistry and Molecular Bio-logy vol 75 no 4-5 pp 229ndash236 2000
[46] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cor-tisol production and on activities of steroidogenic enzymesin human adrenocortical H295R cellsrdquo The Journal of SteroidBiochemistry and Molecular Biology vol 80 no 3 pp 355ndash3632002
[47] D P Dempsher D S Gann and R D Phair ldquoA mechanisticmodel of ACTH-stimulated cortisol secretionrdquo American Jour-nal of Physiology vol 246 no 4 part 2 pp R587ndashR596 1984
[48] N Sarai S Matsuoka and A Noma ldquosimBio a Java package forthe development of detailed cell modelsrdquo Progress in Biophysicsand Molecular Biology vol 90 no 1ndash3 pp 360ndash377 2006
[49] K Levenberg ldquoA method for the solution of certain non-linearproblems in least squaresrdquo Quarterly of Applied Mathematicsvol 2 pp 164ndash168 1944
[50] D W Marquardt ldquoAn algorithm for least-squares estimation ofnonlinear parametersrdquo SIAM Journal on Applied Mathematicsvol 11 no 2 pp 431ndash441 1963
[51] P E Gill and W Murray ldquoAlgorithms for the solution of thenonlinear least-squares problemrdquo SIAM Journal on NumericalAnalysis vol 15 no 5 pp 977ndash992 1978
[52] Y Akimoto R Hasada J Sakuma I Ono and S KobayashildquoGeneration alternation model for real-coded GA using multi-parent proposal and evaluation of just generation gap (JGG)rdquoin Proceedings of the 19th SICE Symposium on DecentralizedAutonomous Systems pp 341ndash346 Tokyo Japan January 2007
[53] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoProposal and evaluation of adaptive real-coded crossoverAREXrdquo Transactions of the Japanese Society for Artificial Intel-ligence vol 24 no 6 pp 446ndash458 2009
Journal of Toxicology 19
[54] S Kobayashi ldquoThe frontiers of real-coded genetic algorithmsrdquoJournal of Japanese Society for Artificial Intelligence vol 24 no1 pp 147ndash162 2009
[55] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoAnalysis of the behavior of MGG and JGG as a selection modelfor real-coded genetic algorithmsrdquo Transactions of the JapaneseSociety for Artificial Intelligence vol 25 no 2 pp 281ndash289 2010
[56] A Struthers H Krum and G H Williams ldquoA comparisonof the aldosterone-blocking agents eplerenone and spironolac-tonerdquo Clinical Cardiology vol 31 no 4 pp 153ndash158 2008
[57] K Yamasaki M Sawaki S Noda et al ldquoComparison of theHershberger assay and androgen receptor binding assay oftwelve chemicalsrdquo Toxicology vol 195 no 2-3 pp 177ndash1862004
[58] C Eil and S K Edelson ldquoThe use of human skin fibroblasts toobtain potency estimates of drug binding to androgen recept-orsrdquo The Journal of Clinical Endocrinology amp Metabolism vol59 no 1 pp 51ndash55 1984
[59] P Ye T Yamashita D M Pollock H Sasano and W ERainey ldquoContrasting effects of eplerenone and spironolac-tone on adrenal cell steroidogenesisrdquo Hormone and MetabolicResearch vol 41 no 1 pp 35ndash39 2009
[60] B D Albertson K L Frederick N C Maronian et al ldquoTheeffect of ketoconazole on steroidogenesis I Leydig cell enzymeactivity in vitrordquo Research Communications in Chemical Patho-logy and Pharmacology vol 61 no 1 pp 17ndash26 1988
[61] B D Albertson N C Maronian K L Frederick et al ldquoTheeffect of ketoconazole on steroidogenesis II Adrenocorticalenzyme activity in vitrordquoResearch Communications in ChemicalPathology and Pharmacology vol 61 no 1 pp 27ndash34 1988
[62] DM Stocco andB J Clark ldquoRegulation of the acute productionof steroids in steroidogenic cellsrdquo Endocrine Reviews vol 17 no3 pp 221ndash244 1996
[63] L K Christenson and J F Strauss III ldquoSteroidogenic acute regu-latory protein (StAR) and the intramitochondrial translocationof cholesterolrdquo Biochimica et Biophysica Acta (BBA)mdashMolecularand Cell Biology of Lipids vol 1529 no 1ndash3 pp 175ndash187 2000
[64] E R Simpson ldquoCholesterol side-chain cleavage cytochromeP450 and the control of steroidogenesisrdquo Molecular and Cel-lular Endocrinology vol 13 no 3 pp 213ndash227 1979
[65] H A LaVoie and S R King ldquoTranscriptional regulation of ster-oidogenic genes STARD1 CYP11A1 and HSD3Brdquo ExperimentalBiology and Medicine vol 234 no 8 pp 880ndash907 2009
[66] F Raucci A DrsquoAniello and M M Di Fiore ldquoStimulation ofandrogen production by d-aspartate through the enhancementof StAR P450scc and 3120573-HSD mRNA levels in vivo rat testisand in culture of immature rat Leydig cellsrdquo Steroids vol 84pp 103ndash110 2014
[67] M Doi Y Takahashi R Komatsu et al ldquoSalt-sensitive hyper-tension in circadian clock-deficient Cry-null mice involves dys-regulated adrenal Hsd3b6rdquo Nature Medicine vol 16 no 1 pp67ndash74 2010
[68] H Okamura M Doi Y Yamaguchi and J-M Fustin ldquoHyper-tension due to loss of clock novel insight from the molecu-lar analysis of Cry1Cry2-deleted micerdquo Current HypertensionReports vol 13 no 2 pp 103ndash108 2011
[69] M Doi F Satoh T Maekawa et al ldquoIsoform-specific mono-clonal antibodies against 3120573-hydroxysteroid dehydrogenaseisomerase family provide markers for subclassification ofhuman primary aldosteronismrdquo The Journal of Clinical Endo-crinology amp Metabolism vol 99 no 2 pp E257ndashE262 2014
[70] S Konosu-Fukaya Y Nakamura F Satoh et al ldquo3120573-hydroxy-steroid dehydrogenase isoforms in human aldosterone-produc-ing adenomardquo Molecular and Cellular Endocrinology vol 408pp 205ndash212 2015
[71] G C Verwoert J Hofland N Amin et al ldquoExpression and genevariation studies deny association of humanHSD3B1 gene withaldosterone production or blood pressurerdquoAmerican Journal ofHypertension vol 28 no 1 pp 113ndash120 2015
[72] D Hough K Storbeck SW P Cloete A C Swart and P SwartldquoRelative contribution of P450c17 towards the acute cortisolresponse lessons from sheep and goatsrdquoMolecular and CellularEndocrinology vol 408 pp 107ndash113 2015
[73] I Sakuma S Suematsu Y Matsuzawa et al ldquoCharacterizationof steroidogenic enzyme expression in aldosterone-producingadenoma a comparison with various human adrenal tumorsrdquoEndocrine Journal vol 60 no 3 pp 329ndash336 2013
[74] A Tong A Jia S Yan Y Zhang Y Xie and G Liu ldquoEctopiccortisol-producing adrenocortical adenoma in the renal hilumhistopathological features and steroidogenic enzyme profilerdquoInternational Journal of Clinical and Experimental Pathologyvol 7 no 7 pp 4415ndash4421 2014
[75] A Hirsch D Hahn P Kempna et al ldquoMetformin inhibitshuman androgen production by regulating steroidogenicenzymes HSD3B2 and CYP17A1 and complex I activity of therespiratory chainrdquo Endocrinology vol 153 no 9 pp 4354ndash43662012
[76] B Xu L Gao Y Cui et al ldquoSET protein up-regulated test-osterone production in the cultured preantral folliclesrdquo Repro-ductive Biology and Endocrinology vol 11 article 9 2013
[77] S Kikuchi D Tominaga M Arita M Takahashi and MTomita ldquoDynamic modeling of genetic networks using geneticalgorithm and S-systemrdquo Bioinformatics vol 19 no 5 pp 643ndash650 2003
[78] N Shikata YMaki M Nakatsui et al ldquoDetermining importantregulatory relations of amino acids from dynamic networkanalysis of plasma amino acidsrdquo Amino Acids vol 38 no 1 pp179ndash187 2010
[79] M Nakatsui T Ueda Y Maki I Ono and M OkamotoldquoMethod for inferring and extracting reliable genetic interac-tions from time-series profile of gene expressionrdquoMathematicalBiosciences vol 215 no 1 pp 105ndash114 2008
[80] A Komori Y Maki M Nakatsui I Ono and M OkamotoldquoEfficient numerical optimization algorithm based on new real-coded genetic algorithm AREX + JGG and application to theinverse problem in systems biologyrdquo Applied Mathematics vol3 no 10 pp 1463ndash1470 2012
[81] J A Roubos G van Straten and A J B van Boxtel ldquoAn evolu-tionary strategy for fed-batch bioreactor optimization conceptsand performancerdquo Journal of Biotechnology vol 67 no 2-3 pp173ndash187 1999
[82] D Sarkar and J M Modak ldquoGenetic algorithms with filters foroptimal control problems in fed-batch bioreactorsrdquo Bioprocessand Biosystems Engineering vol 26 no 5 pp 295ndash306 2004
[83] K Harada K Ikeda J Sakuma I Ono and S KobayashildquoHybridization of genetic algorithm with local search in mul-tiobjective function optimization recommendation of GA thenLSrdquoTransactions of the Japanese Society for Artificial Intelligencevol 21 no 6 pp 482ndash492 2006
Submit your manuscripts athttpwwwhindawicom
PainResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
ToxinsJournal of
VaccinesJournal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AntibioticsInternational Journal of
ToxicologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
StrokeResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Drug DeliveryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Pharmacological Sciences
Tropical MedicineJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AddictionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Emergency Medicine InternationalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Autoimmune Diseases
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anesthesiology Research and Practice
ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Pharmaceutics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
MEDIATORSINFLAMMATION
of
Journal of Toxicology 9
log2
ratio
4
3
2
1
0
minus1
minus2
minus3
minus4
ALDO
CORTICO
CORT
TESTO
DIONE
HPROG
DCORTICO
DCORT
PROG
PREG
HPREG
DHEA
DZ(
100)
SP(1
0)SP
(1)
DZ(
10)
DD
D(0
1)
AN
(01
)A
N(1
)A
N(1
0)SM
(00
1)TG
(00
1)M
P(0
1)FN
(01
)TG
(1)
AGT(
1)AG
T(0
1)SM
(00
0001
)D
DD
(1)
SM(0
1)
AGT(
10)
MP(
1)FN
(1)
MC(
1)D
Z(1)
KC(0
1)
MC(
01)
DZ(
01)
KC( 1
)D
DD
(10)
MP(
10)
MC(
10)
MP(
100)
AGT(
100)
DD
D(2
5)KC
(10)
Figure 3 Hierarchical cluster analysis of differential metabolic profiles of 12 steroid hormones by exposure to adrenal toxicity compoundsAdrenal toxicants were classified by using the differential metabolic profiling of 12 steroid hormones Concentrations of 12 adrenalsteroids secreted from NCI-H295R cells were drastically changed by the exposure of adrenal steroidogenesis inhibitors The 12 adrenalsteroid hormones were quantitatively measured by LC-MSMS simultaneously Four adrenal vasculotoxic compounds acrylonitrile ANfumaronitrile FN salinomycin SM and thioguanine TG were used as negative control compounds for adrenal steroidogenesis inhibitorsSeven adrenal steroidogenesis inhibitors aminoglutethimide AGT op1015840-DDD DDD spironolactone SP metyrapone MP ketoconazoleKC miconazole MC and daidzein DZ showed a characteristic steroid profile each and were classified as each independent singletonat the maximum exposure condition Exposure concentrations of adrenal toxic compounds were described in brackets of the samplename and the units were prepared as 120583M A blue cluster was classified as a group of nonchange samples including negative controlcompounds and low exposure conditions of adrenal steroidogenesis inhibitors PREG pregnenolone HPREG 17120572-hydroxypregnenoloneDHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-hydroxyprogesterone DIONE androstenedione DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICO corticosterone CORT cortisol ALDO aldosterone and TESTO testosterone
reliability of the experimental data because the ALDOconcentration was under the lower limit of quantification at100 120583MAGT Hecker and colleagues reported that 3 120583MAGTdecreases PREG and PROG concentrations and increases theTESTO concentration [10] However AGT did not increasethe TESTO concentration in our study One possibility isthat the concentration of TESTO was already enhanced byabout 33-fold through stimulation with ACTH forskolinand angiotensin II
333 DDD The mechanism of action of DDD in adrenalsteroidogenesis was estimated by dose-dependent inhibitionof CYP11A1 HSD3B2 CYP21A2 and CYP11B1 (estimatedinhibitions at 25120583M were 870 869 769 and 849resp) (Figure 5(f)) DDD has been reported to inhibitCYP11A1 HSD3B2 CYP21A2 CYP11B1 and CYP11B2 [29 3132] Inhibition of DDD by CYP11B2 was not estimated in ourstudy However the concentration of ALDO in the culturemedium decreased to 3 of that in the normal stimulatedcondition Inhibition of DDD by CYP11B2 has been shownusing mitochondrial and microsomal fractions prepared bystandard centrifugation procedures from a bovine adrenal
cortex homogenate [32]The cause of the discrepancy regard-ing the inhibition ofDDDbyCYP11B2 could not be explainedby same effect in the case of AGT
334 SP Themechanism of action of SP in adrenal steroido-genesis was estimated by inhibition of HSD3B2 CYP21A2and HSD17B3 (estimated inhibitions at 10120583M were 702595 and 593 resp) (Figure 5(g)) SP has been reportedto inhibit CYP17H CYP17L CYP11B1 and CYP11B2 [6 33ndash35] The inhibitory effect of SP on the HSD3B2 enzyme wasa novel mechanism of action The main action of SP is asa mineralocorticoid receptor (MR) antagonist SP has alsobeen reported to exert some off-target effects by binding toandrogen glucocorticoid and progesterone receptors [56ndash58] SP has been shown to inhibit the production of ALDOand CORT from PREG induced by angiotensin II in H295Rcells but the specific MR antagonist eplerenone did not showthe inhibitory effects [59] Therefore HSD3B2 inhibition bySP is not mediated via MR and the action might be directinhibition of HSD3B2 enzymes or a part of known off-targeteffects mediated through other nuclear hormone receptorsRegarding CYP17H and CYP17L our results were consistent
10 Journal of Toxicology
CHOL
0
20
40
60
80
100C
once
ntra
tion
in m
ediu
m (120583
M)
10 20 30 40 50 60 70 800Time (hours)
(a)
PREG
DCORT
DCORTICO
CORTICO
CORT
Con
cent
ratio
n in
med
ium
(120583M
)
10
4
2
0
8
6
10 20 30 40 50 60 70 800Time (hours)
(b)
Con
cent
ratio
n in
med
ium
(120583M
)
0
020406
081
1214
0 10 20 30 40 50 60 70 80Time (hours)
HPREG
DIONE
PROGHPROG
E1
(c)
DHEAE2
ALDO
TESTO
0
005
01
015
02
025
Con
cent
ratio
n in
med
ium
(120583M
)
0 10 20 30Time (hours)
40 50 60 70 80
(d)
CHOL
0
5000
10000
15000
20000
25000
30000
Con
cent
ratio
n in
cell
(120583M
)
10 20 30 40 50 60 70 800Time (hours)
(e)
Con
cent
ratio
n in
cell
(120583M
)
0
50
100
150
200
250
0 10 20 30 40 50 60 70 80Time (hours)
PREG
DCORT
DCORTICO
CORTICOCORT
(f)
0 10 20 30 40 50 60 70 80Time (hours)
HPREG
DIONE
PROGHPROG
E1
Con
cent
ratio
n in
cell
(120583M
)
3025201510
35
05
(g)
005
115
2
253
35
Con
cent
ratio
n in
cell
(120583M
)
0 10 20 30 40 50 60 70 80Time (hours)
DHEA
E2ALDO
TESTO
(h)
Figure 4 Comparison of time-course profiles of cholesterol and adrenal steroids produced by NCI-H295R cells between experimentallymeasured and simulated data To intuitively confirm reconstruction of the measured experimental data in the developed simulation modelof NCI-H295R cells dynamics of cholesterol and adrenal steroids produced by NCI-H295R cells were plotted to overlay experimentaldata with the simulated results Graphs show the dynamics of medium concentrations of cholesterol (a) and adrenal steroids ((b)ndash(d))and intracellular concentrations of cholesterol (e) and adrenal steroids ((f)ndash(h)) Steroid hormones were categorized into three groups byconcentration levels Major steroids were PREG pregnenolone DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICOcorticosterone and CORT cortisol ((b) and (f)) Moderate steroids were HPREG 17120572-hydroxypregnenolone PROG progesterone HPROG17120572-hydroxyprogesterone DIONE androstenedione and E1 estrone ((c) and (g)) Minor steroids were DHEA dehydroepiandrosteroneALDO aldosterone TESTO testosterone and E2 17120573-estradiol ((d) and (h)) Experimental data are shown as symbols with dotted lines Alldata represent the mean plusmn SD (119873 = 4) Simulation data are shown as solid lines
Journal of Toxicology 11
0110
10
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
AN (120583M)
(a)
0110
FN (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(b)
00101
SM (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
100E minus 05
(c)
0110
TG (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B1
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(d)
0110
10100
AGT (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(e)
0110
1025
DDD (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(f)
1010
SP (120583M)
2750005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(g)
0110
10100
MP (120583M)
2110005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(h)
Figure 5 Continued
12 Journal of Toxicology
0110
10
KC (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(i)
0110
10
MC (120583M)
4430005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(j)
0110
10100
DZ (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(k)
Figure 5 Estimated mechanism of action of adrenal toxicants by using the mathematical model of adrenal steroidogenesis in NCI-H295Rcells Mechanisms of action of adrenal toxicity compounds were quantitatively estimated from experimental results of differential steroidprofiling using the mathematical model of adrenal steroidogenesis in NCI-H295R cells The drug action was defined as a scaling factorof enzymatic activity in the simulation model These scaling factors were optimized to fit experimental data by a hybrid optimizationmethod of the RCGA and nonlinear least squares Estimated drug actions by the exposure of vasculotoxic agents acrylonitrile AN (a)fumaronitrile FN (b) salinomycin SM (c) and thioguanine TG (d) and the steroidogenic inhibitors aminoglutethimide AGT (e) op1015840-DDD DDD (f) spironolactone SP (g) metyrapone MP (h) ketoconazole KC (i) miconazole MC (j) and daidzein DZ (k) are shownas a spider radar chart CYP11A1 P450 side chain cleavage enzyme CYP17H 17120572-hydroxylase of CYP17 CYP17L C
17ndash20 lyase of CYP17HSD3B2 3120573-hydroxysteroid dehydrogenase CYP21A2 21-hydroxylase CYP11B1 11120573-hydroxylase CYP11B2 18-hydroxylase and HSD17B317120573-hydroxysteroid dehydrogenase
with previous reports [33 34] 7120572-thiospironolactone whichis synthesized by deacetylation of SP inhibits CYP17H andCYP17L [34] The fact that there were no inhibitions ofCYP17HorCYP17L in our study suggests that SPmight not bedeacetylated to 7120572-thiospironolactone in NCI-H295R cellsRegarding CYP11B1 and CYP11B2 our results were unclearcompared with a previous study It has been shown that30 120583M SP inhibits CYP11B1 and CYP11B2 in human andbovine adrenal mitochondria [35]The cause of CYP11B1 andCYP11B2 inhibition by SP could not be determined in ourstudy which might be due to the lower maximum exposureconcentration of SP than that in the previous reportWe couldnot examine SP concentrations over 10 120583M because theseconcentrations were cytotoxic in NCI-H295R cells
335 MP The mechanism of action of MP in adrenalsteroidogenesis was estimated by dose-dependent inhibitionof CYP11B1 (estimated inhibitions at 1 10 and 100 120583M were571 827 and 982 resp) (Figure 5(h)) MP has beenreported to inhibit CYP11B1 as its major effect and CYP11A1and CYP11B2 as a weak effect [6 29 36ndash39] The results wereable to show that MP is a selective inhibitor of CYP11B1 inthe previous report However the estimated effect of MPat a high concentration 100 120583M as the maximum exposureconcentration was unclear According to the previous reportselectivity of MP for CYP11B1CYP11B2 is about five times[39] In addition 20 120583M MP has a slight inhibition effect onCYP11A1 in H295R cells [29]
336 KC The mechanism of action of KC in adrenalsteroidogenesis was estimated by inhibition of CYP11A1CYP17H CYP17L HSD3B2 CYP21A2 CYP11B1 and
CYP11B2 (estimated inhibitions at 10 120583Mwere 926 943518 830 882 974 and 798 resp) (Figure 5(i))KC has been reported to inhibit CYP11A1 CYP17H CYP17LHSD3B2 CYP21A2 and CYP11B1 [6 29 40ndash43] Our resultswere almost consistent with the previous reports KC inhibitsCYP11A1 CYP17H CYP21A2 and CYP11B1 in NCI-H295Rcells at 10 120583M [29] and CYP17H CYP17L CYP21A2 andCYP11B1 in human adrenal mitochondria and Leydig cellmicrosomes at 2ndash5120583M [60 61] However KC has shownonly weak inhibition of HSD3B2 and HSD17B3 in Leydigcells at the millimolar level [60 61] Regarding CYP11B2 andHSD17B3 we considered that these estimated inhibitions ofKC did have sufficient reliability in terms of quantitative pre-diction precision because ALDO and TESTO concentrationswere less than the lower limit of quantification at 10 120583MKC
337 MC Themechanism of action of MC in adrenal stero-idogenesis was estimated by inhibition of CYP17H CYP17LCYP11B1 and HSD17B3 (estimated inhibitions at 10120583Mwere691 530 764 and 571 resp) (Figure 5(j)) MC hasbeen reported to inhibit not only CYP17H and CYP17L butalsoCYP11A1 CYP21A2 andCYP11B1 [41 44 45]The resultsin the previous reports were able to estimate that MC isa CYP17 inhibitor However CYP11A1 inhibition by MCprobably instead of a reduction in StAR expression wasnot clearly detected in our study using NCI-H295R cellsbecause there were no decreases in the concentrations ofPREG and PROG in the culture medium Indirect inhibitionof CYP11A1 via the peripheral-type benzodiazepine receptorhas been reported in mouse adrenocortical Y-1 cells treatedwith MC in the absence of stimuli by measuring PREGproduction [44] On the other hand reductions of StAR
Journal of Toxicology 13
protein expression andor transport activity without affectingtotal steroid synthesis or CYP11A1 and HSD3B2 enzymeexpression or activities have been reported in (BU)
2cAMP-
stimulated MA-10 Leydig tumor cells treated with MC bymeasuring PROG production [45] Therefore the effect ofMC on the initial reaction in adrenal steroidogenesis fromcholesterol should be different according to the cell type andstimulation condition Inhibition of CYP21A2 and CYP11B1by MC has been reported as decreases in the consumptionof PROG and DCORTICO respectively [41] Inhibition ofCYP11B1 was estimated by the action of MC in this studybut that of CYP21A2 was not detected In the previousexperimental report inhibitory sites by MCmight have beenreflected by inhibition of CYP17H activity because CYP21A2activity was measured as a decrease in labeled PROG
338 DZ The mechanism of action of DZ in adrenalsteroidogenesis was estimated by inhibition of CYP11A1HSD3B2 CYP21A2 CYP11B1 and HSD17B3 (estimated inhi-bitions at 100 120583M were 586 941 965 872 and981 resp) (Figure 5(k)) DZ has been reported to inhibitHSD3B2 and CYP21A2 [46] The results of HSD3B2 andCYP21A2 were consistent with the previous report Howeverinhibitions have not been reported for CYP11A1 CYP11B1and HSD17B3These estimated effects of DZ on CYP11B1 andHSD17B3 were unclear because the concentrations of ALDOand TESTO were less than the lower limit of quantificationat 100 120583M In addition these enzymes act downstream of thestrong action points of DZ such as HSD3B2 and CYP21A2
34 The Simulations and the Systematic Model Analysis
341 Dynamic Sensitivity Analysis of Adrenal SteroidogenesisTo comprehensively understand the dynamics of adrenalsteroidogenesis dynamic sensitivities were calculated forsteroid concentrations secreted by NCI-H295R cells usingour constructed mathematical model of steroidogenesis Theresults of dynamic sensitivity analysis at 72 h of duration and6 h of interval time are presented as a heat-map in Figure 6
The top 10 parameters of the total area under the curveof dynamic sensitivity for cholesterol and the 14 steroids inculture medium were 119881HSD3B2
maxA 119881CYP11A1max 119881CYP21A2maxA 119896loc119891 119896loc119887
119870CYP11A1119898
119881CYP17HmaxA 119896Cholesterol Transport 119896acc119891 and 119881CYP21A2maxB in
order from the top Cholesterol uptake (119896Cholesterol Transport)StAR protein (119896loc
119891) and CYP11A1 (119881CYP11A1max ) which are
determining factors of the capacity for steroidogenesis pro-moted the production of mineralocorticoids (DCORTICOCORTICO and ALDO) and restrained the synthesis of glu-cocorticoids (DCORT and CORT) and sex steroids (DIONETESTO and E1) because of the accumulation of interme-diate molecules in steroidogenesis (PREG HPREG DHEAPROG and HPROG) only by self-activation The dynamicpatterns of the intermediate molecules in steroidogenesiswere mainly dependent on the activity of CYP17H andHSD3B2 with PREG as the substrate of these enzymes inwhich the dynamic sensitivities of 119881CYP17HmaxA for HPREG andHPROG and 119881HSD3B2
maxA for PROG HPROG and DCORTICO
reversed the direction of sensitivity at 49ndash66 h after stimu-lation The dynamic sensitivities of the maximum activitiesof HSD3B2 for PREG (119881HSD3B2
maxA ) and CYP21A2 for PROG(119881CYP21A2maxA ) were related to all steroids at 72 h Almost allmodel parameters had positive sensitivity for downstreamsteroids in the adrenal steroidogenic pathway and negativesensitivity for direct-binding steroids as substrates of thesteroidogenic enzyme The sensitivity of 119881max in all steroido-genic enzymes was relatively higher than 119870
119898for the same
steroid substrate
342 Simulation of the Metabolic Balance of AdrenalSteroidogenesis Pathway To clearly show the property of themetabolic shift between mineralocorticoid and glucocorti-coid biosynthesis we performed two-dimensional parameterscanning of the enzymatic activities of CYP17H and HSD3B2(Figure 7) NCI-H295R cells lost the ability to produce allsteroid hormones when enzymatic activities of CYP17H andHSD3B2 were changed by over 60 and 30 respectivelyActivation ofCYP17HandorHSD3B2 induced themetabolicshift that enhanced the glucocorticoid biosynthesis and devi-ated from the mineralocorticoid biosynthesis On the otherhand inhibition of CYP17H andor HSD3B2 induced themetabolic shift that enhanced the mineralocorticoid biosyn-thesis and deviated from the glucocorticoid biosynthesisMoreover the enzymatic activity of HSD3B2 regulated themetabolic balance of sex steroids and the precursors onadrenal steroidogenesis of NCI-H295R cells E1 TESTO andDIONE were produced by NCI-H295R cells when activatingthe enzymatic activity of HSD3B2 Conversely E2 andDHEAwere produced by NCI-H295R cells when suppressing theenzymatic activity of HSD3B2 The biosynthesis of down-stream steroids in adrenal steroidogenesis pathway such asmineralocorticoids and glucocorticoids was almost com-pletely terminated when the enzymatic activity of HSD3B2was decreased by over 80
4 Discussion
41 Importance of 3120573-HSDActivity in Adrenal SteroidogenesisOur systematic analysis using the mathematical model ofadrenal steroidogenesis in NCI-H295R cells revealed thatthe enzymatic activity of 3120573-HSD controls the dynamics ofadrenal steroidogenesis The activity of the StAR proteincontrols the net capacity of steroidogenesis in steroidogeniccells which is the transport of cholesterol from the outer toinner mitochondrial membranes Both the expression levelsof StAR protein and mRNA are rapidly elevated in responseto stimulation by tropic hormones such as ACTH [62 63]Another important factor in adrenal steroidogenesisis is thecholesterol side chain cleavage enzyme CYP11A1 the firstrate-limiting and hormonally regulated step in the synthesisof all steroid hormones which is conversion of cholesterolto pregnenolone in mitochondria [64] According to ourresults of global sensitivity analysis (Supplementary Table 1and Figure 6(d)) in addition to CYP11A1 and StAR proteins3120573-HSD was one of the key regulators in adrenal steroido-genesis of NCI-H295R cells And also this result suggeststhat a significant regulatory mechanism in steroidogenesis
14 Journal of Toxicology
60544830 4236 66 7218 246 120
Time (hours)
0
1
2
3C
once
ntra
tion
(a)
60544830 4236 66 7218 246 120
Time (hours)
minus05
0
05
minus1
1
Sens
itivi
ty
(b)
0
6
12
18
24
30
36
42
48
54
60
66
72
Sens
itivi
ty fo
r med
ium
conc
entr
atio
ns
of ch
oles
tero
l or s
tero
id h
orm
ones
Dur
atio
n tim
e (ho
urs)
Model parameters
+10
+05
00
minus05
minus10
(c)
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
Parameter number in the mathematical model of steroidogenesis in NCI-H295R cells
CHOL
PREG
HPREG
DHEA
PROG
HPROG
DIONE
DCORTICO
DCORT
CORTICO
CORT
ALDO
TESTO
E1
E2
Chol
este
rol o
r ste
roid
hor
mon
es
0
05
10
15
20
minus05
minus10
minus15
minus20
(d)
Figure 6 Heat-map of the global dynamic sensitivity analysis of adrenal steroid concentrations produced by NCI-H295R cells The globaldynamic sensitivity analysis is a powerful tool to comprehensively understand the dependencies of themodel parameters in themathematicalmodel of the biological complex system Dynamic sensitivities of model parameters in the mathematical model of adrenal steroidogenesisin NCI-H295R cells were calculated for all steroid concentrations in the culture medium every 6 h until 72 h after stimulation To clarify theview of heat-map of global dynamic sensitivity analysis imaginary data of the dynamics of steroid concentrations in the original model (blueline) and perturbed model for sensitivity analysis (red line) were prepared (a) Using this imaginary data the calculated dynamic sensitivities(b) and the visualized dynamic sensitivity as one block of the heat-map (c) were shown respectively By the same method that explainedusing imaginary data the large-scale data of the global dynamic sensitivity analysis on the mathematical model of adrenal steroidogenesisin NCI-H295R cells was comprehensively visualized as a big graph of heat-map (d) Parameter numbers in the horizontal axis are (1)119896Choresterol Transport (2) 119896CEH (3) 119896MTR
119891 (4) 119896MTR
119887 (5) 119896acc
119891 (6) 119896acc
119887 (7) 119896loc
119891 (8) 119896loc
119887 (9) 119896Oxysterol Synthesis (10)119870
CYP11A1119898
(11) 119881CYP11A1max (12)119870CYP17H119898A
(13) 119870CYP17H119898B (14) 119881CYP17HmaxA (15) 119881CYP17HmaxB (16) 119881CYP17HmaxB (17) 119870CYP17L
119898B (18) 119881CYP17LmaxA (19) 119881CYP17LmaxB (20) 119870HSD3B2119898A (21) 119870HSD3B2
119898B (22) 119870HSD3B2119898C (23)
119881HSD3B2maxA (24) 119881HSD3B2
maxB (25) 119881HSD3B2maxB (26) 119870CYP21A2
119898A (27) 119870CYP21A2119898B (28) 119881CYP21A2maxA (29) 119881CYP21A2maxB (30) 119870CYP11B1
119898A (31) 119870CYP11B1119898B (32) 119881CYP11B1maxA
(33) 119881CYP11B1maxB (34) 119896CYP11B2 (35) 119870HSD17B3119898A (36) 119870HSD17B3
119898B (37) 119881HSD17B3maxA (38) 119881HSD17B3
maxB (39) 119870CYP19A1119898A (40) 119870CYP19A1
119898B (41) 119881CYP19A1maxA and (42)119881
CYP19A1maxB PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-
hydroxyprogesteroneDIONE androstenedioneDCORTICO 11-deoxycorticosteroneDCORT 11-deoxycortisol CORTICO corticosteroneCORT cortisol ALDO aldosterone and TESTO testosterone
pathway is very reasonable StAR CYP11A1 and 3120573-HSD(isoforms 1 or 2 in humans) proteins generally respond to thesame hormones that stimulate steroid production throughcommon pathways such as cAMP signaling in adrenal glandsand testes [65 66] Moreover our data also support recentexperimental evidence from clinical and in vivo studiessuggesting that the enzymatic activity of 3120573-HSD plays
an important role in the regulation of mineralocorticoidsynthesis in adrenal steroidogenesis and contributes to hyper-tension caused by abnormal overproduction of aldosterone[67ndash70] Circadian clock-deficient Cry-null mice show salt-sensitive hypertension due to abnormally high synthesis ofaldosterone which is caused by constitutively high expressionof HSD3B6mRNA and protein in the adrenal cortex [67 68]
Journal of Toxicology 15
All steroid producing cells
Mineralocorticoid producing cells Glucocorticoid producing cells
Mineralocorticoid and glucocorticoid producing cells
Terminated steroidogenesis
Sex steroids and precursor sex steroidsDIONE TESTO E1DEHA E2
NCI-H295Rcells at controlcondition
Normalized activities of HSD3B2
Nor
mal
ized
activ
ities
of C
YP17
H
2
18
16
14
12
1
08
06
04
02
00 05 1 15 2
Figure 7 Metabolic categories of the steroidogenic cells determined by the balance of HSD3B2 and CYP17H activitiesThe two-dimensionalparameter scanning analysis by the perturbation of two focused parameters clarifies the interaction and the relationship between the twoparameters in the complex system Functional cellular categories of steroidogenic cells were defined by the levels of mineralocorticoid(ALDO) glucocorticoids (DCORTandCORT) and androgens (DHEAandDIONE) at 72 h after stimulation Enzymatic activities ofHSD3B2andCYP17were normalized by standard values of the simulationmodel inNCI-H295R cells Scanning ranges ofHSD3B2 andCYP17 activitieswere 0ndash200 each 10 Green regions are all 14 steroids produced by NCI-H295R cells Red blue and purple regions are mineralocorticoidproducing cells glucocorticoids-producing cells and both corticoid-producing cells respectively Yellow regions are steroidogenesis of NCI-H295R cells terminated upstream of the adrenal steroidogenesis pathway
Recent clinical observations have revealed predominantexpression of HSD3B2 mRNA and protein in tumor cellsof aldosterone-producing adenoma (APA) and HSD3B1mRNA significantly correlated with CYP11B2 mRNA levelsand plasma aldosterone concentrations in APA patients [6970] However the relationship is unclear and disputed ina small-scale clinical study indicating that genetic variationin HSD3B1 affects blood pressure and hypertension in APApatients [71] The results of our simulation study suggest that3120573-HSD protein (human genes are HSD3B1 and HSD3B2) isone of the determination factors for the dynamic propertyof adrenal steroidogenesis Our results support the clinicalevidence of Doi and colleagues [69] and we believe that theHSD3B1 enzyme has a promising potential as novel drugtarget for endocrine hypertension
42 Metabolic Shift of Adrenal Steroidogenesis and the Contri-butions of HSD3B2 and CYP17H The metabolic propertiesof adrenal steroidogenesis in NCI-H295R cells were revealedby dynamic sensitivity analysis using themathematicalmodel(Figure 6) Mineralocorticoids such as DCORTICO COR-TICO and ALDO and intermediate steroids upstream of
the adrenal steroidogenesis pathway such as PREG HPREGDHEA PROG and HPROG were accelerated by reactionsof cholesterol import (119896Cholesterol Transport) StAR protein (119896MTR
119891
and 119896loc119891) and CYP11A1 (119881CYP11A1max ) On the other hand glu-
cocorticoids such as DCORT and CORT and sex hormonessuch as DIONE TESTO and E1 were suppressed by thesemodel parametersTherefore enhancement of the net adrenalsteroidogenesis capacity which supplies PREG precursor tothe pathway causes a production shift from glucocorticoidsto mineralocorticoids by substrate inhibitions of CYP17HHSD3B2 and CYP21A2 caused by accumulation of initialintermediate steroids such as PREG and PROG Sensitivi-ties of CYP17H (119881CYP17HmaxA ) and HSD3B2 (119881HSD3B2
maxA ) for theproducts were dynamically changed and these parametersdetermined the metabolic balance of downstream steroidsin the adrenal steroidogenesis pathway According to theseresults of dynamic sensitivity analysis of StAR CYP11A1CYP17H and HSD3B2 we suggest that the enhancementof CYP17H and HSD3B2 activity during ACTH stimulationwas important to shift the steroidogenic output away fromALDO biosynthesis towards CORT biosynthesis as wellas adrenal androgen production This suggestion partially
16 Journal of Toxicology
supports a comparative animal study in which molecularand cellular variations in CYP17H activity dramatically affectacute cortisol production resulting in distinct physiologicaland behavioral responses [72]
Results of two-dimensional parameter scanning of theenzymatic activities of CYP17H and HSD3B2 quantitativelyshowed the detail of the metabolic relationship betweenmin-eralocorticoid and glucocorticoid biosynthesis (Figure 7)Particularly the results showed that the balance of theseenzymatic activities was very important for the typicalfunction of NCI-H295R cells namely the ability to produceall steroid hormones NCI-H295R cells lost this functionwhen enzymatic activities of CYP17H and HSD3B2 werechanged by over 60 and 30 respectively In addition theybecame mineralocorticoid (ALDO) secreting cells when theenzymatic activity of CYP17H or HSD3B2 was inhibited byover 50 or glucocorticoid (DCORT and CORT) secretingcells when these enzymes were activated by over 50 Inparticular this analysis also showed that HSD3B2 was a keyplayer in the adrenal steroidogenesis of NCI-H295R cellsbecause HSD3B2 inhibition by over 80 almost completelyinhibited the biosynthesis of downstream steroids The ratioof CYP17A1 to HSD3B2 mRNA expression levels has beenrelated to several endocrine diseases with a low level in APAs[73] and high level in cortisol-producing adenomas [74]Furthermore the expression levels or enzymatic activitiesof CYP17A1 and HSD3B1 have been related to androgenproduction in polycystic ovary syndrome [75 76] Theseclinical studies support our simulation results indicating thatthe balance of enzymatic activity of CYP17H and HSD3B2determines the shift in steroidogenic output to mineralocor-ticoids glucocorticoids or androgens
43 Methodologies of Quantitative Mechanistic Analysis forDrug Discovery According to our results obtained using themathematical model of steroidogenesis in NCI-H295R cellssuch as sensitivity analysis comprehensive analysis based onsystems biology is available to quantitatively estimate themechanism of action of steroidogenic disrupting compoundsfrom differential profiling of adrenal steroid hormonesbecause dynamic patterns of steroid hormones in adrenalsteroidogenesis pathway are highly complex Our proposedmethod of quantitative mechanistic analysis of steroidogenicinhibitors was able to predict known action sites in theadrenal steroidogenesis pathway at only one time point (72 hafter drug exposure) Moreover according to the resultsof sensitivity analysis (Figure 6) 119881max of all steroidogenicenzymes was more sensitive than the 119870
119898 because the
intracellular concentrations of steroid hormones were almostmaintained at sufficiently high levels compared with 119870
119898
values of steroidogenic enzymesThese results suggested thatestimation of the mechanism of action of drugs is moreeffective and detectable when using the influences of 119881maxas the searching parameters such as our proposed methodOur data showed that the proposed method based on asystems biology model is a very powerful tool for exploratoryscreening of steroidogenic disrupting compounds
RCGA as a solver of parameter estimation problemsin systems biology has been applied to biological network
identification of gene regulatory networks and metabolicpathways and optimization of biological processes usingexperimentally observed time-course data [77ndash82] In thisstudy RCGA was useful to estimate the mechanism ofaction of novel pharmaceutical drug candidates for adrenalsteroidogenesis as a new application of RCGA in systemsbiology We had two issues when applying RCGA to thequantitative mechanistic analysis of drug actions Theseissues were the vast calculation cost and multimodality ofquasi-optimum solutions in solving the optimization prob-lem because the mathematical model in systems biologyconsists of many equations and parameters A proposedoptimization strategy using RCGA based on REXJGG wasa highly stable and efficient calculation method for a betterquasi-optimum solution than the unimodal normal distri-bution crossover (UNDX)minimum generation gap (MGG)method that is well applied in the engineering field In addi-tion we expanded the RCGA optimization program basedon REXJGG to a hybrid method of GA and then applieda local search as recommended by Harada and Kobayashi[28 83] A final optimal solution was obtained with a goodconvergence property Because these problems are generalin systems biology studies we suggest that the proposedhybrid method based on REXJGG is a very useful tool forquantitative mechanistic analysis of novel pharmaceuticaldrugs not limited to steroidogenic disrupting compounds
5 Conclusions
Thenovelmathematicalmodel of adrenal steroidogenesis wasconstructed in this study including cholesterol transport anddistribution the C
21-steroid hormone pathway steroid trans-
port and cell proliferation which could reproduce adrenalsteroidogenesis in NCI-H295R cells According to the resultsof dynamic sensitivity analysis using the newmodel HSD3B2plays the most important role in the metabolic balance ofadrenal steroidogenesis in NCI-H295R cells Moreover toquantitatively estimate mechanisms of action of adrenal toxiccompounds we analyzed differential metabolic profiles of12 steroid hormones at 3 days after exposure to 11 adrenaltoxic compounds by using the new mathematical modeland a hybrid optimization method of the RCGA and a localsearch (nonlinear least squares) We could estimate whichsteroidogenic enzymes were affected by these compoundsusing the hybrid optimization method Vasculotoxic agentswere estimated to have no effect according to the resultsobtained by our method In terms of adrenal steroidogenicinhibitors the predicted action sites were approximatelymatched to the target enzymes as reported in the literatureThus our computer-aided method based on a systems biol-ogy approach may be useful to analyze the mechanism ofaction of endocrine-disrupting compounds and to assess thehuman adrenal toxicity of novel pharmaceutical drugs basedon steroid hormone profiling
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Journal of Toxicology 17
Authorsrsquo Contribution
Ryuta Saito Natsuko Terasaki and Naohisa Tsutsui designedthe study Naohisa Tsutsui and Naoya Masutomi supervisedthe experiments Natsuko Terasaki performed all in vitroexperiments and collected dataMakotoYamazaki performedall measurements of steroids Ryuta Saito conceived the insilico strategies analyzed data developed all programs andthe mathematical model and performed all simulation anal-yses with Masahiro Okamoto Ryuta Saito Natsuko TerasakiMakoto Yamazaki NaoyaMasutomi andMasahiroOkamotowrote the paper
Acknowledgments
The authors would like to thank Dr Michael S Breenand Dr Miyuki Breen for helpful discussions concerningprevious mathematical models of adrenal steroidogenesisThis work was partially supported by the Grant-in-Aid forScientific Research on Innovative Areas ldquoSynthetic Biologyrdquo(No 23119001 (MO)) from Ministry of Education CultureSports Science and Technology in Japan
References
[1] P W Harvey D J Everett and C J Springall ldquoAdrenal toxi-cology a strategy for assessment of functional toxicity to theadrenal cortex and steroidogenesisrdquo Journal of Applied Toxi-cology vol 27 no 2 pp 103ndash115 2007
[2] W E Ribelin ldquoThe effects of drugs and chemicals upon thestructure of the adrenal glandrdquo Fundamental and Applied Toxi-cology vol 4 no 1 pp 105ndash119 1984
[3] H D Colby and P A Longhurst ldquoToxicology of the adrenalglandrdquo in Endocrine Toxicology C K Atterwill and J D FlackEds pp 243ndash281 CambridgeUniversity Press Cambridge UK1992
[4] H D Colby ldquoIn vitro assessment of adrenocortical toxicityrdquoJournal of Pharmacological and Toxicological Methods vol 32no 1 pp 1ndash6 1994
[5] T J Rosol J T Yarrington J Latendresse and C C CapenldquoAdrenal gland structure function and mechanisms of toxic-ityrdquo Toxicologic Pathology vol 29 no 1 pp 41ndash48 2001
[6] P W Harvey and D J Everett ldquoThe adrenal cortex and ster-oidogenesis as cellular and molecular targets for toxicity crit-ical omissions from regulatory endocrine disrupter screeningstrategies for human healthrdquo Journal of Applied Toxicology vol23 no 2 pp 81ndash87 2003
[7] A F Gazdar H K Oie C H Shackleton et al ldquoEstablishmentand characterization of a human adrenocortical carcinoma cellline that expresses multiple pathways of steroid biosynthesisrdquoCancer Research vol 50 no 17 pp 5488ndash5496 1990
[8] W E Rainey I M Bird C Sawetawan et al ldquoRegulation ofhuman adrenal carcinoma cell (NCI-H295) production of C19steroidsrdquo The Journal of Clinical Endocrinology amp Metabolismvol 77 no 3 pp 731ndash737 1993
[9] B Staels D W Hum andW L Miller ldquoRegulation of steroido-genesis in NCI-H295 cells a cellular model of the human fetaladrenalrdquo Molecular Endocrinology vol 7 no 3 pp 423ndash4331993
[10] M Hecker J L Newsted M B Murphy et al ldquoHuman adreno-carcinoma (H295R) cells for rapid in vitro determination of
effects on steroidogenesis hormone productionrdquo Toxicologyand Applied Pharmacology vol 217 no 1 pp 114ndash124 2006
[11] A Oskarsson E Ulleras K E Plant J P Hinson and P SGoldfarb ldquoSteroidogenic gene expression in H295R cells andthe human adrenal gland adrenotoxic effects of lindane invitrordquo Journal of Applied Toxicology vol 26 no 6 pp 484ndash4922006
[12] UMuller-Vieira M Angotti and RW Hartman ldquoThe adreno-cortical tumor cell line NCI-H295R as an in vitro screeningsystem for the evaluation of CYP11B2 (aldosterone synthase)and CYP11B1 (steroid-11120573-hydroxylase) inhibitorsrdquoThe Journalof Steroid Biochemistry and Molecular Biology vol 96 no 3-4pp 259ndash270 2005
[13] K Imagawa S Okayama M Takaoka et al ldquoInhibitory effectof efonidipine on aldosterone synthesis and secretion in humanadrenocarcinoma (H295R) cellsrdquo Journal of CardiovascularPharmacology vol 47 no 1 pp 133ndash138 2006
[14] M Voets UMuller-Vieira S Marchais-Oberwinkler and RWHartmann ldquoSynthesis of amidinohydrazones and evaluation oftheir inhibitory effect towards aldosterone synthase (CYP11B2)and the formation of selected steroidsrdquo Archiv der Pharmazievol 337 no 7 pp 411ndash416 2004
[15] F K Nielsen C H Hansen J A Fey et al ldquoH295R cells as amodel for steroidogenic disruption a broader perspective usingsimultaneous chemical analysis of 7 key steroid hormonesrdquoToxicology in Vitro vol 26 no 2 pp 343ndash350 2012
[16] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cortisolproduction and on activities of steroidogenic enzymes inhuman adrenocortical H295R cellsrdquo Journal of Steroid Biochem-istry and Molecular Biology vol 80 no 3 pp 355ndash363 2002
[17] R F Canton J T Sanderson S Nijmeijer A Bergman R JLetcher and M van den Berg ldquoIn vitro effects of brominatedflame retardants and metabolites on CYP17 catalytic activity anovel mechanism of actionrdquo Toxicology and Applied Pharma-cology vol 216 no 2 pp 274ndash281 2006
[18] K Hilscherova P D Jones T Gracia et al ldquoAssessment ofthe effects of chemicals on the expression of ten steroidogenicgenes in the H295R cell line using real-time PCRrdquo ToxicologicalSciences vol 81 no 1 pp 78ndash89 2004
[19] T Gracia K Hilscherova P D Jones et al ldquoThe H295R systemfor evaluation of endocrine-disrupting effectsrdquo Ecotoxicologyand Environmental Safety vol 65 no 3 pp 293ndash305 2006
[20] X Zhang J L NewstedM Hecker E B Higley P D Jones andJ P Giesy ldquoClassification of chemicals based on concentration-dependent toxicological data using ToxClustrdquo EnvironmentalScience and Technology vol 43 no 10 pp 3926ndash3932 2009
[21] M S Breen M Breen N Terasaki M Yamazaki and R BConolly ldquoComputational model of steroidogenesis in humanH295R cells to predict biochemical response to endocrine-active chemicalsmodel development formetyraponerdquoEnviron-mental Health Perspectives vol 118 no 2 pp 265ndash272 2010
[22] M Breen M S Breen N Terasaki M Yamazaki A LLloyd and R B Conolly ldquoMechanistic computational model ofsteroidogenesis in H295R cells role of oxysterols and cell pro-liferation to improve predictability of biochemical response toendocrine active chemical-metyraponerdquo Toxicological Sciencesvol 123 no 1 pp 80ndash93 2011
[23] F B Kraemer ldquoAdrenal cholesterol utilizationrdquo Molecular andCellular Endocrinology vol 265-266 pp 42ndash45 2007
[24] W L Miller ldquoSteroidogenic acute regulatory protein (StAR)a novel mitochondrial cholesterol transporterrdquo Biochimica et
18 Journal of Toxicology
Biophysica Acta (BBA)mdashMolecular and Cell Biology of Lipidsvol 1771 no 6 pp 663ndash676 2007
[25] S Szabo and I T Lippe ldquoAdrenal gland chemically inducedstructural and functional changes in the cortexrdquo ToxicologicPathology vol 17 no 2 pp 317ndash329 1989
[26] C Chen-Pan I-J Pan Y Yamamoto T Sakogawa J Yamadaand YHayashi ldquoPrompt recovery of damaged adrenalmedullaeinduced by salinomycinrdquoToxicologic Pathology vol 27 no 5 pp563ndash572 1999
[27] R J Santen and R I Misbin ldquoAminoglutethimide review ofpharmacology and clinical userdquo Pharmacotherapy vol 1 no 2pp 95ndash120 1981
[28] M J Carella N V Dimitrov V V Gossain L Srivastava andD R Rovner ldquoAdrenal effects of low-dose aminoglutethimidewhen used alone in postmenopausal women with advancedbreast cancerrdquoMetabolism vol 43 no 6 pp 723ndash727 1994
[29] M K Johansson J T Sanderson and B-O Lund ldquoEffects of3-MeSO2-DDE and some CYP inhibitors on glucocorticoidsteroidogenesis in the H295R human adrenocortical carcinomacell linerdquo Toxicology in Vitro vol 16 no 2 pp 113ndash121 2002
[30] Y Touitou A Bogdan J C Legrand and P Desgrez ldquoAmino-glutethimide and glutethimide effects on 18-hydroxycorti-costerone biosynthesis by human and sheep adrenals in vitrordquoActa Endocrinologica vol 80 no 3 pp 517ndash526 1975
[31] M M Hart and J A Straw ldquoStudies on the site of action ofop1015840-DDD in the dog adrenal cortex 1 Inhibition of ACTH-mediated pregnenolone synthesisrdquo Steroids vol 17 no 1ndash5 pp559ndash574 1971
[32] M Ojima M Saitoh N Itoh Y Kusano S Fukuchi and HNaganuma ldquoThe effects of op1015840-DDD on adrenal steroidogen-esis and hepatic steroid metabolismrdquo Nippon Naibunpi Gakkaizasshi vol 61 no 3 pp 168ndash178 1985
[33] L F Canosa and N R Ceballos ldquoEffects of different steroid-biosynthesis inhibitors on the testicular steroidogenesis of thetoad Bufo arenarumrdquo Journal of Comparative Physiology B vol171 no 6 pp 519ndash526 2001
[34] D C Kossor S Kominami S Takemori and H D ColbyldquoRole of the steroid 17 alpha-hydroxylase in spironolactone-mediated destruction of adrenal cytochrome P-450rdquoMolecularPharmacology vol 40 no 2 pp 321ndash325 1991
[35] S C Cheng K Suzuki W Sadee and B W Harding ldquoEffectsof spironolactone canrenone and canrenoate K on cytochromeP450 and 11120573 and 18 hydroxylation in bovine and humanadrenal corticalmitochondriardquo Endocrinology vol 99 no 4 pp1097ndash1106 1976
[36] K Yanagibashi M Haniu J E Shively W H Shen and P HallldquoThe synthesis of aldosterone by the adrenal cortex Two zones(fasciculata and glomerulosa) possess one enzyme for 11120573- 18-hydroxylation and aldehyde synthesisrdquoThe Journal of BiologicalChemistry vol 261 no 8 pp 3556ndash3562 1986
[37] K Morishita H Okumura N Ito and N Takahashi ldquoPrimaryculture system of adrenocortical cells from dogs to evaluatedirect effects of chemicals on steroidogenesisrdquo Toxicology vol165 no 2-3 pp 171ndash178 2001
[38] S W J Lamberts E G Bons H A Bruining and F H de JongldquoDifferential effects of the imidazole derivatives etomidateketoconazole and miconazole and of metyrapone on the secre-tion of cortisol and its precursors by human adrenocorticalcellsrdquo Journal of Pharmacology and Experimental Therapeuticsvol 240 no 1 pp 259ndash264 1987
[39] L Yin S Lucas F Maurer U Kazmaier Q Hu and RW Hartmann ldquoNovel imidazol-1-ylmethyl substituted 1256-tetrahydropyrrolo[321-ij]quinolin-4-ones as potent and selec-tive CYP11B1 inhibitors for the treatment of Cushingrsquos syn-dromerdquo Journal ofMedicinal Chemistry vol 55 no 14 pp 6629ndash6633 2012
[40] M DiMattina N Maronian H Ashby D L Loriaux andB D Albertson ldquoKetoconazole inhibits multiple steroidogenicenzymes involved in androgen biosynthesis in the humanovaryrdquo Fertility and Sterility vol 49 no 1 pp 62ndash65 1988
[41] M Ayub and M J Levell ldquoInhibition of human adrenal steroi-dogenic enzymes in vitro by imidazole drugs including keto-conazolerdquo Journal of Steroid Biochemistry vol 32 no 4 pp 515ndash524 1989
[42] D Engelhardt M M Weber T Miksch F Abedinpour andC Jaspers ldquoThe influence of ketoconazole on human adrenalsteroidogenesis incubation studies with tissue slicesrdquo ClinicalEndocrinology vol 35 no 2 pp 163ndash168 1991
[43] S Bhasin S Sikka T Fielder et al ldquoHormonal effects of keto-conazole in vivo in the male rat mechanism of actionrdquo Endo-crinology vol 118 no 3 pp 1229ndash1232 1986
[44] D M Zisterer and D C Williams ldquoCalmidazolium and otherimidazole compounds affect steroidogenesis in Y1 cells lack ofinvolvement of the peripheral-type benzodiazepine receptorrdquoJournal of Steroid Biochemistry and Molecular Biology vol 60no 3-4 pp 189ndash195 1997
[45] L P Walsh C N Kuratko and D M Stocco ldquoEconazoleand miconazole inhibit steroidogenesis and disrupt steroido-genic acute regulatory (StAR) protein expression post-trans-criptionallyrdquo Journal of Steroid Biochemistry and Molecular Bio-logy vol 75 no 4-5 pp 229ndash236 2000
[46] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cor-tisol production and on activities of steroidogenic enzymesin human adrenocortical H295R cellsrdquo The Journal of SteroidBiochemistry and Molecular Biology vol 80 no 3 pp 355ndash3632002
[47] D P Dempsher D S Gann and R D Phair ldquoA mechanisticmodel of ACTH-stimulated cortisol secretionrdquo American Jour-nal of Physiology vol 246 no 4 part 2 pp R587ndashR596 1984
[48] N Sarai S Matsuoka and A Noma ldquosimBio a Java package forthe development of detailed cell modelsrdquo Progress in Biophysicsand Molecular Biology vol 90 no 1ndash3 pp 360ndash377 2006
[49] K Levenberg ldquoA method for the solution of certain non-linearproblems in least squaresrdquo Quarterly of Applied Mathematicsvol 2 pp 164ndash168 1944
[50] D W Marquardt ldquoAn algorithm for least-squares estimation ofnonlinear parametersrdquo SIAM Journal on Applied Mathematicsvol 11 no 2 pp 431ndash441 1963
[51] P E Gill and W Murray ldquoAlgorithms for the solution of thenonlinear least-squares problemrdquo SIAM Journal on NumericalAnalysis vol 15 no 5 pp 977ndash992 1978
[52] Y Akimoto R Hasada J Sakuma I Ono and S KobayashildquoGeneration alternation model for real-coded GA using multi-parent proposal and evaluation of just generation gap (JGG)rdquoin Proceedings of the 19th SICE Symposium on DecentralizedAutonomous Systems pp 341ndash346 Tokyo Japan January 2007
[53] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoProposal and evaluation of adaptive real-coded crossoverAREXrdquo Transactions of the Japanese Society for Artificial Intel-ligence vol 24 no 6 pp 446ndash458 2009
Journal of Toxicology 19
[54] S Kobayashi ldquoThe frontiers of real-coded genetic algorithmsrdquoJournal of Japanese Society for Artificial Intelligence vol 24 no1 pp 147ndash162 2009
[55] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoAnalysis of the behavior of MGG and JGG as a selection modelfor real-coded genetic algorithmsrdquo Transactions of the JapaneseSociety for Artificial Intelligence vol 25 no 2 pp 281ndash289 2010
[56] A Struthers H Krum and G H Williams ldquoA comparisonof the aldosterone-blocking agents eplerenone and spironolac-tonerdquo Clinical Cardiology vol 31 no 4 pp 153ndash158 2008
[57] K Yamasaki M Sawaki S Noda et al ldquoComparison of theHershberger assay and androgen receptor binding assay oftwelve chemicalsrdquo Toxicology vol 195 no 2-3 pp 177ndash1862004
[58] C Eil and S K Edelson ldquoThe use of human skin fibroblasts toobtain potency estimates of drug binding to androgen recept-orsrdquo The Journal of Clinical Endocrinology amp Metabolism vol59 no 1 pp 51ndash55 1984
[59] P Ye T Yamashita D M Pollock H Sasano and W ERainey ldquoContrasting effects of eplerenone and spironolac-tone on adrenal cell steroidogenesisrdquo Hormone and MetabolicResearch vol 41 no 1 pp 35ndash39 2009
[60] B D Albertson K L Frederick N C Maronian et al ldquoTheeffect of ketoconazole on steroidogenesis I Leydig cell enzymeactivity in vitrordquo Research Communications in Chemical Patho-logy and Pharmacology vol 61 no 1 pp 17ndash26 1988
[61] B D Albertson N C Maronian K L Frederick et al ldquoTheeffect of ketoconazole on steroidogenesis II Adrenocorticalenzyme activity in vitrordquoResearch Communications in ChemicalPathology and Pharmacology vol 61 no 1 pp 27ndash34 1988
[62] DM Stocco andB J Clark ldquoRegulation of the acute productionof steroids in steroidogenic cellsrdquo Endocrine Reviews vol 17 no3 pp 221ndash244 1996
[63] L K Christenson and J F Strauss III ldquoSteroidogenic acute regu-latory protein (StAR) and the intramitochondrial translocationof cholesterolrdquo Biochimica et Biophysica Acta (BBA)mdashMolecularand Cell Biology of Lipids vol 1529 no 1ndash3 pp 175ndash187 2000
[64] E R Simpson ldquoCholesterol side-chain cleavage cytochromeP450 and the control of steroidogenesisrdquo Molecular and Cel-lular Endocrinology vol 13 no 3 pp 213ndash227 1979
[65] H A LaVoie and S R King ldquoTranscriptional regulation of ster-oidogenic genes STARD1 CYP11A1 and HSD3Brdquo ExperimentalBiology and Medicine vol 234 no 8 pp 880ndash907 2009
[66] F Raucci A DrsquoAniello and M M Di Fiore ldquoStimulation ofandrogen production by d-aspartate through the enhancementof StAR P450scc and 3120573-HSD mRNA levels in vivo rat testisand in culture of immature rat Leydig cellsrdquo Steroids vol 84pp 103ndash110 2014
[67] M Doi Y Takahashi R Komatsu et al ldquoSalt-sensitive hyper-tension in circadian clock-deficient Cry-null mice involves dys-regulated adrenal Hsd3b6rdquo Nature Medicine vol 16 no 1 pp67ndash74 2010
[68] H Okamura M Doi Y Yamaguchi and J-M Fustin ldquoHyper-tension due to loss of clock novel insight from the molecu-lar analysis of Cry1Cry2-deleted micerdquo Current HypertensionReports vol 13 no 2 pp 103ndash108 2011
[69] M Doi F Satoh T Maekawa et al ldquoIsoform-specific mono-clonal antibodies against 3120573-hydroxysteroid dehydrogenaseisomerase family provide markers for subclassification ofhuman primary aldosteronismrdquo The Journal of Clinical Endo-crinology amp Metabolism vol 99 no 2 pp E257ndashE262 2014
[70] S Konosu-Fukaya Y Nakamura F Satoh et al ldquo3120573-hydroxy-steroid dehydrogenase isoforms in human aldosterone-produc-ing adenomardquo Molecular and Cellular Endocrinology vol 408pp 205ndash212 2015
[71] G C Verwoert J Hofland N Amin et al ldquoExpression and genevariation studies deny association of humanHSD3B1 gene withaldosterone production or blood pressurerdquoAmerican Journal ofHypertension vol 28 no 1 pp 113ndash120 2015
[72] D Hough K Storbeck SW P Cloete A C Swart and P SwartldquoRelative contribution of P450c17 towards the acute cortisolresponse lessons from sheep and goatsrdquoMolecular and CellularEndocrinology vol 408 pp 107ndash113 2015
[73] I Sakuma S Suematsu Y Matsuzawa et al ldquoCharacterizationof steroidogenic enzyme expression in aldosterone-producingadenoma a comparison with various human adrenal tumorsrdquoEndocrine Journal vol 60 no 3 pp 329ndash336 2013
[74] A Tong A Jia S Yan Y Zhang Y Xie and G Liu ldquoEctopiccortisol-producing adrenocortical adenoma in the renal hilumhistopathological features and steroidogenic enzyme profilerdquoInternational Journal of Clinical and Experimental Pathologyvol 7 no 7 pp 4415ndash4421 2014
[75] A Hirsch D Hahn P Kempna et al ldquoMetformin inhibitshuman androgen production by regulating steroidogenicenzymes HSD3B2 and CYP17A1 and complex I activity of therespiratory chainrdquo Endocrinology vol 153 no 9 pp 4354ndash43662012
[76] B Xu L Gao Y Cui et al ldquoSET protein up-regulated test-osterone production in the cultured preantral folliclesrdquo Repro-ductive Biology and Endocrinology vol 11 article 9 2013
[77] S Kikuchi D Tominaga M Arita M Takahashi and MTomita ldquoDynamic modeling of genetic networks using geneticalgorithm and S-systemrdquo Bioinformatics vol 19 no 5 pp 643ndash650 2003
[78] N Shikata YMaki M Nakatsui et al ldquoDetermining importantregulatory relations of amino acids from dynamic networkanalysis of plasma amino acidsrdquo Amino Acids vol 38 no 1 pp179ndash187 2010
[79] M Nakatsui T Ueda Y Maki I Ono and M OkamotoldquoMethod for inferring and extracting reliable genetic interac-tions from time-series profile of gene expressionrdquoMathematicalBiosciences vol 215 no 1 pp 105ndash114 2008
[80] A Komori Y Maki M Nakatsui I Ono and M OkamotoldquoEfficient numerical optimization algorithm based on new real-coded genetic algorithm AREX + JGG and application to theinverse problem in systems biologyrdquo Applied Mathematics vol3 no 10 pp 1463ndash1470 2012
[81] J A Roubos G van Straten and A J B van Boxtel ldquoAn evolu-tionary strategy for fed-batch bioreactor optimization conceptsand performancerdquo Journal of Biotechnology vol 67 no 2-3 pp173ndash187 1999
[82] D Sarkar and J M Modak ldquoGenetic algorithms with filters foroptimal control problems in fed-batch bioreactorsrdquo Bioprocessand Biosystems Engineering vol 26 no 5 pp 295ndash306 2004
[83] K Harada K Ikeda J Sakuma I Ono and S KobayashildquoHybridization of genetic algorithm with local search in mul-tiobjective function optimization recommendation of GA thenLSrdquoTransactions of the Japanese Society for Artificial Intelligencevol 21 no 6 pp 482ndash492 2006
Submit your manuscripts athttpwwwhindawicom
PainResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
ToxinsJournal of
VaccinesJournal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AntibioticsInternational Journal of
ToxicologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
StrokeResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Drug DeliveryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Pharmacological Sciences
Tropical MedicineJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AddictionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Emergency Medicine InternationalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Autoimmune Diseases
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anesthesiology Research and Practice
ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Pharmaceutics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
MEDIATORSINFLAMMATION
of
10 Journal of Toxicology
CHOL
0
20
40
60
80
100C
once
ntra
tion
in m
ediu
m (120583
M)
10 20 30 40 50 60 70 800Time (hours)
(a)
PREG
DCORT
DCORTICO
CORTICO
CORT
Con
cent
ratio
n in
med
ium
(120583M
)
10
4
2
0
8
6
10 20 30 40 50 60 70 800Time (hours)
(b)
Con
cent
ratio
n in
med
ium
(120583M
)
0
020406
081
1214
0 10 20 30 40 50 60 70 80Time (hours)
HPREG
DIONE
PROGHPROG
E1
(c)
DHEAE2
ALDO
TESTO
0
005
01
015
02
025
Con
cent
ratio
n in
med
ium
(120583M
)
0 10 20 30Time (hours)
40 50 60 70 80
(d)
CHOL
0
5000
10000
15000
20000
25000
30000
Con
cent
ratio
n in
cell
(120583M
)
10 20 30 40 50 60 70 800Time (hours)
(e)
Con
cent
ratio
n in
cell
(120583M
)
0
50
100
150
200
250
0 10 20 30 40 50 60 70 80Time (hours)
PREG
DCORT
DCORTICO
CORTICOCORT
(f)
0 10 20 30 40 50 60 70 80Time (hours)
HPREG
DIONE
PROGHPROG
E1
Con
cent
ratio
n in
cell
(120583M
)
3025201510
35
05
(g)
005
115
2
253
35
Con
cent
ratio
n in
cell
(120583M
)
0 10 20 30 40 50 60 70 80Time (hours)
DHEA
E2ALDO
TESTO
(h)
Figure 4 Comparison of time-course profiles of cholesterol and adrenal steroids produced by NCI-H295R cells between experimentallymeasured and simulated data To intuitively confirm reconstruction of the measured experimental data in the developed simulation modelof NCI-H295R cells dynamics of cholesterol and adrenal steroids produced by NCI-H295R cells were plotted to overlay experimentaldata with the simulated results Graphs show the dynamics of medium concentrations of cholesterol (a) and adrenal steroids ((b)ndash(d))and intracellular concentrations of cholesterol (e) and adrenal steroids ((f)ndash(h)) Steroid hormones were categorized into three groups byconcentration levels Major steroids were PREG pregnenolone DCORTICO 11-deoxycorticosterone DCORT 11-deoxycortisol CORTICOcorticosterone and CORT cortisol ((b) and (f)) Moderate steroids were HPREG 17120572-hydroxypregnenolone PROG progesterone HPROG17120572-hydroxyprogesterone DIONE androstenedione and E1 estrone ((c) and (g)) Minor steroids were DHEA dehydroepiandrosteroneALDO aldosterone TESTO testosterone and E2 17120573-estradiol ((d) and (h)) Experimental data are shown as symbols with dotted lines Alldata represent the mean plusmn SD (119873 = 4) Simulation data are shown as solid lines
Journal of Toxicology 11
0110
10
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
AN (120583M)
(a)
0110
FN (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(b)
00101
SM (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
100E minus 05
(c)
0110
TG (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B1
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(d)
0110
10100
AGT (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(e)
0110
1025
DDD (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(f)
1010
SP (120583M)
2750005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(g)
0110
10100
MP (120583M)
2110005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(h)
Figure 5 Continued
12 Journal of Toxicology
0110
10
KC (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(i)
0110
10
MC (120583M)
4430005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(j)
0110
10100
DZ (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(k)
Figure 5 Estimated mechanism of action of adrenal toxicants by using the mathematical model of adrenal steroidogenesis in NCI-H295Rcells Mechanisms of action of adrenal toxicity compounds were quantitatively estimated from experimental results of differential steroidprofiling using the mathematical model of adrenal steroidogenesis in NCI-H295R cells The drug action was defined as a scaling factorof enzymatic activity in the simulation model These scaling factors were optimized to fit experimental data by a hybrid optimizationmethod of the RCGA and nonlinear least squares Estimated drug actions by the exposure of vasculotoxic agents acrylonitrile AN (a)fumaronitrile FN (b) salinomycin SM (c) and thioguanine TG (d) and the steroidogenic inhibitors aminoglutethimide AGT (e) op1015840-DDD DDD (f) spironolactone SP (g) metyrapone MP (h) ketoconazole KC (i) miconazole MC (j) and daidzein DZ (k) are shownas a spider radar chart CYP11A1 P450 side chain cleavage enzyme CYP17H 17120572-hydroxylase of CYP17 CYP17L C
17ndash20 lyase of CYP17HSD3B2 3120573-hydroxysteroid dehydrogenase CYP21A2 21-hydroxylase CYP11B1 11120573-hydroxylase CYP11B2 18-hydroxylase and HSD17B317120573-hydroxysteroid dehydrogenase
with previous reports [33 34] 7120572-thiospironolactone whichis synthesized by deacetylation of SP inhibits CYP17H andCYP17L [34] The fact that there were no inhibitions ofCYP17HorCYP17L in our study suggests that SPmight not bedeacetylated to 7120572-thiospironolactone in NCI-H295R cellsRegarding CYP11B1 and CYP11B2 our results were unclearcompared with a previous study It has been shown that30 120583M SP inhibits CYP11B1 and CYP11B2 in human andbovine adrenal mitochondria [35]The cause of CYP11B1 andCYP11B2 inhibition by SP could not be determined in ourstudy which might be due to the lower maximum exposureconcentration of SP than that in the previous reportWe couldnot examine SP concentrations over 10 120583M because theseconcentrations were cytotoxic in NCI-H295R cells
335 MP The mechanism of action of MP in adrenalsteroidogenesis was estimated by dose-dependent inhibitionof CYP11B1 (estimated inhibitions at 1 10 and 100 120583M were571 827 and 982 resp) (Figure 5(h)) MP has beenreported to inhibit CYP11B1 as its major effect and CYP11A1and CYP11B2 as a weak effect [6 29 36ndash39] The results wereable to show that MP is a selective inhibitor of CYP11B1 inthe previous report However the estimated effect of MPat a high concentration 100 120583M as the maximum exposureconcentration was unclear According to the previous reportselectivity of MP for CYP11B1CYP11B2 is about five times[39] In addition 20 120583M MP has a slight inhibition effect onCYP11A1 in H295R cells [29]
336 KC The mechanism of action of KC in adrenalsteroidogenesis was estimated by inhibition of CYP11A1CYP17H CYP17L HSD3B2 CYP21A2 CYP11B1 and
CYP11B2 (estimated inhibitions at 10 120583Mwere 926 943518 830 882 974 and 798 resp) (Figure 5(i))KC has been reported to inhibit CYP11A1 CYP17H CYP17LHSD3B2 CYP21A2 and CYP11B1 [6 29 40ndash43] Our resultswere almost consistent with the previous reports KC inhibitsCYP11A1 CYP17H CYP21A2 and CYP11B1 in NCI-H295Rcells at 10 120583M [29] and CYP17H CYP17L CYP21A2 andCYP11B1 in human adrenal mitochondria and Leydig cellmicrosomes at 2ndash5120583M [60 61] However KC has shownonly weak inhibition of HSD3B2 and HSD17B3 in Leydigcells at the millimolar level [60 61] Regarding CYP11B2 andHSD17B3 we considered that these estimated inhibitions ofKC did have sufficient reliability in terms of quantitative pre-diction precision because ALDO and TESTO concentrationswere less than the lower limit of quantification at 10 120583MKC
337 MC Themechanism of action of MC in adrenal stero-idogenesis was estimated by inhibition of CYP17H CYP17LCYP11B1 and HSD17B3 (estimated inhibitions at 10120583Mwere691 530 764 and 571 resp) (Figure 5(j)) MC hasbeen reported to inhibit not only CYP17H and CYP17L butalsoCYP11A1 CYP21A2 andCYP11B1 [41 44 45]The resultsin the previous reports were able to estimate that MC isa CYP17 inhibitor However CYP11A1 inhibition by MCprobably instead of a reduction in StAR expression wasnot clearly detected in our study using NCI-H295R cellsbecause there were no decreases in the concentrations ofPREG and PROG in the culture medium Indirect inhibitionof CYP11A1 via the peripheral-type benzodiazepine receptorhas been reported in mouse adrenocortical Y-1 cells treatedwith MC in the absence of stimuli by measuring PREGproduction [44] On the other hand reductions of StAR
Journal of Toxicology 13
protein expression andor transport activity without affectingtotal steroid synthesis or CYP11A1 and HSD3B2 enzymeexpression or activities have been reported in (BU)
2cAMP-
stimulated MA-10 Leydig tumor cells treated with MC bymeasuring PROG production [45] Therefore the effect ofMC on the initial reaction in adrenal steroidogenesis fromcholesterol should be different according to the cell type andstimulation condition Inhibition of CYP21A2 and CYP11B1by MC has been reported as decreases in the consumptionof PROG and DCORTICO respectively [41] Inhibition ofCYP11B1 was estimated by the action of MC in this studybut that of CYP21A2 was not detected In the previousexperimental report inhibitory sites by MCmight have beenreflected by inhibition of CYP17H activity because CYP21A2activity was measured as a decrease in labeled PROG
338 DZ The mechanism of action of DZ in adrenalsteroidogenesis was estimated by inhibition of CYP11A1HSD3B2 CYP21A2 CYP11B1 and HSD17B3 (estimated inhi-bitions at 100 120583M were 586 941 965 872 and981 resp) (Figure 5(k)) DZ has been reported to inhibitHSD3B2 and CYP21A2 [46] The results of HSD3B2 andCYP21A2 were consistent with the previous report Howeverinhibitions have not been reported for CYP11A1 CYP11B1and HSD17B3These estimated effects of DZ on CYP11B1 andHSD17B3 were unclear because the concentrations of ALDOand TESTO were less than the lower limit of quantificationat 100 120583M In addition these enzymes act downstream of thestrong action points of DZ such as HSD3B2 and CYP21A2
34 The Simulations and the Systematic Model Analysis
341 Dynamic Sensitivity Analysis of Adrenal SteroidogenesisTo comprehensively understand the dynamics of adrenalsteroidogenesis dynamic sensitivities were calculated forsteroid concentrations secreted by NCI-H295R cells usingour constructed mathematical model of steroidogenesis Theresults of dynamic sensitivity analysis at 72 h of duration and6 h of interval time are presented as a heat-map in Figure 6
The top 10 parameters of the total area under the curveof dynamic sensitivity for cholesterol and the 14 steroids inculture medium were 119881HSD3B2
maxA 119881CYP11A1max 119881CYP21A2maxA 119896loc119891 119896loc119887
119870CYP11A1119898
119881CYP17HmaxA 119896Cholesterol Transport 119896acc119891 and 119881CYP21A2maxB in
order from the top Cholesterol uptake (119896Cholesterol Transport)StAR protein (119896loc
119891) and CYP11A1 (119881CYP11A1max ) which are
determining factors of the capacity for steroidogenesis pro-moted the production of mineralocorticoids (DCORTICOCORTICO and ALDO) and restrained the synthesis of glu-cocorticoids (DCORT and CORT) and sex steroids (DIONETESTO and E1) because of the accumulation of interme-diate molecules in steroidogenesis (PREG HPREG DHEAPROG and HPROG) only by self-activation The dynamicpatterns of the intermediate molecules in steroidogenesiswere mainly dependent on the activity of CYP17H andHSD3B2 with PREG as the substrate of these enzymes inwhich the dynamic sensitivities of 119881CYP17HmaxA for HPREG andHPROG and 119881HSD3B2
maxA for PROG HPROG and DCORTICO
reversed the direction of sensitivity at 49ndash66 h after stimu-lation The dynamic sensitivities of the maximum activitiesof HSD3B2 for PREG (119881HSD3B2
maxA ) and CYP21A2 for PROG(119881CYP21A2maxA ) were related to all steroids at 72 h Almost allmodel parameters had positive sensitivity for downstreamsteroids in the adrenal steroidogenic pathway and negativesensitivity for direct-binding steroids as substrates of thesteroidogenic enzyme The sensitivity of 119881max in all steroido-genic enzymes was relatively higher than 119870
119898for the same
steroid substrate
342 Simulation of the Metabolic Balance of AdrenalSteroidogenesis Pathway To clearly show the property of themetabolic shift between mineralocorticoid and glucocorti-coid biosynthesis we performed two-dimensional parameterscanning of the enzymatic activities of CYP17H and HSD3B2(Figure 7) NCI-H295R cells lost the ability to produce allsteroid hormones when enzymatic activities of CYP17H andHSD3B2 were changed by over 60 and 30 respectivelyActivation ofCYP17HandorHSD3B2 induced themetabolicshift that enhanced the glucocorticoid biosynthesis and devi-ated from the mineralocorticoid biosynthesis On the otherhand inhibition of CYP17H andor HSD3B2 induced themetabolic shift that enhanced the mineralocorticoid biosyn-thesis and deviated from the glucocorticoid biosynthesisMoreover the enzymatic activity of HSD3B2 regulated themetabolic balance of sex steroids and the precursors onadrenal steroidogenesis of NCI-H295R cells E1 TESTO andDIONE were produced by NCI-H295R cells when activatingthe enzymatic activity of HSD3B2 Conversely E2 andDHEAwere produced by NCI-H295R cells when suppressing theenzymatic activity of HSD3B2 The biosynthesis of down-stream steroids in adrenal steroidogenesis pathway such asmineralocorticoids and glucocorticoids was almost com-pletely terminated when the enzymatic activity of HSD3B2was decreased by over 80
4 Discussion
41 Importance of 3120573-HSDActivity in Adrenal SteroidogenesisOur systematic analysis using the mathematical model ofadrenal steroidogenesis in NCI-H295R cells revealed thatthe enzymatic activity of 3120573-HSD controls the dynamics ofadrenal steroidogenesis The activity of the StAR proteincontrols the net capacity of steroidogenesis in steroidogeniccells which is the transport of cholesterol from the outer toinner mitochondrial membranes Both the expression levelsof StAR protein and mRNA are rapidly elevated in responseto stimulation by tropic hormones such as ACTH [62 63]Another important factor in adrenal steroidogenesisis is thecholesterol side chain cleavage enzyme CYP11A1 the firstrate-limiting and hormonally regulated step in the synthesisof all steroid hormones which is conversion of cholesterolto pregnenolone in mitochondria [64] According to ourresults of global sensitivity analysis (Supplementary Table 1and Figure 6(d)) in addition to CYP11A1 and StAR proteins3120573-HSD was one of the key regulators in adrenal steroido-genesis of NCI-H295R cells And also this result suggeststhat a significant regulatory mechanism in steroidogenesis
14 Journal of Toxicology
60544830 4236 66 7218 246 120
Time (hours)
0
1
2
3C
once
ntra
tion
(a)
60544830 4236 66 7218 246 120
Time (hours)
minus05
0
05
minus1
1
Sens
itivi
ty
(b)
0
6
12
18
24
30
36
42
48
54
60
66
72
Sens
itivi
ty fo
r med
ium
conc
entr
atio
ns
of ch
oles
tero
l or s
tero
id h
orm
ones
Dur
atio
n tim
e (ho
urs)
Model parameters
+10
+05
00
minus05
minus10
(c)
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
Parameter number in the mathematical model of steroidogenesis in NCI-H295R cells
CHOL
PREG
HPREG
DHEA
PROG
HPROG
DIONE
DCORTICO
DCORT
CORTICO
CORT
ALDO
TESTO
E1
E2
Chol
este
rol o
r ste
roid
hor
mon
es
0
05
10
15
20
minus05
minus10
minus15
minus20
(d)
Figure 6 Heat-map of the global dynamic sensitivity analysis of adrenal steroid concentrations produced by NCI-H295R cells The globaldynamic sensitivity analysis is a powerful tool to comprehensively understand the dependencies of themodel parameters in themathematicalmodel of the biological complex system Dynamic sensitivities of model parameters in the mathematical model of adrenal steroidogenesisin NCI-H295R cells were calculated for all steroid concentrations in the culture medium every 6 h until 72 h after stimulation To clarify theview of heat-map of global dynamic sensitivity analysis imaginary data of the dynamics of steroid concentrations in the original model (blueline) and perturbed model for sensitivity analysis (red line) were prepared (a) Using this imaginary data the calculated dynamic sensitivities(b) and the visualized dynamic sensitivity as one block of the heat-map (c) were shown respectively By the same method that explainedusing imaginary data the large-scale data of the global dynamic sensitivity analysis on the mathematical model of adrenal steroidogenesisin NCI-H295R cells was comprehensively visualized as a big graph of heat-map (d) Parameter numbers in the horizontal axis are (1)119896Choresterol Transport (2) 119896CEH (3) 119896MTR
119891 (4) 119896MTR
119887 (5) 119896acc
119891 (6) 119896acc
119887 (7) 119896loc
119891 (8) 119896loc
119887 (9) 119896Oxysterol Synthesis (10)119870
CYP11A1119898
(11) 119881CYP11A1max (12)119870CYP17H119898A
(13) 119870CYP17H119898B (14) 119881CYP17HmaxA (15) 119881CYP17HmaxB (16) 119881CYP17HmaxB (17) 119870CYP17L
119898B (18) 119881CYP17LmaxA (19) 119881CYP17LmaxB (20) 119870HSD3B2119898A (21) 119870HSD3B2
119898B (22) 119870HSD3B2119898C (23)
119881HSD3B2maxA (24) 119881HSD3B2
maxB (25) 119881HSD3B2maxB (26) 119870CYP21A2
119898A (27) 119870CYP21A2119898B (28) 119881CYP21A2maxA (29) 119881CYP21A2maxB (30) 119870CYP11B1
119898A (31) 119870CYP11B1119898B (32) 119881CYP11B1maxA
(33) 119881CYP11B1maxB (34) 119896CYP11B2 (35) 119870HSD17B3119898A (36) 119870HSD17B3
119898B (37) 119881HSD17B3maxA (38) 119881HSD17B3
maxB (39) 119870CYP19A1119898A (40) 119870CYP19A1
119898B (41) 119881CYP19A1maxA and (42)119881
CYP19A1maxB PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-
hydroxyprogesteroneDIONE androstenedioneDCORTICO 11-deoxycorticosteroneDCORT 11-deoxycortisol CORTICO corticosteroneCORT cortisol ALDO aldosterone and TESTO testosterone
pathway is very reasonable StAR CYP11A1 and 3120573-HSD(isoforms 1 or 2 in humans) proteins generally respond to thesame hormones that stimulate steroid production throughcommon pathways such as cAMP signaling in adrenal glandsand testes [65 66] Moreover our data also support recentexperimental evidence from clinical and in vivo studiessuggesting that the enzymatic activity of 3120573-HSD plays
an important role in the regulation of mineralocorticoidsynthesis in adrenal steroidogenesis and contributes to hyper-tension caused by abnormal overproduction of aldosterone[67ndash70] Circadian clock-deficient Cry-null mice show salt-sensitive hypertension due to abnormally high synthesis ofaldosterone which is caused by constitutively high expressionof HSD3B6mRNA and protein in the adrenal cortex [67 68]
Journal of Toxicology 15
All steroid producing cells
Mineralocorticoid producing cells Glucocorticoid producing cells
Mineralocorticoid and glucocorticoid producing cells
Terminated steroidogenesis
Sex steroids and precursor sex steroidsDIONE TESTO E1DEHA E2
NCI-H295Rcells at controlcondition
Normalized activities of HSD3B2
Nor
mal
ized
activ
ities
of C
YP17
H
2
18
16
14
12
1
08
06
04
02
00 05 1 15 2
Figure 7 Metabolic categories of the steroidogenic cells determined by the balance of HSD3B2 and CYP17H activitiesThe two-dimensionalparameter scanning analysis by the perturbation of two focused parameters clarifies the interaction and the relationship between the twoparameters in the complex system Functional cellular categories of steroidogenic cells were defined by the levels of mineralocorticoid(ALDO) glucocorticoids (DCORTandCORT) and androgens (DHEAandDIONE) at 72 h after stimulation Enzymatic activities ofHSD3B2andCYP17were normalized by standard values of the simulationmodel inNCI-H295R cells Scanning ranges ofHSD3B2 andCYP17 activitieswere 0ndash200 each 10 Green regions are all 14 steroids produced by NCI-H295R cells Red blue and purple regions are mineralocorticoidproducing cells glucocorticoids-producing cells and both corticoid-producing cells respectively Yellow regions are steroidogenesis of NCI-H295R cells terminated upstream of the adrenal steroidogenesis pathway
Recent clinical observations have revealed predominantexpression of HSD3B2 mRNA and protein in tumor cellsof aldosterone-producing adenoma (APA) and HSD3B1mRNA significantly correlated with CYP11B2 mRNA levelsand plasma aldosterone concentrations in APA patients [6970] However the relationship is unclear and disputed ina small-scale clinical study indicating that genetic variationin HSD3B1 affects blood pressure and hypertension in APApatients [71] The results of our simulation study suggest that3120573-HSD protein (human genes are HSD3B1 and HSD3B2) isone of the determination factors for the dynamic propertyof adrenal steroidogenesis Our results support the clinicalevidence of Doi and colleagues [69] and we believe that theHSD3B1 enzyme has a promising potential as novel drugtarget for endocrine hypertension
42 Metabolic Shift of Adrenal Steroidogenesis and the Contri-butions of HSD3B2 and CYP17H The metabolic propertiesof adrenal steroidogenesis in NCI-H295R cells were revealedby dynamic sensitivity analysis using themathematicalmodel(Figure 6) Mineralocorticoids such as DCORTICO COR-TICO and ALDO and intermediate steroids upstream of
the adrenal steroidogenesis pathway such as PREG HPREGDHEA PROG and HPROG were accelerated by reactionsof cholesterol import (119896Cholesterol Transport) StAR protein (119896MTR
119891
and 119896loc119891) and CYP11A1 (119881CYP11A1max ) On the other hand glu-
cocorticoids such as DCORT and CORT and sex hormonessuch as DIONE TESTO and E1 were suppressed by thesemodel parametersTherefore enhancement of the net adrenalsteroidogenesis capacity which supplies PREG precursor tothe pathway causes a production shift from glucocorticoidsto mineralocorticoids by substrate inhibitions of CYP17HHSD3B2 and CYP21A2 caused by accumulation of initialintermediate steroids such as PREG and PROG Sensitivi-ties of CYP17H (119881CYP17HmaxA ) and HSD3B2 (119881HSD3B2
maxA ) for theproducts were dynamically changed and these parametersdetermined the metabolic balance of downstream steroidsin the adrenal steroidogenesis pathway According to theseresults of dynamic sensitivity analysis of StAR CYP11A1CYP17H and HSD3B2 we suggest that the enhancementof CYP17H and HSD3B2 activity during ACTH stimulationwas important to shift the steroidogenic output away fromALDO biosynthesis towards CORT biosynthesis as wellas adrenal androgen production This suggestion partially
16 Journal of Toxicology
supports a comparative animal study in which molecularand cellular variations in CYP17H activity dramatically affectacute cortisol production resulting in distinct physiologicaland behavioral responses [72]
Results of two-dimensional parameter scanning of theenzymatic activities of CYP17H and HSD3B2 quantitativelyshowed the detail of the metabolic relationship betweenmin-eralocorticoid and glucocorticoid biosynthesis (Figure 7)Particularly the results showed that the balance of theseenzymatic activities was very important for the typicalfunction of NCI-H295R cells namely the ability to produceall steroid hormones NCI-H295R cells lost this functionwhen enzymatic activities of CYP17H and HSD3B2 werechanged by over 60 and 30 respectively In addition theybecame mineralocorticoid (ALDO) secreting cells when theenzymatic activity of CYP17H or HSD3B2 was inhibited byover 50 or glucocorticoid (DCORT and CORT) secretingcells when these enzymes were activated by over 50 Inparticular this analysis also showed that HSD3B2 was a keyplayer in the adrenal steroidogenesis of NCI-H295R cellsbecause HSD3B2 inhibition by over 80 almost completelyinhibited the biosynthesis of downstream steroids The ratioof CYP17A1 to HSD3B2 mRNA expression levels has beenrelated to several endocrine diseases with a low level in APAs[73] and high level in cortisol-producing adenomas [74]Furthermore the expression levels or enzymatic activitiesof CYP17A1 and HSD3B1 have been related to androgenproduction in polycystic ovary syndrome [75 76] Theseclinical studies support our simulation results indicating thatthe balance of enzymatic activity of CYP17H and HSD3B2determines the shift in steroidogenic output to mineralocor-ticoids glucocorticoids or androgens
43 Methodologies of Quantitative Mechanistic Analysis forDrug Discovery According to our results obtained using themathematical model of steroidogenesis in NCI-H295R cellssuch as sensitivity analysis comprehensive analysis based onsystems biology is available to quantitatively estimate themechanism of action of steroidogenic disrupting compoundsfrom differential profiling of adrenal steroid hormonesbecause dynamic patterns of steroid hormones in adrenalsteroidogenesis pathway are highly complex Our proposedmethod of quantitative mechanistic analysis of steroidogenicinhibitors was able to predict known action sites in theadrenal steroidogenesis pathway at only one time point (72 hafter drug exposure) Moreover according to the resultsof sensitivity analysis (Figure 6) 119881max of all steroidogenicenzymes was more sensitive than the 119870
119898 because the
intracellular concentrations of steroid hormones were almostmaintained at sufficiently high levels compared with 119870
119898
values of steroidogenic enzymesThese results suggested thatestimation of the mechanism of action of drugs is moreeffective and detectable when using the influences of 119881maxas the searching parameters such as our proposed methodOur data showed that the proposed method based on asystems biology model is a very powerful tool for exploratoryscreening of steroidogenic disrupting compounds
RCGA as a solver of parameter estimation problemsin systems biology has been applied to biological network
identification of gene regulatory networks and metabolicpathways and optimization of biological processes usingexperimentally observed time-course data [77ndash82] In thisstudy RCGA was useful to estimate the mechanism ofaction of novel pharmaceutical drug candidates for adrenalsteroidogenesis as a new application of RCGA in systemsbiology We had two issues when applying RCGA to thequantitative mechanistic analysis of drug actions Theseissues were the vast calculation cost and multimodality ofquasi-optimum solutions in solving the optimization prob-lem because the mathematical model in systems biologyconsists of many equations and parameters A proposedoptimization strategy using RCGA based on REXJGG wasa highly stable and efficient calculation method for a betterquasi-optimum solution than the unimodal normal distri-bution crossover (UNDX)minimum generation gap (MGG)method that is well applied in the engineering field In addi-tion we expanded the RCGA optimization program basedon REXJGG to a hybrid method of GA and then applieda local search as recommended by Harada and Kobayashi[28 83] A final optimal solution was obtained with a goodconvergence property Because these problems are generalin systems biology studies we suggest that the proposedhybrid method based on REXJGG is a very useful tool forquantitative mechanistic analysis of novel pharmaceuticaldrugs not limited to steroidogenic disrupting compounds
5 Conclusions
Thenovelmathematicalmodel of adrenal steroidogenesis wasconstructed in this study including cholesterol transport anddistribution the C
21-steroid hormone pathway steroid trans-
port and cell proliferation which could reproduce adrenalsteroidogenesis in NCI-H295R cells According to the resultsof dynamic sensitivity analysis using the newmodel HSD3B2plays the most important role in the metabolic balance ofadrenal steroidogenesis in NCI-H295R cells Moreover toquantitatively estimate mechanisms of action of adrenal toxiccompounds we analyzed differential metabolic profiles of12 steroid hormones at 3 days after exposure to 11 adrenaltoxic compounds by using the new mathematical modeland a hybrid optimization method of the RCGA and a localsearch (nonlinear least squares) We could estimate whichsteroidogenic enzymes were affected by these compoundsusing the hybrid optimization method Vasculotoxic agentswere estimated to have no effect according to the resultsobtained by our method In terms of adrenal steroidogenicinhibitors the predicted action sites were approximatelymatched to the target enzymes as reported in the literatureThus our computer-aided method based on a systems biol-ogy approach may be useful to analyze the mechanism ofaction of endocrine-disrupting compounds and to assess thehuman adrenal toxicity of novel pharmaceutical drugs basedon steroid hormone profiling
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Journal of Toxicology 17
Authorsrsquo Contribution
Ryuta Saito Natsuko Terasaki and Naohisa Tsutsui designedthe study Naohisa Tsutsui and Naoya Masutomi supervisedthe experiments Natsuko Terasaki performed all in vitroexperiments and collected dataMakotoYamazaki performedall measurements of steroids Ryuta Saito conceived the insilico strategies analyzed data developed all programs andthe mathematical model and performed all simulation anal-yses with Masahiro Okamoto Ryuta Saito Natsuko TerasakiMakoto Yamazaki NaoyaMasutomi andMasahiroOkamotowrote the paper
Acknowledgments
The authors would like to thank Dr Michael S Breenand Dr Miyuki Breen for helpful discussions concerningprevious mathematical models of adrenal steroidogenesisThis work was partially supported by the Grant-in-Aid forScientific Research on Innovative Areas ldquoSynthetic Biologyrdquo(No 23119001 (MO)) from Ministry of Education CultureSports Science and Technology in Japan
References
[1] P W Harvey D J Everett and C J Springall ldquoAdrenal toxi-cology a strategy for assessment of functional toxicity to theadrenal cortex and steroidogenesisrdquo Journal of Applied Toxi-cology vol 27 no 2 pp 103ndash115 2007
[2] W E Ribelin ldquoThe effects of drugs and chemicals upon thestructure of the adrenal glandrdquo Fundamental and Applied Toxi-cology vol 4 no 1 pp 105ndash119 1984
[3] H D Colby and P A Longhurst ldquoToxicology of the adrenalglandrdquo in Endocrine Toxicology C K Atterwill and J D FlackEds pp 243ndash281 CambridgeUniversity Press Cambridge UK1992
[4] H D Colby ldquoIn vitro assessment of adrenocortical toxicityrdquoJournal of Pharmacological and Toxicological Methods vol 32no 1 pp 1ndash6 1994
[5] T J Rosol J T Yarrington J Latendresse and C C CapenldquoAdrenal gland structure function and mechanisms of toxic-ityrdquo Toxicologic Pathology vol 29 no 1 pp 41ndash48 2001
[6] P W Harvey and D J Everett ldquoThe adrenal cortex and ster-oidogenesis as cellular and molecular targets for toxicity crit-ical omissions from regulatory endocrine disrupter screeningstrategies for human healthrdquo Journal of Applied Toxicology vol23 no 2 pp 81ndash87 2003
[7] A F Gazdar H K Oie C H Shackleton et al ldquoEstablishmentand characterization of a human adrenocortical carcinoma cellline that expresses multiple pathways of steroid biosynthesisrdquoCancer Research vol 50 no 17 pp 5488ndash5496 1990
[8] W E Rainey I M Bird C Sawetawan et al ldquoRegulation ofhuman adrenal carcinoma cell (NCI-H295) production of C19steroidsrdquo The Journal of Clinical Endocrinology amp Metabolismvol 77 no 3 pp 731ndash737 1993
[9] B Staels D W Hum andW L Miller ldquoRegulation of steroido-genesis in NCI-H295 cells a cellular model of the human fetaladrenalrdquo Molecular Endocrinology vol 7 no 3 pp 423ndash4331993
[10] M Hecker J L Newsted M B Murphy et al ldquoHuman adreno-carcinoma (H295R) cells for rapid in vitro determination of
effects on steroidogenesis hormone productionrdquo Toxicologyand Applied Pharmacology vol 217 no 1 pp 114ndash124 2006
[11] A Oskarsson E Ulleras K E Plant J P Hinson and P SGoldfarb ldquoSteroidogenic gene expression in H295R cells andthe human adrenal gland adrenotoxic effects of lindane invitrordquo Journal of Applied Toxicology vol 26 no 6 pp 484ndash4922006
[12] UMuller-Vieira M Angotti and RW Hartman ldquoThe adreno-cortical tumor cell line NCI-H295R as an in vitro screeningsystem for the evaluation of CYP11B2 (aldosterone synthase)and CYP11B1 (steroid-11120573-hydroxylase) inhibitorsrdquoThe Journalof Steroid Biochemistry and Molecular Biology vol 96 no 3-4pp 259ndash270 2005
[13] K Imagawa S Okayama M Takaoka et al ldquoInhibitory effectof efonidipine on aldosterone synthesis and secretion in humanadrenocarcinoma (H295R) cellsrdquo Journal of CardiovascularPharmacology vol 47 no 1 pp 133ndash138 2006
[14] M Voets UMuller-Vieira S Marchais-Oberwinkler and RWHartmann ldquoSynthesis of amidinohydrazones and evaluation oftheir inhibitory effect towards aldosterone synthase (CYP11B2)and the formation of selected steroidsrdquo Archiv der Pharmazievol 337 no 7 pp 411ndash416 2004
[15] F K Nielsen C H Hansen J A Fey et al ldquoH295R cells as amodel for steroidogenic disruption a broader perspective usingsimultaneous chemical analysis of 7 key steroid hormonesrdquoToxicology in Vitro vol 26 no 2 pp 343ndash350 2012
[16] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cortisolproduction and on activities of steroidogenic enzymes inhuman adrenocortical H295R cellsrdquo Journal of Steroid Biochem-istry and Molecular Biology vol 80 no 3 pp 355ndash363 2002
[17] R F Canton J T Sanderson S Nijmeijer A Bergman R JLetcher and M van den Berg ldquoIn vitro effects of brominatedflame retardants and metabolites on CYP17 catalytic activity anovel mechanism of actionrdquo Toxicology and Applied Pharma-cology vol 216 no 2 pp 274ndash281 2006
[18] K Hilscherova P D Jones T Gracia et al ldquoAssessment ofthe effects of chemicals on the expression of ten steroidogenicgenes in the H295R cell line using real-time PCRrdquo ToxicologicalSciences vol 81 no 1 pp 78ndash89 2004
[19] T Gracia K Hilscherova P D Jones et al ldquoThe H295R systemfor evaluation of endocrine-disrupting effectsrdquo Ecotoxicologyand Environmental Safety vol 65 no 3 pp 293ndash305 2006
[20] X Zhang J L NewstedM Hecker E B Higley P D Jones andJ P Giesy ldquoClassification of chemicals based on concentration-dependent toxicological data using ToxClustrdquo EnvironmentalScience and Technology vol 43 no 10 pp 3926ndash3932 2009
[21] M S Breen M Breen N Terasaki M Yamazaki and R BConolly ldquoComputational model of steroidogenesis in humanH295R cells to predict biochemical response to endocrine-active chemicalsmodel development formetyraponerdquoEnviron-mental Health Perspectives vol 118 no 2 pp 265ndash272 2010
[22] M Breen M S Breen N Terasaki M Yamazaki A LLloyd and R B Conolly ldquoMechanistic computational model ofsteroidogenesis in H295R cells role of oxysterols and cell pro-liferation to improve predictability of biochemical response toendocrine active chemical-metyraponerdquo Toxicological Sciencesvol 123 no 1 pp 80ndash93 2011
[23] F B Kraemer ldquoAdrenal cholesterol utilizationrdquo Molecular andCellular Endocrinology vol 265-266 pp 42ndash45 2007
[24] W L Miller ldquoSteroidogenic acute regulatory protein (StAR)a novel mitochondrial cholesterol transporterrdquo Biochimica et
18 Journal of Toxicology
Biophysica Acta (BBA)mdashMolecular and Cell Biology of Lipidsvol 1771 no 6 pp 663ndash676 2007
[25] S Szabo and I T Lippe ldquoAdrenal gland chemically inducedstructural and functional changes in the cortexrdquo ToxicologicPathology vol 17 no 2 pp 317ndash329 1989
[26] C Chen-Pan I-J Pan Y Yamamoto T Sakogawa J Yamadaand YHayashi ldquoPrompt recovery of damaged adrenalmedullaeinduced by salinomycinrdquoToxicologic Pathology vol 27 no 5 pp563ndash572 1999
[27] R J Santen and R I Misbin ldquoAminoglutethimide review ofpharmacology and clinical userdquo Pharmacotherapy vol 1 no 2pp 95ndash120 1981
[28] M J Carella N V Dimitrov V V Gossain L Srivastava andD R Rovner ldquoAdrenal effects of low-dose aminoglutethimidewhen used alone in postmenopausal women with advancedbreast cancerrdquoMetabolism vol 43 no 6 pp 723ndash727 1994
[29] M K Johansson J T Sanderson and B-O Lund ldquoEffects of3-MeSO2-DDE and some CYP inhibitors on glucocorticoidsteroidogenesis in the H295R human adrenocortical carcinomacell linerdquo Toxicology in Vitro vol 16 no 2 pp 113ndash121 2002
[30] Y Touitou A Bogdan J C Legrand and P Desgrez ldquoAmino-glutethimide and glutethimide effects on 18-hydroxycorti-costerone biosynthesis by human and sheep adrenals in vitrordquoActa Endocrinologica vol 80 no 3 pp 517ndash526 1975
[31] M M Hart and J A Straw ldquoStudies on the site of action ofop1015840-DDD in the dog adrenal cortex 1 Inhibition of ACTH-mediated pregnenolone synthesisrdquo Steroids vol 17 no 1ndash5 pp559ndash574 1971
[32] M Ojima M Saitoh N Itoh Y Kusano S Fukuchi and HNaganuma ldquoThe effects of op1015840-DDD on adrenal steroidogen-esis and hepatic steroid metabolismrdquo Nippon Naibunpi Gakkaizasshi vol 61 no 3 pp 168ndash178 1985
[33] L F Canosa and N R Ceballos ldquoEffects of different steroid-biosynthesis inhibitors on the testicular steroidogenesis of thetoad Bufo arenarumrdquo Journal of Comparative Physiology B vol171 no 6 pp 519ndash526 2001
[34] D C Kossor S Kominami S Takemori and H D ColbyldquoRole of the steroid 17 alpha-hydroxylase in spironolactone-mediated destruction of adrenal cytochrome P-450rdquoMolecularPharmacology vol 40 no 2 pp 321ndash325 1991
[35] S C Cheng K Suzuki W Sadee and B W Harding ldquoEffectsof spironolactone canrenone and canrenoate K on cytochromeP450 and 11120573 and 18 hydroxylation in bovine and humanadrenal corticalmitochondriardquo Endocrinology vol 99 no 4 pp1097ndash1106 1976
[36] K Yanagibashi M Haniu J E Shively W H Shen and P HallldquoThe synthesis of aldosterone by the adrenal cortex Two zones(fasciculata and glomerulosa) possess one enzyme for 11120573- 18-hydroxylation and aldehyde synthesisrdquoThe Journal of BiologicalChemistry vol 261 no 8 pp 3556ndash3562 1986
[37] K Morishita H Okumura N Ito and N Takahashi ldquoPrimaryculture system of adrenocortical cells from dogs to evaluatedirect effects of chemicals on steroidogenesisrdquo Toxicology vol165 no 2-3 pp 171ndash178 2001
[38] S W J Lamberts E G Bons H A Bruining and F H de JongldquoDifferential effects of the imidazole derivatives etomidateketoconazole and miconazole and of metyrapone on the secre-tion of cortisol and its precursors by human adrenocorticalcellsrdquo Journal of Pharmacology and Experimental Therapeuticsvol 240 no 1 pp 259ndash264 1987
[39] L Yin S Lucas F Maurer U Kazmaier Q Hu and RW Hartmann ldquoNovel imidazol-1-ylmethyl substituted 1256-tetrahydropyrrolo[321-ij]quinolin-4-ones as potent and selec-tive CYP11B1 inhibitors for the treatment of Cushingrsquos syn-dromerdquo Journal ofMedicinal Chemistry vol 55 no 14 pp 6629ndash6633 2012
[40] M DiMattina N Maronian H Ashby D L Loriaux andB D Albertson ldquoKetoconazole inhibits multiple steroidogenicenzymes involved in androgen biosynthesis in the humanovaryrdquo Fertility and Sterility vol 49 no 1 pp 62ndash65 1988
[41] M Ayub and M J Levell ldquoInhibition of human adrenal steroi-dogenic enzymes in vitro by imidazole drugs including keto-conazolerdquo Journal of Steroid Biochemistry vol 32 no 4 pp 515ndash524 1989
[42] D Engelhardt M M Weber T Miksch F Abedinpour andC Jaspers ldquoThe influence of ketoconazole on human adrenalsteroidogenesis incubation studies with tissue slicesrdquo ClinicalEndocrinology vol 35 no 2 pp 163ndash168 1991
[43] S Bhasin S Sikka T Fielder et al ldquoHormonal effects of keto-conazole in vivo in the male rat mechanism of actionrdquo Endo-crinology vol 118 no 3 pp 1229ndash1232 1986
[44] D M Zisterer and D C Williams ldquoCalmidazolium and otherimidazole compounds affect steroidogenesis in Y1 cells lack ofinvolvement of the peripheral-type benzodiazepine receptorrdquoJournal of Steroid Biochemistry and Molecular Biology vol 60no 3-4 pp 189ndash195 1997
[45] L P Walsh C N Kuratko and D M Stocco ldquoEconazoleand miconazole inhibit steroidogenesis and disrupt steroido-genic acute regulatory (StAR) protein expression post-trans-criptionallyrdquo Journal of Steroid Biochemistry and Molecular Bio-logy vol 75 no 4-5 pp 229ndash236 2000
[46] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cor-tisol production and on activities of steroidogenic enzymesin human adrenocortical H295R cellsrdquo The Journal of SteroidBiochemistry and Molecular Biology vol 80 no 3 pp 355ndash3632002
[47] D P Dempsher D S Gann and R D Phair ldquoA mechanisticmodel of ACTH-stimulated cortisol secretionrdquo American Jour-nal of Physiology vol 246 no 4 part 2 pp R587ndashR596 1984
[48] N Sarai S Matsuoka and A Noma ldquosimBio a Java package forthe development of detailed cell modelsrdquo Progress in Biophysicsand Molecular Biology vol 90 no 1ndash3 pp 360ndash377 2006
[49] K Levenberg ldquoA method for the solution of certain non-linearproblems in least squaresrdquo Quarterly of Applied Mathematicsvol 2 pp 164ndash168 1944
[50] D W Marquardt ldquoAn algorithm for least-squares estimation ofnonlinear parametersrdquo SIAM Journal on Applied Mathematicsvol 11 no 2 pp 431ndash441 1963
[51] P E Gill and W Murray ldquoAlgorithms for the solution of thenonlinear least-squares problemrdquo SIAM Journal on NumericalAnalysis vol 15 no 5 pp 977ndash992 1978
[52] Y Akimoto R Hasada J Sakuma I Ono and S KobayashildquoGeneration alternation model for real-coded GA using multi-parent proposal and evaluation of just generation gap (JGG)rdquoin Proceedings of the 19th SICE Symposium on DecentralizedAutonomous Systems pp 341ndash346 Tokyo Japan January 2007
[53] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoProposal and evaluation of adaptive real-coded crossoverAREXrdquo Transactions of the Japanese Society for Artificial Intel-ligence vol 24 no 6 pp 446ndash458 2009
Journal of Toxicology 19
[54] S Kobayashi ldquoThe frontiers of real-coded genetic algorithmsrdquoJournal of Japanese Society for Artificial Intelligence vol 24 no1 pp 147ndash162 2009
[55] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoAnalysis of the behavior of MGG and JGG as a selection modelfor real-coded genetic algorithmsrdquo Transactions of the JapaneseSociety for Artificial Intelligence vol 25 no 2 pp 281ndash289 2010
[56] A Struthers H Krum and G H Williams ldquoA comparisonof the aldosterone-blocking agents eplerenone and spironolac-tonerdquo Clinical Cardiology vol 31 no 4 pp 153ndash158 2008
[57] K Yamasaki M Sawaki S Noda et al ldquoComparison of theHershberger assay and androgen receptor binding assay oftwelve chemicalsrdquo Toxicology vol 195 no 2-3 pp 177ndash1862004
[58] C Eil and S K Edelson ldquoThe use of human skin fibroblasts toobtain potency estimates of drug binding to androgen recept-orsrdquo The Journal of Clinical Endocrinology amp Metabolism vol59 no 1 pp 51ndash55 1984
[59] P Ye T Yamashita D M Pollock H Sasano and W ERainey ldquoContrasting effects of eplerenone and spironolac-tone on adrenal cell steroidogenesisrdquo Hormone and MetabolicResearch vol 41 no 1 pp 35ndash39 2009
[60] B D Albertson K L Frederick N C Maronian et al ldquoTheeffect of ketoconazole on steroidogenesis I Leydig cell enzymeactivity in vitrordquo Research Communications in Chemical Patho-logy and Pharmacology vol 61 no 1 pp 17ndash26 1988
[61] B D Albertson N C Maronian K L Frederick et al ldquoTheeffect of ketoconazole on steroidogenesis II Adrenocorticalenzyme activity in vitrordquoResearch Communications in ChemicalPathology and Pharmacology vol 61 no 1 pp 27ndash34 1988
[62] DM Stocco andB J Clark ldquoRegulation of the acute productionof steroids in steroidogenic cellsrdquo Endocrine Reviews vol 17 no3 pp 221ndash244 1996
[63] L K Christenson and J F Strauss III ldquoSteroidogenic acute regu-latory protein (StAR) and the intramitochondrial translocationof cholesterolrdquo Biochimica et Biophysica Acta (BBA)mdashMolecularand Cell Biology of Lipids vol 1529 no 1ndash3 pp 175ndash187 2000
[64] E R Simpson ldquoCholesterol side-chain cleavage cytochromeP450 and the control of steroidogenesisrdquo Molecular and Cel-lular Endocrinology vol 13 no 3 pp 213ndash227 1979
[65] H A LaVoie and S R King ldquoTranscriptional regulation of ster-oidogenic genes STARD1 CYP11A1 and HSD3Brdquo ExperimentalBiology and Medicine vol 234 no 8 pp 880ndash907 2009
[66] F Raucci A DrsquoAniello and M M Di Fiore ldquoStimulation ofandrogen production by d-aspartate through the enhancementof StAR P450scc and 3120573-HSD mRNA levels in vivo rat testisand in culture of immature rat Leydig cellsrdquo Steroids vol 84pp 103ndash110 2014
[67] M Doi Y Takahashi R Komatsu et al ldquoSalt-sensitive hyper-tension in circadian clock-deficient Cry-null mice involves dys-regulated adrenal Hsd3b6rdquo Nature Medicine vol 16 no 1 pp67ndash74 2010
[68] H Okamura M Doi Y Yamaguchi and J-M Fustin ldquoHyper-tension due to loss of clock novel insight from the molecu-lar analysis of Cry1Cry2-deleted micerdquo Current HypertensionReports vol 13 no 2 pp 103ndash108 2011
[69] M Doi F Satoh T Maekawa et al ldquoIsoform-specific mono-clonal antibodies against 3120573-hydroxysteroid dehydrogenaseisomerase family provide markers for subclassification ofhuman primary aldosteronismrdquo The Journal of Clinical Endo-crinology amp Metabolism vol 99 no 2 pp E257ndashE262 2014
[70] S Konosu-Fukaya Y Nakamura F Satoh et al ldquo3120573-hydroxy-steroid dehydrogenase isoforms in human aldosterone-produc-ing adenomardquo Molecular and Cellular Endocrinology vol 408pp 205ndash212 2015
[71] G C Verwoert J Hofland N Amin et al ldquoExpression and genevariation studies deny association of humanHSD3B1 gene withaldosterone production or blood pressurerdquoAmerican Journal ofHypertension vol 28 no 1 pp 113ndash120 2015
[72] D Hough K Storbeck SW P Cloete A C Swart and P SwartldquoRelative contribution of P450c17 towards the acute cortisolresponse lessons from sheep and goatsrdquoMolecular and CellularEndocrinology vol 408 pp 107ndash113 2015
[73] I Sakuma S Suematsu Y Matsuzawa et al ldquoCharacterizationof steroidogenic enzyme expression in aldosterone-producingadenoma a comparison with various human adrenal tumorsrdquoEndocrine Journal vol 60 no 3 pp 329ndash336 2013
[74] A Tong A Jia S Yan Y Zhang Y Xie and G Liu ldquoEctopiccortisol-producing adrenocortical adenoma in the renal hilumhistopathological features and steroidogenic enzyme profilerdquoInternational Journal of Clinical and Experimental Pathologyvol 7 no 7 pp 4415ndash4421 2014
[75] A Hirsch D Hahn P Kempna et al ldquoMetformin inhibitshuman androgen production by regulating steroidogenicenzymes HSD3B2 and CYP17A1 and complex I activity of therespiratory chainrdquo Endocrinology vol 153 no 9 pp 4354ndash43662012
[76] B Xu L Gao Y Cui et al ldquoSET protein up-regulated test-osterone production in the cultured preantral folliclesrdquo Repro-ductive Biology and Endocrinology vol 11 article 9 2013
[77] S Kikuchi D Tominaga M Arita M Takahashi and MTomita ldquoDynamic modeling of genetic networks using geneticalgorithm and S-systemrdquo Bioinformatics vol 19 no 5 pp 643ndash650 2003
[78] N Shikata YMaki M Nakatsui et al ldquoDetermining importantregulatory relations of amino acids from dynamic networkanalysis of plasma amino acidsrdquo Amino Acids vol 38 no 1 pp179ndash187 2010
[79] M Nakatsui T Ueda Y Maki I Ono and M OkamotoldquoMethod for inferring and extracting reliable genetic interac-tions from time-series profile of gene expressionrdquoMathematicalBiosciences vol 215 no 1 pp 105ndash114 2008
[80] A Komori Y Maki M Nakatsui I Ono and M OkamotoldquoEfficient numerical optimization algorithm based on new real-coded genetic algorithm AREX + JGG and application to theinverse problem in systems biologyrdquo Applied Mathematics vol3 no 10 pp 1463ndash1470 2012
[81] J A Roubos G van Straten and A J B van Boxtel ldquoAn evolu-tionary strategy for fed-batch bioreactor optimization conceptsand performancerdquo Journal of Biotechnology vol 67 no 2-3 pp173ndash187 1999
[82] D Sarkar and J M Modak ldquoGenetic algorithms with filters foroptimal control problems in fed-batch bioreactorsrdquo Bioprocessand Biosystems Engineering vol 26 no 5 pp 295ndash306 2004
[83] K Harada K Ikeda J Sakuma I Ono and S KobayashildquoHybridization of genetic algorithm with local search in mul-tiobjective function optimization recommendation of GA thenLSrdquoTransactions of the Japanese Society for Artificial Intelligencevol 21 no 6 pp 482ndash492 2006
Submit your manuscripts athttpwwwhindawicom
PainResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
ToxinsJournal of
VaccinesJournal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AntibioticsInternational Journal of
ToxicologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
StrokeResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Drug DeliveryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Pharmacological Sciences
Tropical MedicineJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AddictionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Emergency Medicine InternationalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Autoimmune Diseases
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anesthesiology Research and Practice
ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Pharmaceutics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
MEDIATORSINFLAMMATION
of
Journal of Toxicology 11
0110
10
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
AN (120583M)
(a)
0110
FN (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(b)
00101
SM (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
100E minus 05
(c)
0110
TG (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B1
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(d)
0110
10100
AGT (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(e)
0110
1025
DDD (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(f)
1010
SP (120583M)
2750005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(g)
0110
10100
MP (120583M)
2110005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(h)
Figure 5 Continued
12 Journal of Toxicology
0110
10
KC (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(i)
0110
10
MC (120583M)
4430005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(j)
0110
10100
DZ (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(k)
Figure 5 Estimated mechanism of action of adrenal toxicants by using the mathematical model of adrenal steroidogenesis in NCI-H295Rcells Mechanisms of action of adrenal toxicity compounds were quantitatively estimated from experimental results of differential steroidprofiling using the mathematical model of adrenal steroidogenesis in NCI-H295R cells The drug action was defined as a scaling factorof enzymatic activity in the simulation model These scaling factors were optimized to fit experimental data by a hybrid optimizationmethod of the RCGA and nonlinear least squares Estimated drug actions by the exposure of vasculotoxic agents acrylonitrile AN (a)fumaronitrile FN (b) salinomycin SM (c) and thioguanine TG (d) and the steroidogenic inhibitors aminoglutethimide AGT (e) op1015840-DDD DDD (f) spironolactone SP (g) metyrapone MP (h) ketoconazole KC (i) miconazole MC (j) and daidzein DZ (k) are shownas a spider radar chart CYP11A1 P450 side chain cleavage enzyme CYP17H 17120572-hydroxylase of CYP17 CYP17L C
17ndash20 lyase of CYP17HSD3B2 3120573-hydroxysteroid dehydrogenase CYP21A2 21-hydroxylase CYP11B1 11120573-hydroxylase CYP11B2 18-hydroxylase and HSD17B317120573-hydroxysteroid dehydrogenase
with previous reports [33 34] 7120572-thiospironolactone whichis synthesized by deacetylation of SP inhibits CYP17H andCYP17L [34] The fact that there were no inhibitions ofCYP17HorCYP17L in our study suggests that SPmight not bedeacetylated to 7120572-thiospironolactone in NCI-H295R cellsRegarding CYP11B1 and CYP11B2 our results were unclearcompared with a previous study It has been shown that30 120583M SP inhibits CYP11B1 and CYP11B2 in human andbovine adrenal mitochondria [35]The cause of CYP11B1 andCYP11B2 inhibition by SP could not be determined in ourstudy which might be due to the lower maximum exposureconcentration of SP than that in the previous reportWe couldnot examine SP concentrations over 10 120583M because theseconcentrations were cytotoxic in NCI-H295R cells
335 MP The mechanism of action of MP in adrenalsteroidogenesis was estimated by dose-dependent inhibitionof CYP11B1 (estimated inhibitions at 1 10 and 100 120583M were571 827 and 982 resp) (Figure 5(h)) MP has beenreported to inhibit CYP11B1 as its major effect and CYP11A1and CYP11B2 as a weak effect [6 29 36ndash39] The results wereable to show that MP is a selective inhibitor of CYP11B1 inthe previous report However the estimated effect of MPat a high concentration 100 120583M as the maximum exposureconcentration was unclear According to the previous reportselectivity of MP for CYP11B1CYP11B2 is about five times[39] In addition 20 120583M MP has a slight inhibition effect onCYP11A1 in H295R cells [29]
336 KC The mechanism of action of KC in adrenalsteroidogenesis was estimated by inhibition of CYP11A1CYP17H CYP17L HSD3B2 CYP21A2 CYP11B1 and
CYP11B2 (estimated inhibitions at 10 120583Mwere 926 943518 830 882 974 and 798 resp) (Figure 5(i))KC has been reported to inhibit CYP11A1 CYP17H CYP17LHSD3B2 CYP21A2 and CYP11B1 [6 29 40ndash43] Our resultswere almost consistent with the previous reports KC inhibitsCYP11A1 CYP17H CYP21A2 and CYP11B1 in NCI-H295Rcells at 10 120583M [29] and CYP17H CYP17L CYP21A2 andCYP11B1 in human adrenal mitochondria and Leydig cellmicrosomes at 2ndash5120583M [60 61] However KC has shownonly weak inhibition of HSD3B2 and HSD17B3 in Leydigcells at the millimolar level [60 61] Regarding CYP11B2 andHSD17B3 we considered that these estimated inhibitions ofKC did have sufficient reliability in terms of quantitative pre-diction precision because ALDO and TESTO concentrationswere less than the lower limit of quantification at 10 120583MKC
337 MC Themechanism of action of MC in adrenal stero-idogenesis was estimated by inhibition of CYP17H CYP17LCYP11B1 and HSD17B3 (estimated inhibitions at 10120583Mwere691 530 764 and 571 resp) (Figure 5(j)) MC hasbeen reported to inhibit not only CYP17H and CYP17L butalsoCYP11A1 CYP21A2 andCYP11B1 [41 44 45]The resultsin the previous reports were able to estimate that MC isa CYP17 inhibitor However CYP11A1 inhibition by MCprobably instead of a reduction in StAR expression wasnot clearly detected in our study using NCI-H295R cellsbecause there were no decreases in the concentrations ofPREG and PROG in the culture medium Indirect inhibitionof CYP11A1 via the peripheral-type benzodiazepine receptorhas been reported in mouse adrenocortical Y-1 cells treatedwith MC in the absence of stimuli by measuring PREGproduction [44] On the other hand reductions of StAR
Journal of Toxicology 13
protein expression andor transport activity without affectingtotal steroid synthesis or CYP11A1 and HSD3B2 enzymeexpression or activities have been reported in (BU)
2cAMP-
stimulated MA-10 Leydig tumor cells treated with MC bymeasuring PROG production [45] Therefore the effect ofMC on the initial reaction in adrenal steroidogenesis fromcholesterol should be different according to the cell type andstimulation condition Inhibition of CYP21A2 and CYP11B1by MC has been reported as decreases in the consumptionof PROG and DCORTICO respectively [41] Inhibition ofCYP11B1 was estimated by the action of MC in this studybut that of CYP21A2 was not detected In the previousexperimental report inhibitory sites by MCmight have beenreflected by inhibition of CYP17H activity because CYP21A2activity was measured as a decrease in labeled PROG
338 DZ The mechanism of action of DZ in adrenalsteroidogenesis was estimated by inhibition of CYP11A1HSD3B2 CYP21A2 CYP11B1 and HSD17B3 (estimated inhi-bitions at 100 120583M were 586 941 965 872 and981 resp) (Figure 5(k)) DZ has been reported to inhibitHSD3B2 and CYP21A2 [46] The results of HSD3B2 andCYP21A2 were consistent with the previous report Howeverinhibitions have not been reported for CYP11A1 CYP11B1and HSD17B3These estimated effects of DZ on CYP11B1 andHSD17B3 were unclear because the concentrations of ALDOand TESTO were less than the lower limit of quantificationat 100 120583M In addition these enzymes act downstream of thestrong action points of DZ such as HSD3B2 and CYP21A2
34 The Simulations and the Systematic Model Analysis
341 Dynamic Sensitivity Analysis of Adrenal SteroidogenesisTo comprehensively understand the dynamics of adrenalsteroidogenesis dynamic sensitivities were calculated forsteroid concentrations secreted by NCI-H295R cells usingour constructed mathematical model of steroidogenesis Theresults of dynamic sensitivity analysis at 72 h of duration and6 h of interval time are presented as a heat-map in Figure 6
The top 10 parameters of the total area under the curveof dynamic sensitivity for cholesterol and the 14 steroids inculture medium were 119881HSD3B2
maxA 119881CYP11A1max 119881CYP21A2maxA 119896loc119891 119896loc119887
119870CYP11A1119898
119881CYP17HmaxA 119896Cholesterol Transport 119896acc119891 and 119881CYP21A2maxB in
order from the top Cholesterol uptake (119896Cholesterol Transport)StAR protein (119896loc
119891) and CYP11A1 (119881CYP11A1max ) which are
determining factors of the capacity for steroidogenesis pro-moted the production of mineralocorticoids (DCORTICOCORTICO and ALDO) and restrained the synthesis of glu-cocorticoids (DCORT and CORT) and sex steroids (DIONETESTO and E1) because of the accumulation of interme-diate molecules in steroidogenesis (PREG HPREG DHEAPROG and HPROG) only by self-activation The dynamicpatterns of the intermediate molecules in steroidogenesiswere mainly dependent on the activity of CYP17H andHSD3B2 with PREG as the substrate of these enzymes inwhich the dynamic sensitivities of 119881CYP17HmaxA for HPREG andHPROG and 119881HSD3B2
maxA for PROG HPROG and DCORTICO
reversed the direction of sensitivity at 49ndash66 h after stimu-lation The dynamic sensitivities of the maximum activitiesof HSD3B2 for PREG (119881HSD3B2
maxA ) and CYP21A2 for PROG(119881CYP21A2maxA ) were related to all steroids at 72 h Almost allmodel parameters had positive sensitivity for downstreamsteroids in the adrenal steroidogenic pathway and negativesensitivity for direct-binding steroids as substrates of thesteroidogenic enzyme The sensitivity of 119881max in all steroido-genic enzymes was relatively higher than 119870
119898for the same
steroid substrate
342 Simulation of the Metabolic Balance of AdrenalSteroidogenesis Pathway To clearly show the property of themetabolic shift between mineralocorticoid and glucocorti-coid biosynthesis we performed two-dimensional parameterscanning of the enzymatic activities of CYP17H and HSD3B2(Figure 7) NCI-H295R cells lost the ability to produce allsteroid hormones when enzymatic activities of CYP17H andHSD3B2 were changed by over 60 and 30 respectivelyActivation ofCYP17HandorHSD3B2 induced themetabolicshift that enhanced the glucocorticoid biosynthesis and devi-ated from the mineralocorticoid biosynthesis On the otherhand inhibition of CYP17H andor HSD3B2 induced themetabolic shift that enhanced the mineralocorticoid biosyn-thesis and deviated from the glucocorticoid biosynthesisMoreover the enzymatic activity of HSD3B2 regulated themetabolic balance of sex steroids and the precursors onadrenal steroidogenesis of NCI-H295R cells E1 TESTO andDIONE were produced by NCI-H295R cells when activatingthe enzymatic activity of HSD3B2 Conversely E2 andDHEAwere produced by NCI-H295R cells when suppressing theenzymatic activity of HSD3B2 The biosynthesis of down-stream steroids in adrenal steroidogenesis pathway such asmineralocorticoids and glucocorticoids was almost com-pletely terminated when the enzymatic activity of HSD3B2was decreased by over 80
4 Discussion
41 Importance of 3120573-HSDActivity in Adrenal SteroidogenesisOur systematic analysis using the mathematical model ofadrenal steroidogenesis in NCI-H295R cells revealed thatthe enzymatic activity of 3120573-HSD controls the dynamics ofadrenal steroidogenesis The activity of the StAR proteincontrols the net capacity of steroidogenesis in steroidogeniccells which is the transport of cholesterol from the outer toinner mitochondrial membranes Both the expression levelsof StAR protein and mRNA are rapidly elevated in responseto stimulation by tropic hormones such as ACTH [62 63]Another important factor in adrenal steroidogenesisis is thecholesterol side chain cleavage enzyme CYP11A1 the firstrate-limiting and hormonally regulated step in the synthesisof all steroid hormones which is conversion of cholesterolto pregnenolone in mitochondria [64] According to ourresults of global sensitivity analysis (Supplementary Table 1and Figure 6(d)) in addition to CYP11A1 and StAR proteins3120573-HSD was one of the key regulators in adrenal steroido-genesis of NCI-H295R cells And also this result suggeststhat a significant regulatory mechanism in steroidogenesis
14 Journal of Toxicology
60544830 4236 66 7218 246 120
Time (hours)
0
1
2
3C
once
ntra
tion
(a)
60544830 4236 66 7218 246 120
Time (hours)
minus05
0
05
minus1
1
Sens
itivi
ty
(b)
0
6
12
18
24
30
36
42
48
54
60
66
72
Sens
itivi
ty fo
r med
ium
conc
entr
atio
ns
of ch
oles
tero
l or s
tero
id h
orm
ones
Dur
atio
n tim
e (ho
urs)
Model parameters
+10
+05
00
minus05
minus10
(c)
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
Parameter number in the mathematical model of steroidogenesis in NCI-H295R cells
CHOL
PREG
HPREG
DHEA
PROG
HPROG
DIONE
DCORTICO
DCORT
CORTICO
CORT
ALDO
TESTO
E1
E2
Chol
este
rol o
r ste
roid
hor
mon
es
0
05
10
15
20
minus05
minus10
minus15
minus20
(d)
Figure 6 Heat-map of the global dynamic sensitivity analysis of adrenal steroid concentrations produced by NCI-H295R cells The globaldynamic sensitivity analysis is a powerful tool to comprehensively understand the dependencies of themodel parameters in themathematicalmodel of the biological complex system Dynamic sensitivities of model parameters in the mathematical model of adrenal steroidogenesisin NCI-H295R cells were calculated for all steroid concentrations in the culture medium every 6 h until 72 h after stimulation To clarify theview of heat-map of global dynamic sensitivity analysis imaginary data of the dynamics of steroid concentrations in the original model (blueline) and perturbed model for sensitivity analysis (red line) were prepared (a) Using this imaginary data the calculated dynamic sensitivities(b) and the visualized dynamic sensitivity as one block of the heat-map (c) were shown respectively By the same method that explainedusing imaginary data the large-scale data of the global dynamic sensitivity analysis on the mathematical model of adrenal steroidogenesisin NCI-H295R cells was comprehensively visualized as a big graph of heat-map (d) Parameter numbers in the horizontal axis are (1)119896Choresterol Transport (2) 119896CEH (3) 119896MTR
119891 (4) 119896MTR
119887 (5) 119896acc
119891 (6) 119896acc
119887 (7) 119896loc
119891 (8) 119896loc
119887 (9) 119896Oxysterol Synthesis (10)119870
CYP11A1119898
(11) 119881CYP11A1max (12)119870CYP17H119898A
(13) 119870CYP17H119898B (14) 119881CYP17HmaxA (15) 119881CYP17HmaxB (16) 119881CYP17HmaxB (17) 119870CYP17L
119898B (18) 119881CYP17LmaxA (19) 119881CYP17LmaxB (20) 119870HSD3B2119898A (21) 119870HSD3B2
119898B (22) 119870HSD3B2119898C (23)
119881HSD3B2maxA (24) 119881HSD3B2
maxB (25) 119881HSD3B2maxB (26) 119870CYP21A2
119898A (27) 119870CYP21A2119898B (28) 119881CYP21A2maxA (29) 119881CYP21A2maxB (30) 119870CYP11B1
119898A (31) 119870CYP11B1119898B (32) 119881CYP11B1maxA
(33) 119881CYP11B1maxB (34) 119896CYP11B2 (35) 119870HSD17B3119898A (36) 119870HSD17B3
119898B (37) 119881HSD17B3maxA (38) 119881HSD17B3
maxB (39) 119870CYP19A1119898A (40) 119870CYP19A1
119898B (41) 119881CYP19A1maxA and (42)119881
CYP19A1maxB PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-
hydroxyprogesteroneDIONE androstenedioneDCORTICO 11-deoxycorticosteroneDCORT 11-deoxycortisol CORTICO corticosteroneCORT cortisol ALDO aldosterone and TESTO testosterone
pathway is very reasonable StAR CYP11A1 and 3120573-HSD(isoforms 1 or 2 in humans) proteins generally respond to thesame hormones that stimulate steroid production throughcommon pathways such as cAMP signaling in adrenal glandsand testes [65 66] Moreover our data also support recentexperimental evidence from clinical and in vivo studiessuggesting that the enzymatic activity of 3120573-HSD plays
an important role in the regulation of mineralocorticoidsynthesis in adrenal steroidogenesis and contributes to hyper-tension caused by abnormal overproduction of aldosterone[67ndash70] Circadian clock-deficient Cry-null mice show salt-sensitive hypertension due to abnormally high synthesis ofaldosterone which is caused by constitutively high expressionof HSD3B6mRNA and protein in the adrenal cortex [67 68]
Journal of Toxicology 15
All steroid producing cells
Mineralocorticoid producing cells Glucocorticoid producing cells
Mineralocorticoid and glucocorticoid producing cells
Terminated steroidogenesis
Sex steroids and precursor sex steroidsDIONE TESTO E1DEHA E2
NCI-H295Rcells at controlcondition
Normalized activities of HSD3B2
Nor
mal
ized
activ
ities
of C
YP17
H
2
18
16
14
12
1
08
06
04
02
00 05 1 15 2
Figure 7 Metabolic categories of the steroidogenic cells determined by the balance of HSD3B2 and CYP17H activitiesThe two-dimensionalparameter scanning analysis by the perturbation of two focused parameters clarifies the interaction and the relationship between the twoparameters in the complex system Functional cellular categories of steroidogenic cells were defined by the levels of mineralocorticoid(ALDO) glucocorticoids (DCORTandCORT) and androgens (DHEAandDIONE) at 72 h after stimulation Enzymatic activities ofHSD3B2andCYP17were normalized by standard values of the simulationmodel inNCI-H295R cells Scanning ranges ofHSD3B2 andCYP17 activitieswere 0ndash200 each 10 Green regions are all 14 steroids produced by NCI-H295R cells Red blue and purple regions are mineralocorticoidproducing cells glucocorticoids-producing cells and both corticoid-producing cells respectively Yellow regions are steroidogenesis of NCI-H295R cells terminated upstream of the adrenal steroidogenesis pathway
Recent clinical observations have revealed predominantexpression of HSD3B2 mRNA and protein in tumor cellsof aldosterone-producing adenoma (APA) and HSD3B1mRNA significantly correlated with CYP11B2 mRNA levelsand plasma aldosterone concentrations in APA patients [6970] However the relationship is unclear and disputed ina small-scale clinical study indicating that genetic variationin HSD3B1 affects blood pressure and hypertension in APApatients [71] The results of our simulation study suggest that3120573-HSD protein (human genes are HSD3B1 and HSD3B2) isone of the determination factors for the dynamic propertyof adrenal steroidogenesis Our results support the clinicalevidence of Doi and colleagues [69] and we believe that theHSD3B1 enzyme has a promising potential as novel drugtarget for endocrine hypertension
42 Metabolic Shift of Adrenal Steroidogenesis and the Contri-butions of HSD3B2 and CYP17H The metabolic propertiesof adrenal steroidogenesis in NCI-H295R cells were revealedby dynamic sensitivity analysis using themathematicalmodel(Figure 6) Mineralocorticoids such as DCORTICO COR-TICO and ALDO and intermediate steroids upstream of
the adrenal steroidogenesis pathway such as PREG HPREGDHEA PROG and HPROG were accelerated by reactionsof cholesterol import (119896Cholesterol Transport) StAR protein (119896MTR
119891
and 119896loc119891) and CYP11A1 (119881CYP11A1max ) On the other hand glu-
cocorticoids such as DCORT and CORT and sex hormonessuch as DIONE TESTO and E1 were suppressed by thesemodel parametersTherefore enhancement of the net adrenalsteroidogenesis capacity which supplies PREG precursor tothe pathway causes a production shift from glucocorticoidsto mineralocorticoids by substrate inhibitions of CYP17HHSD3B2 and CYP21A2 caused by accumulation of initialintermediate steroids such as PREG and PROG Sensitivi-ties of CYP17H (119881CYP17HmaxA ) and HSD3B2 (119881HSD3B2
maxA ) for theproducts were dynamically changed and these parametersdetermined the metabolic balance of downstream steroidsin the adrenal steroidogenesis pathway According to theseresults of dynamic sensitivity analysis of StAR CYP11A1CYP17H and HSD3B2 we suggest that the enhancementof CYP17H and HSD3B2 activity during ACTH stimulationwas important to shift the steroidogenic output away fromALDO biosynthesis towards CORT biosynthesis as wellas adrenal androgen production This suggestion partially
16 Journal of Toxicology
supports a comparative animal study in which molecularand cellular variations in CYP17H activity dramatically affectacute cortisol production resulting in distinct physiologicaland behavioral responses [72]
Results of two-dimensional parameter scanning of theenzymatic activities of CYP17H and HSD3B2 quantitativelyshowed the detail of the metabolic relationship betweenmin-eralocorticoid and glucocorticoid biosynthesis (Figure 7)Particularly the results showed that the balance of theseenzymatic activities was very important for the typicalfunction of NCI-H295R cells namely the ability to produceall steroid hormones NCI-H295R cells lost this functionwhen enzymatic activities of CYP17H and HSD3B2 werechanged by over 60 and 30 respectively In addition theybecame mineralocorticoid (ALDO) secreting cells when theenzymatic activity of CYP17H or HSD3B2 was inhibited byover 50 or glucocorticoid (DCORT and CORT) secretingcells when these enzymes were activated by over 50 Inparticular this analysis also showed that HSD3B2 was a keyplayer in the adrenal steroidogenesis of NCI-H295R cellsbecause HSD3B2 inhibition by over 80 almost completelyinhibited the biosynthesis of downstream steroids The ratioof CYP17A1 to HSD3B2 mRNA expression levels has beenrelated to several endocrine diseases with a low level in APAs[73] and high level in cortisol-producing adenomas [74]Furthermore the expression levels or enzymatic activitiesof CYP17A1 and HSD3B1 have been related to androgenproduction in polycystic ovary syndrome [75 76] Theseclinical studies support our simulation results indicating thatthe balance of enzymatic activity of CYP17H and HSD3B2determines the shift in steroidogenic output to mineralocor-ticoids glucocorticoids or androgens
43 Methodologies of Quantitative Mechanistic Analysis forDrug Discovery According to our results obtained using themathematical model of steroidogenesis in NCI-H295R cellssuch as sensitivity analysis comprehensive analysis based onsystems biology is available to quantitatively estimate themechanism of action of steroidogenic disrupting compoundsfrom differential profiling of adrenal steroid hormonesbecause dynamic patterns of steroid hormones in adrenalsteroidogenesis pathway are highly complex Our proposedmethod of quantitative mechanistic analysis of steroidogenicinhibitors was able to predict known action sites in theadrenal steroidogenesis pathway at only one time point (72 hafter drug exposure) Moreover according to the resultsof sensitivity analysis (Figure 6) 119881max of all steroidogenicenzymes was more sensitive than the 119870
119898 because the
intracellular concentrations of steroid hormones were almostmaintained at sufficiently high levels compared with 119870
119898
values of steroidogenic enzymesThese results suggested thatestimation of the mechanism of action of drugs is moreeffective and detectable when using the influences of 119881maxas the searching parameters such as our proposed methodOur data showed that the proposed method based on asystems biology model is a very powerful tool for exploratoryscreening of steroidogenic disrupting compounds
RCGA as a solver of parameter estimation problemsin systems biology has been applied to biological network
identification of gene regulatory networks and metabolicpathways and optimization of biological processes usingexperimentally observed time-course data [77ndash82] In thisstudy RCGA was useful to estimate the mechanism ofaction of novel pharmaceutical drug candidates for adrenalsteroidogenesis as a new application of RCGA in systemsbiology We had two issues when applying RCGA to thequantitative mechanistic analysis of drug actions Theseissues were the vast calculation cost and multimodality ofquasi-optimum solutions in solving the optimization prob-lem because the mathematical model in systems biologyconsists of many equations and parameters A proposedoptimization strategy using RCGA based on REXJGG wasa highly stable and efficient calculation method for a betterquasi-optimum solution than the unimodal normal distri-bution crossover (UNDX)minimum generation gap (MGG)method that is well applied in the engineering field In addi-tion we expanded the RCGA optimization program basedon REXJGG to a hybrid method of GA and then applieda local search as recommended by Harada and Kobayashi[28 83] A final optimal solution was obtained with a goodconvergence property Because these problems are generalin systems biology studies we suggest that the proposedhybrid method based on REXJGG is a very useful tool forquantitative mechanistic analysis of novel pharmaceuticaldrugs not limited to steroidogenic disrupting compounds
5 Conclusions
Thenovelmathematicalmodel of adrenal steroidogenesis wasconstructed in this study including cholesterol transport anddistribution the C
21-steroid hormone pathway steroid trans-
port and cell proliferation which could reproduce adrenalsteroidogenesis in NCI-H295R cells According to the resultsof dynamic sensitivity analysis using the newmodel HSD3B2plays the most important role in the metabolic balance ofadrenal steroidogenesis in NCI-H295R cells Moreover toquantitatively estimate mechanisms of action of adrenal toxiccompounds we analyzed differential metabolic profiles of12 steroid hormones at 3 days after exposure to 11 adrenaltoxic compounds by using the new mathematical modeland a hybrid optimization method of the RCGA and a localsearch (nonlinear least squares) We could estimate whichsteroidogenic enzymes were affected by these compoundsusing the hybrid optimization method Vasculotoxic agentswere estimated to have no effect according to the resultsobtained by our method In terms of adrenal steroidogenicinhibitors the predicted action sites were approximatelymatched to the target enzymes as reported in the literatureThus our computer-aided method based on a systems biol-ogy approach may be useful to analyze the mechanism ofaction of endocrine-disrupting compounds and to assess thehuman adrenal toxicity of novel pharmaceutical drugs basedon steroid hormone profiling
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Journal of Toxicology 17
Authorsrsquo Contribution
Ryuta Saito Natsuko Terasaki and Naohisa Tsutsui designedthe study Naohisa Tsutsui and Naoya Masutomi supervisedthe experiments Natsuko Terasaki performed all in vitroexperiments and collected dataMakotoYamazaki performedall measurements of steroids Ryuta Saito conceived the insilico strategies analyzed data developed all programs andthe mathematical model and performed all simulation anal-yses with Masahiro Okamoto Ryuta Saito Natsuko TerasakiMakoto Yamazaki NaoyaMasutomi andMasahiroOkamotowrote the paper
Acknowledgments
The authors would like to thank Dr Michael S Breenand Dr Miyuki Breen for helpful discussions concerningprevious mathematical models of adrenal steroidogenesisThis work was partially supported by the Grant-in-Aid forScientific Research on Innovative Areas ldquoSynthetic Biologyrdquo(No 23119001 (MO)) from Ministry of Education CultureSports Science and Technology in Japan
References
[1] P W Harvey D J Everett and C J Springall ldquoAdrenal toxi-cology a strategy for assessment of functional toxicity to theadrenal cortex and steroidogenesisrdquo Journal of Applied Toxi-cology vol 27 no 2 pp 103ndash115 2007
[2] W E Ribelin ldquoThe effects of drugs and chemicals upon thestructure of the adrenal glandrdquo Fundamental and Applied Toxi-cology vol 4 no 1 pp 105ndash119 1984
[3] H D Colby and P A Longhurst ldquoToxicology of the adrenalglandrdquo in Endocrine Toxicology C K Atterwill and J D FlackEds pp 243ndash281 CambridgeUniversity Press Cambridge UK1992
[4] H D Colby ldquoIn vitro assessment of adrenocortical toxicityrdquoJournal of Pharmacological and Toxicological Methods vol 32no 1 pp 1ndash6 1994
[5] T J Rosol J T Yarrington J Latendresse and C C CapenldquoAdrenal gland structure function and mechanisms of toxic-ityrdquo Toxicologic Pathology vol 29 no 1 pp 41ndash48 2001
[6] P W Harvey and D J Everett ldquoThe adrenal cortex and ster-oidogenesis as cellular and molecular targets for toxicity crit-ical omissions from regulatory endocrine disrupter screeningstrategies for human healthrdquo Journal of Applied Toxicology vol23 no 2 pp 81ndash87 2003
[7] A F Gazdar H K Oie C H Shackleton et al ldquoEstablishmentand characterization of a human adrenocortical carcinoma cellline that expresses multiple pathways of steroid biosynthesisrdquoCancer Research vol 50 no 17 pp 5488ndash5496 1990
[8] W E Rainey I M Bird C Sawetawan et al ldquoRegulation ofhuman adrenal carcinoma cell (NCI-H295) production of C19steroidsrdquo The Journal of Clinical Endocrinology amp Metabolismvol 77 no 3 pp 731ndash737 1993
[9] B Staels D W Hum andW L Miller ldquoRegulation of steroido-genesis in NCI-H295 cells a cellular model of the human fetaladrenalrdquo Molecular Endocrinology vol 7 no 3 pp 423ndash4331993
[10] M Hecker J L Newsted M B Murphy et al ldquoHuman adreno-carcinoma (H295R) cells for rapid in vitro determination of
effects on steroidogenesis hormone productionrdquo Toxicologyand Applied Pharmacology vol 217 no 1 pp 114ndash124 2006
[11] A Oskarsson E Ulleras K E Plant J P Hinson and P SGoldfarb ldquoSteroidogenic gene expression in H295R cells andthe human adrenal gland adrenotoxic effects of lindane invitrordquo Journal of Applied Toxicology vol 26 no 6 pp 484ndash4922006
[12] UMuller-Vieira M Angotti and RW Hartman ldquoThe adreno-cortical tumor cell line NCI-H295R as an in vitro screeningsystem for the evaluation of CYP11B2 (aldosterone synthase)and CYP11B1 (steroid-11120573-hydroxylase) inhibitorsrdquoThe Journalof Steroid Biochemistry and Molecular Biology vol 96 no 3-4pp 259ndash270 2005
[13] K Imagawa S Okayama M Takaoka et al ldquoInhibitory effectof efonidipine on aldosterone synthesis and secretion in humanadrenocarcinoma (H295R) cellsrdquo Journal of CardiovascularPharmacology vol 47 no 1 pp 133ndash138 2006
[14] M Voets UMuller-Vieira S Marchais-Oberwinkler and RWHartmann ldquoSynthesis of amidinohydrazones and evaluation oftheir inhibitory effect towards aldosterone synthase (CYP11B2)and the formation of selected steroidsrdquo Archiv der Pharmazievol 337 no 7 pp 411ndash416 2004
[15] F K Nielsen C H Hansen J A Fey et al ldquoH295R cells as amodel for steroidogenic disruption a broader perspective usingsimultaneous chemical analysis of 7 key steroid hormonesrdquoToxicology in Vitro vol 26 no 2 pp 343ndash350 2012
[16] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cortisolproduction and on activities of steroidogenic enzymes inhuman adrenocortical H295R cellsrdquo Journal of Steroid Biochem-istry and Molecular Biology vol 80 no 3 pp 355ndash363 2002
[17] R F Canton J T Sanderson S Nijmeijer A Bergman R JLetcher and M van den Berg ldquoIn vitro effects of brominatedflame retardants and metabolites on CYP17 catalytic activity anovel mechanism of actionrdquo Toxicology and Applied Pharma-cology vol 216 no 2 pp 274ndash281 2006
[18] K Hilscherova P D Jones T Gracia et al ldquoAssessment ofthe effects of chemicals on the expression of ten steroidogenicgenes in the H295R cell line using real-time PCRrdquo ToxicologicalSciences vol 81 no 1 pp 78ndash89 2004
[19] T Gracia K Hilscherova P D Jones et al ldquoThe H295R systemfor evaluation of endocrine-disrupting effectsrdquo Ecotoxicologyand Environmental Safety vol 65 no 3 pp 293ndash305 2006
[20] X Zhang J L NewstedM Hecker E B Higley P D Jones andJ P Giesy ldquoClassification of chemicals based on concentration-dependent toxicological data using ToxClustrdquo EnvironmentalScience and Technology vol 43 no 10 pp 3926ndash3932 2009
[21] M S Breen M Breen N Terasaki M Yamazaki and R BConolly ldquoComputational model of steroidogenesis in humanH295R cells to predict biochemical response to endocrine-active chemicalsmodel development formetyraponerdquoEnviron-mental Health Perspectives vol 118 no 2 pp 265ndash272 2010
[22] M Breen M S Breen N Terasaki M Yamazaki A LLloyd and R B Conolly ldquoMechanistic computational model ofsteroidogenesis in H295R cells role of oxysterols and cell pro-liferation to improve predictability of biochemical response toendocrine active chemical-metyraponerdquo Toxicological Sciencesvol 123 no 1 pp 80ndash93 2011
[23] F B Kraemer ldquoAdrenal cholesterol utilizationrdquo Molecular andCellular Endocrinology vol 265-266 pp 42ndash45 2007
[24] W L Miller ldquoSteroidogenic acute regulatory protein (StAR)a novel mitochondrial cholesterol transporterrdquo Biochimica et
18 Journal of Toxicology
Biophysica Acta (BBA)mdashMolecular and Cell Biology of Lipidsvol 1771 no 6 pp 663ndash676 2007
[25] S Szabo and I T Lippe ldquoAdrenal gland chemically inducedstructural and functional changes in the cortexrdquo ToxicologicPathology vol 17 no 2 pp 317ndash329 1989
[26] C Chen-Pan I-J Pan Y Yamamoto T Sakogawa J Yamadaand YHayashi ldquoPrompt recovery of damaged adrenalmedullaeinduced by salinomycinrdquoToxicologic Pathology vol 27 no 5 pp563ndash572 1999
[27] R J Santen and R I Misbin ldquoAminoglutethimide review ofpharmacology and clinical userdquo Pharmacotherapy vol 1 no 2pp 95ndash120 1981
[28] M J Carella N V Dimitrov V V Gossain L Srivastava andD R Rovner ldquoAdrenal effects of low-dose aminoglutethimidewhen used alone in postmenopausal women with advancedbreast cancerrdquoMetabolism vol 43 no 6 pp 723ndash727 1994
[29] M K Johansson J T Sanderson and B-O Lund ldquoEffects of3-MeSO2-DDE and some CYP inhibitors on glucocorticoidsteroidogenesis in the H295R human adrenocortical carcinomacell linerdquo Toxicology in Vitro vol 16 no 2 pp 113ndash121 2002
[30] Y Touitou A Bogdan J C Legrand and P Desgrez ldquoAmino-glutethimide and glutethimide effects on 18-hydroxycorti-costerone biosynthesis by human and sheep adrenals in vitrordquoActa Endocrinologica vol 80 no 3 pp 517ndash526 1975
[31] M M Hart and J A Straw ldquoStudies on the site of action ofop1015840-DDD in the dog adrenal cortex 1 Inhibition of ACTH-mediated pregnenolone synthesisrdquo Steroids vol 17 no 1ndash5 pp559ndash574 1971
[32] M Ojima M Saitoh N Itoh Y Kusano S Fukuchi and HNaganuma ldquoThe effects of op1015840-DDD on adrenal steroidogen-esis and hepatic steroid metabolismrdquo Nippon Naibunpi Gakkaizasshi vol 61 no 3 pp 168ndash178 1985
[33] L F Canosa and N R Ceballos ldquoEffects of different steroid-biosynthesis inhibitors on the testicular steroidogenesis of thetoad Bufo arenarumrdquo Journal of Comparative Physiology B vol171 no 6 pp 519ndash526 2001
[34] D C Kossor S Kominami S Takemori and H D ColbyldquoRole of the steroid 17 alpha-hydroxylase in spironolactone-mediated destruction of adrenal cytochrome P-450rdquoMolecularPharmacology vol 40 no 2 pp 321ndash325 1991
[35] S C Cheng K Suzuki W Sadee and B W Harding ldquoEffectsof spironolactone canrenone and canrenoate K on cytochromeP450 and 11120573 and 18 hydroxylation in bovine and humanadrenal corticalmitochondriardquo Endocrinology vol 99 no 4 pp1097ndash1106 1976
[36] K Yanagibashi M Haniu J E Shively W H Shen and P HallldquoThe synthesis of aldosterone by the adrenal cortex Two zones(fasciculata and glomerulosa) possess one enzyme for 11120573- 18-hydroxylation and aldehyde synthesisrdquoThe Journal of BiologicalChemistry vol 261 no 8 pp 3556ndash3562 1986
[37] K Morishita H Okumura N Ito and N Takahashi ldquoPrimaryculture system of adrenocortical cells from dogs to evaluatedirect effects of chemicals on steroidogenesisrdquo Toxicology vol165 no 2-3 pp 171ndash178 2001
[38] S W J Lamberts E G Bons H A Bruining and F H de JongldquoDifferential effects of the imidazole derivatives etomidateketoconazole and miconazole and of metyrapone on the secre-tion of cortisol and its precursors by human adrenocorticalcellsrdquo Journal of Pharmacology and Experimental Therapeuticsvol 240 no 1 pp 259ndash264 1987
[39] L Yin S Lucas F Maurer U Kazmaier Q Hu and RW Hartmann ldquoNovel imidazol-1-ylmethyl substituted 1256-tetrahydropyrrolo[321-ij]quinolin-4-ones as potent and selec-tive CYP11B1 inhibitors for the treatment of Cushingrsquos syn-dromerdquo Journal ofMedicinal Chemistry vol 55 no 14 pp 6629ndash6633 2012
[40] M DiMattina N Maronian H Ashby D L Loriaux andB D Albertson ldquoKetoconazole inhibits multiple steroidogenicenzymes involved in androgen biosynthesis in the humanovaryrdquo Fertility and Sterility vol 49 no 1 pp 62ndash65 1988
[41] M Ayub and M J Levell ldquoInhibition of human adrenal steroi-dogenic enzymes in vitro by imidazole drugs including keto-conazolerdquo Journal of Steroid Biochemistry vol 32 no 4 pp 515ndash524 1989
[42] D Engelhardt M M Weber T Miksch F Abedinpour andC Jaspers ldquoThe influence of ketoconazole on human adrenalsteroidogenesis incubation studies with tissue slicesrdquo ClinicalEndocrinology vol 35 no 2 pp 163ndash168 1991
[43] S Bhasin S Sikka T Fielder et al ldquoHormonal effects of keto-conazole in vivo in the male rat mechanism of actionrdquo Endo-crinology vol 118 no 3 pp 1229ndash1232 1986
[44] D M Zisterer and D C Williams ldquoCalmidazolium and otherimidazole compounds affect steroidogenesis in Y1 cells lack ofinvolvement of the peripheral-type benzodiazepine receptorrdquoJournal of Steroid Biochemistry and Molecular Biology vol 60no 3-4 pp 189ndash195 1997
[45] L P Walsh C N Kuratko and D M Stocco ldquoEconazoleand miconazole inhibit steroidogenesis and disrupt steroido-genic acute regulatory (StAR) protein expression post-trans-criptionallyrdquo Journal of Steroid Biochemistry and Molecular Bio-logy vol 75 no 4-5 pp 229ndash236 2000
[46] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cor-tisol production and on activities of steroidogenic enzymesin human adrenocortical H295R cellsrdquo The Journal of SteroidBiochemistry and Molecular Biology vol 80 no 3 pp 355ndash3632002
[47] D P Dempsher D S Gann and R D Phair ldquoA mechanisticmodel of ACTH-stimulated cortisol secretionrdquo American Jour-nal of Physiology vol 246 no 4 part 2 pp R587ndashR596 1984
[48] N Sarai S Matsuoka and A Noma ldquosimBio a Java package forthe development of detailed cell modelsrdquo Progress in Biophysicsand Molecular Biology vol 90 no 1ndash3 pp 360ndash377 2006
[49] K Levenberg ldquoA method for the solution of certain non-linearproblems in least squaresrdquo Quarterly of Applied Mathematicsvol 2 pp 164ndash168 1944
[50] D W Marquardt ldquoAn algorithm for least-squares estimation ofnonlinear parametersrdquo SIAM Journal on Applied Mathematicsvol 11 no 2 pp 431ndash441 1963
[51] P E Gill and W Murray ldquoAlgorithms for the solution of thenonlinear least-squares problemrdquo SIAM Journal on NumericalAnalysis vol 15 no 5 pp 977ndash992 1978
[52] Y Akimoto R Hasada J Sakuma I Ono and S KobayashildquoGeneration alternation model for real-coded GA using multi-parent proposal and evaluation of just generation gap (JGG)rdquoin Proceedings of the 19th SICE Symposium on DecentralizedAutonomous Systems pp 341ndash346 Tokyo Japan January 2007
[53] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoProposal and evaluation of adaptive real-coded crossoverAREXrdquo Transactions of the Japanese Society for Artificial Intel-ligence vol 24 no 6 pp 446ndash458 2009
Journal of Toxicology 19
[54] S Kobayashi ldquoThe frontiers of real-coded genetic algorithmsrdquoJournal of Japanese Society for Artificial Intelligence vol 24 no1 pp 147ndash162 2009
[55] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoAnalysis of the behavior of MGG and JGG as a selection modelfor real-coded genetic algorithmsrdquo Transactions of the JapaneseSociety for Artificial Intelligence vol 25 no 2 pp 281ndash289 2010
[56] A Struthers H Krum and G H Williams ldquoA comparisonof the aldosterone-blocking agents eplerenone and spironolac-tonerdquo Clinical Cardiology vol 31 no 4 pp 153ndash158 2008
[57] K Yamasaki M Sawaki S Noda et al ldquoComparison of theHershberger assay and androgen receptor binding assay oftwelve chemicalsrdquo Toxicology vol 195 no 2-3 pp 177ndash1862004
[58] C Eil and S K Edelson ldquoThe use of human skin fibroblasts toobtain potency estimates of drug binding to androgen recept-orsrdquo The Journal of Clinical Endocrinology amp Metabolism vol59 no 1 pp 51ndash55 1984
[59] P Ye T Yamashita D M Pollock H Sasano and W ERainey ldquoContrasting effects of eplerenone and spironolac-tone on adrenal cell steroidogenesisrdquo Hormone and MetabolicResearch vol 41 no 1 pp 35ndash39 2009
[60] B D Albertson K L Frederick N C Maronian et al ldquoTheeffect of ketoconazole on steroidogenesis I Leydig cell enzymeactivity in vitrordquo Research Communications in Chemical Patho-logy and Pharmacology vol 61 no 1 pp 17ndash26 1988
[61] B D Albertson N C Maronian K L Frederick et al ldquoTheeffect of ketoconazole on steroidogenesis II Adrenocorticalenzyme activity in vitrordquoResearch Communications in ChemicalPathology and Pharmacology vol 61 no 1 pp 27ndash34 1988
[62] DM Stocco andB J Clark ldquoRegulation of the acute productionof steroids in steroidogenic cellsrdquo Endocrine Reviews vol 17 no3 pp 221ndash244 1996
[63] L K Christenson and J F Strauss III ldquoSteroidogenic acute regu-latory protein (StAR) and the intramitochondrial translocationof cholesterolrdquo Biochimica et Biophysica Acta (BBA)mdashMolecularand Cell Biology of Lipids vol 1529 no 1ndash3 pp 175ndash187 2000
[64] E R Simpson ldquoCholesterol side-chain cleavage cytochromeP450 and the control of steroidogenesisrdquo Molecular and Cel-lular Endocrinology vol 13 no 3 pp 213ndash227 1979
[65] H A LaVoie and S R King ldquoTranscriptional regulation of ster-oidogenic genes STARD1 CYP11A1 and HSD3Brdquo ExperimentalBiology and Medicine vol 234 no 8 pp 880ndash907 2009
[66] F Raucci A DrsquoAniello and M M Di Fiore ldquoStimulation ofandrogen production by d-aspartate through the enhancementof StAR P450scc and 3120573-HSD mRNA levels in vivo rat testisand in culture of immature rat Leydig cellsrdquo Steroids vol 84pp 103ndash110 2014
[67] M Doi Y Takahashi R Komatsu et al ldquoSalt-sensitive hyper-tension in circadian clock-deficient Cry-null mice involves dys-regulated adrenal Hsd3b6rdquo Nature Medicine vol 16 no 1 pp67ndash74 2010
[68] H Okamura M Doi Y Yamaguchi and J-M Fustin ldquoHyper-tension due to loss of clock novel insight from the molecu-lar analysis of Cry1Cry2-deleted micerdquo Current HypertensionReports vol 13 no 2 pp 103ndash108 2011
[69] M Doi F Satoh T Maekawa et al ldquoIsoform-specific mono-clonal antibodies against 3120573-hydroxysteroid dehydrogenaseisomerase family provide markers for subclassification ofhuman primary aldosteronismrdquo The Journal of Clinical Endo-crinology amp Metabolism vol 99 no 2 pp E257ndashE262 2014
[70] S Konosu-Fukaya Y Nakamura F Satoh et al ldquo3120573-hydroxy-steroid dehydrogenase isoforms in human aldosterone-produc-ing adenomardquo Molecular and Cellular Endocrinology vol 408pp 205ndash212 2015
[71] G C Verwoert J Hofland N Amin et al ldquoExpression and genevariation studies deny association of humanHSD3B1 gene withaldosterone production or blood pressurerdquoAmerican Journal ofHypertension vol 28 no 1 pp 113ndash120 2015
[72] D Hough K Storbeck SW P Cloete A C Swart and P SwartldquoRelative contribution of P450c17 towards the acute cortisolresponse lessons from sheep and goatsrdquoMolecular and CellularEndocrinology vol 408 pp 107ndash113 2015
[73] I Sakuma S Suematsu Y Matsuzawa et al ldquoCharacterizationof steroidogenic enzyme expression in aldosterone-producingadenoma a comparison with various human adrenal tumorsrdquoEndocrine Journal vol 60 no 3 pp 329ndash336 2013
[74] A Tong A Jia S Yan Y Zhang Y Xie and G Liu ldquoEctopiccortisol-producing adrenocortical adenoma in the renal hilumhistopathological features and steroidogenic enzyme profilerdquoInternational Journal of Clinical and Experimental Pathologyvol 7 no 7 pp 4415ndash4421 2014
[75] A Hirsch D Hahn P Kempna et al ldquoMetformin inhibitshuman androgen production by regulating steroidogenicenzymes HSD3B2 and CYP17A1 and complex I activity of therespiratory chainrdquo Endocrinology vol 153 no 9 pp 4354ndash43662012
[76] B Xu L Gao Y Cui et al ldquoSET protein up-regulated test-osterone production in the cultured preantral folliclesrdquo Repro-ductive Biology and Endocrinology vol 11 article 9 2013
[77] S Kikuchi D Tominaga M Arita M Takahashi and MTomita ldquoDynamic modeling of genetic networks using geneticalgorithm and S-systemrdquo Bioinformatics vol 19 no 5 pp 643ndash650 2003
[78] N Shikata YMaki M Nakatsui et al ldquoDetermining importantregulatory relations of amino acids from dynamic networkanalysis of plasma amino acidsrdquo Amino Acids vol 38 no 1 pp179ndash187 2010
[79] M Nakatsui T Ueda Y Maki I Ono and M OkamotoldquoMethod for inferring and extracting reliable genetic interac-tions from time-series profile of gene expressionrdquoMathematicalBiosciences vol 215 no 1 pp 105ndash114 2008
[80] A Komori Y Maki M Nakatsui I Ono and M OkamotoldquoEfficient numerical optimization algorithm based on new real-coded genetic algorithm AREX + JGG and application to theinverse problem in systems biologyrdquo Applied Mathematics vol3 no 10 pp 1463ndash1470 2012
[81] J A Roubos G van Straten and A J B van Boxtel ldquoAn evolu-tionary strategy for fed-batch bioreactor optimization conceptsand performancerdquo Journal of Biotechnology vol 67 no 2-3 pp173ndash187 1999
[82] D Sarkar and J M Modak ldquoGenetic algorithms with filters foroptimal control problems in fed-batch bioreactorsrdquo Bioprocessand Biosystems Engineering vol 26 no 5 pp 295ndash306 2004
[83] K Harada K Ikeda J Sakuma I Ono and S KobayashildquoHybridization of genetic algorithm with local search in mul-tiobjective function optimization recommendation of GA thenLSrdquoTransactions of the Japanese Society for Artificial Intelligencevol 21 no 6 pp 482ndash492 2006
Submit your manuscripts athttpwwwhindawicom
PainResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
ToxinsJournal of
VaccinesJournal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AntibioticsInternational Journal of
ToxicologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
StrokeResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Drug DeliveryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Pharmacological Sciences
Tropical MedicineJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AddictionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Emergency Medicine InternationalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Autoimmune Diseases
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anesthesiology Research and Practice
ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Pharmaceutics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
MEDIATORSINFLAMMATION
of
12 Journal of Toxicology
0110
10
KC (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(i)
0110
10
MC (120583M)
4430005101520
CYP11A1
CYP17H
CYP17L
HSD3B2
CYP21A2
CYP11B1
CYP11B2
HSD17B3
(j)
0110
10100
DZ (120583M)
0005101520
CYP11A1
CYP17H
CYP17L
HSD3B2CYP21A2
CYP11B1
CYP11B2
HSD17B3
(k)
Figure 5 Estimated mechanism of action of adrenal toxicants by using the mathematical model of adrenal steroidogenesis in NCI-H295Rcells Mechanisms of action of adrenal toxicity compounds were quantitatively estimated from experimental results of differential steroidprofiling using the mathematical model of adrenal steroidogenesis in NCI-H295R cells The drug action was defined as a scaling factorof enzymatic activity in the simulation model These scaling factors were optimized to fit experimental data by a hybrid optimizationmethod of the RCGA and nonlinear least squares Estimated drug actions by the exposure of vasculotoxic agents acrylonitrile AN (a)fumaronitrile FN (b) salinomycin SM (c) and thioguanine TG (d) and the steroidogenic inhibitors aminoglutethimide AGT (e) op1015840-DDD DDD (f) spironolactone SP (g) metyrapone MP (h) ketoconazole KC (i) miconazole MC (j) and daidzein DZ (k) are shownas a spider radar chart CYP11A1 P450 side chain cleavage enzyme CYP17H 17120572-hydroxylase of CYP17 CYP17L C
17ndash20 lyase of CYP17HSD3B2 3120573-hydroxysteroid dehydrogenase CYP21A2 21-hydroxylase CYP11B1 11120573-hydroxylase CYP11B2 18-hydroxylase and HSD17B317120573-hydroxysteroid dehydrogenase
with previous reports [33 34] 7120572-thiospironolactone whichis synthesized by deacetylation of SP inhibits CYP17H andCYP17L [34] The fact that there were no inhibitions ofCYP17HorCYP17L in our study suggests that SPmight not bedeacetylated to 7120572-thiospironolactone in NCI-H295R cellsRegarding CYP11B1 and CYP11B2 our results were unclearcompared with a previous study It has been shown that30 120583M SP inhibits CYP11B1 and CYP11B2 in human andbovine adrenal mitochondria [35]The cause of CYP11B1 andCYP11B2 inhibition by SP could not be determined in ourstudy which might be due to the lower maximum exposureconcentration of SP than that in the previous reportWe couldnot examine SP concentrations over 10 120583M because theseconcentrations were cytotoxic in NCI-H295R cells
335 MP The mechanism of action of MP in adrenalsteroidogenesis was estimated by dose-dependent inhibitionof CYP11B1 (estimated inhibitions at 1 10 and 100 120583M were571 827 and 982 resp) (Figure 5(h)) MP has beenreported to inhibit CYP11B1 as its major effect and CYP11A1and CYP11B2 as a weak effect [6 29 36ndash39] The results wereable to show that MP is a selective inhibitor of CYP11B1 inthe previous report However the estimated effect of MPat a high concentration 100 120583M as the maximum exposureconcentration was unclear According to the previous reportselectivity of MP for CYP11B1CYP11B2 is about five times[39] In addition 20 120583M MP has a slight inhibition effect onCYP11A1 in H295R cells [29]
336 KC The mechanism of action of KC in adrenalsteroidogenesis was estimated by inhibition of CYP11A1CYP17H CYP17L HSD3B2 CYP21A2 CYP11B1 and
CYP11B2 (estimated inhibitions at 10 120583Mwere 926 943518 830 882 974 and 798 resp) (Figure 5(i))KC has been reported to inhibit CYP11A1 CYP17H CYP17LHSD3B2 CYP21A2 and CYP11B1 [6 29 40ndash43] Our resultswere almost consistent with the previous reports KC inhibitsCYP11A1 CYP17H CYP21A2 and CYP11B1 in NCI-H295Rcells at 10 120583M [29] and CYP17H CYP17L CYP21A2 andCYP11B1 in human adrenal mitochondria and Leydig cellmicrosomes at 2ndash5120583M [60 61] However KC has shownonly weak inhibition of HSD3B2 and HSD17B3 in Leydigcells at the millimolar level [60 61] Regarding CYP11B2 andHSD17B3 we considered that these estimated inhibitions ofKC did have sufficient reliability in terms of quantitative pre-diction precision because ALDO and TESTO concentrationswere less than the lower limit of quantification at 10 120583MKC
337 MC Themechanism of action of MC in adrenal stero-idogenesis was estimated by inhibition of CYP17H CYP17LCYP11B1 and HSD17B3 (estimated inhibitions at 10120583Mwere691 530 764 and 571 resp) (Figure 5(j)) MC hasbeen reported to inhibit not only CYP17H and CYP17L butalsoCYP11A1 CYP21A2 andCYP11B1 [41 44 45]The resultsin the previous reports were able to estimate that MC isa CYP17 inhibitor However CYP11A1 inhibition by MCprobably instead of a reduction in StAR expression wasnot clearly detected in our study using NCI-H295R cellsbecause there were no decreases in the concentrations ofPREG and PROG in the culture medium Indirect inhibitionof CYP11A1 via the peripheral-type benzodiazepine receptorhas been reported in mouse adrenocortical Y-1 cells treatedwith MC in the absence of stimuli by measuring PREGproduction [44] On the other hand reductions of StAR
Journal of Toxicology 13
protein expression andor transport activity without affectingtotal steroid synthesis or CYP11A1 and HSD3B2 enzymeexpression or activities have been reported in (BU)
2cAMP-
stimulated MA-10 Leydig tumor cells treated with MC bymeasuring PROG production [45] Therefore the effect ofMC on the initial reaction in adrenal steroidogenesis fromcholesterol should be different according to the cell type andstimulation condition Inhibition of CYP21A2 and CYP11B1by MC has been reported as decreases in the consumptionof PROG and DCORTICO respectively [41] Inhibition ofCYP11B1 was estimated by the action of MC in this studybut that of CYP21A2 was not detected In the previousexperimental report inhibitory sites by MCmight have beenreflected by inhibition of CYP17H activity because CYP21A2activity was measured as a decrease in labeled PROG
338 DZ The mechanism of action of DZ in adrenalsteroidogenesis was estimated by inhibition of CYP11A1HSD3B2 CYP21A2 CYP11B1 and HSD17B3 (estimated inhi-bitions at 100 120583M were 586 941 965 872 and981 resp) (Figure 5(k)) DZ has been reported to inhibitHSD3B2 and CYP21A2 [46] The results of HSD3B2 andCYP21A2 were consistent with the previous report Howeverinhibitions have not been reported for CYP11A1 CYP11B1and HSD17B3These estimated effects of DZ on CYP11B1 andHSD17B3 were unclear because the concentrations of ALDOand TESTO were less than the lower limit of quantificationat 100 120583M In addition these enzymes act downstream of thestrong action points of DZ such as HSD3B2 and CYP21A2
34 The Simulations and the Systematic Model Analysis
341 Dynamic Sensitivity Analysis of Adrenal SteroidogenesisTo comprehensively understand the dynamics of adrenalsteroidogenesis dynamic sensitivities were calculated forsteroid concentrations secreted by NCI-H295R cells usingour constructed mathematical model of steroidogenesis Theresults of dynamic sensitivity analysis at 72 h of duration and6 h of interval time are presented as a heat-map in Figure 6
The top 10 parameters of the total area under the curveof dynamic sensitivity for cholesterol and the 14 steroids inculture medium were 119881HSD3B2
maxA 119881CYP11A1max 119881CYP21A2maxA 119896loc119891 119896loc119887
119870CYP11A1119898
119881CYP17HmaxA 119896Cholesterol Transport 119896acc119891 and 119881CYP21A2maxB in
order from the top Cholesterol uptake (119896Cholesterol Transport)StAR protein (119896loc
119891) and CYP11A1 (119881CYP11A1max ) which are
determining factors of the capacity for steroidogenesis pro-moted the production of mineralocorticoids (DCORTICOCORTICO and ALDO) and restrained the synthesis of glu-cocorticoids (DCORT and CORT) and sex steroids (DIONETESTO and E1) because of the accumulation of interme-diate molecules in steroidogenesis (PREG HPREG DHEAPROG and HPROG) only by self-activation The dynamicpatterns of the intermediate molecules in steroidogenesiswere mainly dependent on the activity of CYP17H andHSD3B2 with PREG as the substrate of these enzymes inwhich the dynamic sensitivities of 119881CYP17HmaxA for HPREG andHPROG and 119881HSD3B2
maxA for PROG HPROG and DCORTICO
reversed the direction of sensitivity at 49ndash66 h after stimu-lation The dynamic sensitivities of the maximum activitiesof HSD3B2 for PREG (119881HSD3B2
maxA ) and CYP21A2 for PROG(119881CYP21A2maxA ) were related to all steroids at 72 h Almost allmodel parameters had positive sensitivity for downstreamsteroids in the adrenal steroidogenic pathway and negativesensitivity for direct-binding steroids as substrates of thesteroidogenic enzyme The sensitivity of 119881max in all steroido-genic enzymes was relatively higher than 119870
119898for the same
steroid substrate
342 Simulation of the Metabolic Balance of AdrenalSteroidogenesis Pathway To clearly show the property of themetabolic shift between mineralocorticoid and glucocorti-coid biosynthesis we performed two-dimensional parameterscanning of the enzymatic activities of CYP17H and HSD3B2(Figure 7) NCI-H295R cells lost the ability to produce allsteroid hormones when enzymatic activities of CYP17H andHSD3B2 were changed by over 60 and 30 respectivelyActivation ofCYP17HandorHSD3B2 induced themetabolicshift that enhanced the glucocorticoid biosynthesis and devi-ated from the mineralocorticoid biosynthesis On the otherhand inhibition of CYP17H andor HSD3B2 induced themetabolic shift that enhanced the mineralocorticoid biosyn-thesis and deviated from the glucocorticoid biosynthesisMoreover the enzymatic activity of HSD3B2 regulated themetabolic balance of sex steroids and the precursors onadrenal steroidogenesis of NCI-H295R cells E1 TESTO andDIONE were produced by NCI-H295R cells when activatingthe enzymatic activity of HSD3B2 Conversely E2 andDHEAwere produced by NCI-H295R cells when suppressing theenzymatic activity of HSD3B2 The biosynthesis of down-stream steroids in adrenal steroidogenesis pathway such asmineralocorticoids and glucocorticoids was almost com-pletely terminated when the enzymatic activity of HSD3B2was decreased by over 80
4 Discussion
41 Importance of 3120573-HSDActivity in Adrenal SteroidogenesisOur systematic analysis using the mathematical model ofadrenal steroidogenesis in NCI-H295R cells revealed thatthe enzymatic activity of 3120573-HSD controls the dynamics ofadrenal steroidogenesis The activity of the StAR proteincontrols the net capacity of steroidogenesis in steroidogeniccells which is the transport of cholesterol from the outer toinner mitochondrial membranes Both the expression levelsof StAR protein and mRNA are rapidly elevated in responseto stimulation by tropic hormones such as ACTH [62 63]Another important factor in adrenal steroidogenesisis is thecholesterol side chain cleavage enzyme CYP11A1 the firstrate-limiting and hormonally regulated step in the synthesisof all steroid hormones which is conversion of cholesterolto pregnenolone in mitochondria [64] According to ourresults of global sensitivity analysis (Supplementary Table 1and Figure 6(d)) in addition to CYP11A1 and StAR proteins3120573-HSD was one of the key regulators in adrenal steroido-genesis of NCI-H295R cells And also this result suggeststhat a significant regulatory mechanism in steroidogenesis
14 Journal of Toxicology
60544830 4236 66 7218 246 120
Time (hours)
0
1
2
3C
once
ntra
tion
(a)
60544830 4236 66 7218 246 120
Time (hours)
minus05
0
05
minus1
1
Sens
itivi
ty
(b)
0
6
12
18
24
30
36
42
48
54
60
66
72
Sens
itivi
ty fo
r med
ium
conc
entr
atio
ns
of ch
oles
tero
l or s
tero
id h
orm
ones
Dur
atio
n tim
e (ho
urs)
Model parameters
+10
+05
00
minus05
minus10
(c)
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
Parameter number in the mathematical model of steroidogenesis in NCI-H295R cells
CHOL
PREG
HPREG
DHEA
PROG
HPROG
DIONE
DCORTICO
DCORT
CORTICO
CORT
ALDO
TESTO
E1
E2
Chol
este
rol o
r ste
roid
hor
mon
es
0
05
10
15
20
minus05
minus10
minus15
minus20
(d)
Figure 6 Heat-map of the global dynamic sensitivity analysis of adrenal steroid concentrations produced by NCI-H295R cells The globaldynamic sensitivity analysis is a powerful tool to comprehensively understand the dependencies of themodel parameters in themathematicalmodel of the biological complex system Dynamic sensitivities of model parameters in the mathematical model of adrenal steroidogenesisin NCI-H295R cells were calculated for all steroid concentrations in the culture medium every 6 h until 72 h after stimulation To clarify theview of heat-map of global dynamic sensitivity analysis imaginary data of the dynamics of steroid concentrations in the original model (blueline) and perturbed model for sensitivity analysis (red line) were prepared (a) Using this imaginary data the calculated dynamic sensitivities(b) and the visualized dynamic sensitivity as one block of the heat-map (c) were shown respectively By the same method that explainedusing imaginary data the large-scale data of the global dynamic sensitivity analysis on the mathematical model of adrenal steroidogenesisin NCI-H295R cells was comprehensively visualized as a big graph of heat-map (d) Parameter numbers in the horizontal axis are (1)119896Choresterol Transport (2) 119896CEH (3) 119896MTR
119891 (4) 119896MTR
119887 (5) 119896acc
119891 (6) 119896acc
119887 (7) 119896loc
119891 (8) 119896loc
119887 (9) 119896Oxysterol Synthesis (10)119870
CYP11A1119898
(11) 119881CYP11A1max (12)119870CYP17H119898A
(13) 119870CYP17H119898B (14) 119881CYP17HmaxA (15) 119881CYP17HmaxB (16) 119881CYP17HmaxB (17) 119870CYP17L
119898B (18) 119881CYP17LmaxA (19) 119881CYP17LmaxB (20) 119870HSD3B2119898A (21) 119870HSD3B2
119898B (22) 119870HSD3B2119898C (23)
119881HSD3B2maxA (24) 119881HSD3B2
maxB (25) 119881HSD3B2maxB (26) 119870CYP21A2
119898A (27) 119870CYP21A2119898B (28) 119881CYP21A2maxA (29) 119881CYP21A2maxB (30) 119870CYP11B1
119898A (31) 119870CYP11B1119898B (32) 119881CYP11B1maxA
(33) 119881CYP11B1maxB (34) 119896CYP11B2 (35) 119870HSD17B3119898A (36) 119870HSD17B3
119898B (37) 119881HSD17B3maxA (38) 119881HSD17B3
maxB (39) 119870CYP19A1119898A (40) 119870CYP19A1
119898B (41) 119881CYP19A1maxA and (42)119881
CYP19A1maxB PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-
hydroxyprogesteroneDIONE androstenedioneDCORTICO 11-deoxycorticosteroneDCORT 11-deoxycortisol CORTICO corticosteroneCORT cortisol ALDO aldosterone and TESTO testosterone
pathway is very reasonable StAR CYP11A1 and 3120573-HSD(isoforms 1 or 2 in humans) proteins generally respond to thesame hormones that stimulate steroid production throughcommon pathways such as cAMP signaling in adrenal glandsand testes [65 66] Moreover our data also support recentexperimental evidence from clinical and in vivo studiessuggesting that the enzymatic activity of 3120573-HSD plays
an important role in the regulation of mineralocorticoidsynthesis in adrenal steroidogenesis and contributes to hyper-tension caused by abnormal overproduction of aldosterone[67ndash70] Circadian clock-deficient Cry-null mice show salt-sensitive hypertension due to abnormally high synthesis ofaldosterone which is caused by constitutively high expressionof HSD3B6mRNA and protein in the adrenal cortex [67 68]
Journal of Toxicology 15
All steroid producing cells
Mineralocorticoid producing cells Glucocorticoid producing cells
Mineralocorticoid and glucocorticoid producing cells
Terminated steroidogenesis
Sex steroids and precursor sex steroidsDIONE TESTO E1DEHA E2
NCI-H295Rcells at controlcondition
Normalized activities of HSD3B2
Nor
mal
ized
activ
ities
of C
YP17
H
2
18
16
14
12
1
08
06
04
02
00 05 1 15 2
Figure 7 Metabolic categories of the steroidogenic cells determined by the balance of HSD3B2 and CYP17H activitiesThe two-dimensionalparameter scanning analysis by the perturbation of two focused parameters clarifies the interaction and the relationship between the twoparameters in the complex system Functional cellular categories of steroidogenic cells were defined by the levels of mineralocorticoid(ALDO) glucocorticoids (DCORTandCORT) and androgens (DHEAandDIONE) at 72 h after stimulation Enzymatic activities ofHSD3B2andCYP17were normalized by standard values of the simulationmodel inNCI-H295R cells Scanning ranges ofHSD3B2 andCYP17 activitieswere 0ndash200 each 10 Green regions are all 14 steroids produced by NCI-H295R cells Red blue and purple regions are mineralocorticoidproducing cells glucocorticoids-producing cells and both corticoid-producing cells respectively Yellow regions are steroidogenesis of NCI-H295R cells terminated upstream of the adrenal steroidogenesis pathway
Recent clinical observations have revealed predominantexpression of HSD3B2 mRNA and protein in tumor cellsof aldosterone-producing adenoma (APA) and HSD3B1mRNA significantly correlated with CYP11B2 mRNA levelsand plasma aldosterone concentrations in APA patients [6970] However the relationship is unclear and disputed ina small-scale clinical study indicating that genetic variationin HSD3B1 affects blood pressure and hypertension in APApatients [71] The results of our simulation study suggest that3120573-HSD protein (human genes are HSD3B1 and HSD3B2) isone of the determination factors for the dynamic propertyof adrenal steroidogenesis Our results support the clinicalevidence of Doi and colleagues [69] and we believe that theHSD3B1 enzyme has a promising potential as novel drugtarget for endocrine hypertension
42 Metabolic Shift of Adrenal Steroidogenesis and the Contri-butions of HSD3B2 and CYP17H The metabolic propertiesof adrenal steroidogenesis in NCI-H295R cells were revealedby dynamic sensitivity analysis using themathematicalmodel(Figure 6) Mineralocorticoids such as DCORTICO COR-TICO and ALDO and intermediate steroids upstream of
the adrenal steroidogenesis pathway such as PREG HPREGDHEA PROG and HPROG were accelerated by reactionsof cholesterol import (119896Cholesterol Transport) StAR protein (119896MTR
119891
and 119896loc119891) and CYP11A1 (119881CYP11A1max ) On the other hand glu-
cocorticoids such as DCORT and CORT and sex hormonessuch as DIONE TESTO and E1 were suppressed by thesemodel parametersTherefore enhancement of the net adrenalsteroidogenesis capacity which supplies PREG precursor tothe pathway causes a production shift from glucocorticoidsto mineralocorticoids by substrate inhibitions of CYP17HHSD3B2 and CYP21A2 caused by accumulation of initialintermediate steroids such as PREG and PROG Sensitivi-ties of CYP17H (119881CYP17HmaxA ) and HSD3B2 (119881HSD3B2
maxA ) for theproducts were dynamically changed and these parametersdetermined the metabolic balance of downstream steroidsin the adrenal steroidogenesis pathway According to theseresults of dynamic sensitivity analysis of StAR CYP11A1CYP17H and HSD3B2 we suggest that the enhancementof CYP17H and HSD3B2 activity during ACTH stimulationwas important to shift the steroidogenic output away fromALDO biosynthesis towards CORT biosynthesis as wellas adrenal androgen production This suggestion partially
16 Journal of Toxicology
supports a comparative animal study in which molecularand cellular variations in CYP17H activity dramatically affectacute cortisol production resulting in distinct physiologicaland behavioral responses [72]
Results of two-dimensional parameter scanning of theenzymatic activities of CYP17H and HSD3B2 quantitativelyshowed the detail of the metabolic relationship betweenmin-eralocorticoid and glucocorticoid biosynthesis (Figure 7)Particularly the results showed that the balance of theseenzymatic activities was very important for the typicalfunction of NCI-H295R cells namely the ability to produceall steroid hormones NCI-H295R cells lost this functionwhen enzymatic activities of CYP17H and HSD3B2 werechanged by over 60 and 30 respectively In addition theybecame mineralocorticoid (ALDO) secreting cells when theenzymatic activity of CYP17H or HSD3B2 was inhibited byover 50 or glucocorticoid (DCORT and CORT) secretingcells when these enzymes were activated by over 50 Inparticular this analysis also showed that HSD3B2 was a keyplayer in the adrenal steroidogenesis of NCI-H295R cellsbecause HSD3B2 inhibition by over 80 almost completelyinhibited the biosynthesis of downstream steroids The ratioof CYP17A1 to HSD3B2 mRNA expression levels has beenrelated to several endocrine diseases with a low level in APAs[73] and high level in cortisol-producing adenomas [74]Furthermore the expression levels or enzymatic activitiesof CYP17A1 and HSD3B1 have been related to androgenproduction in polycystic ovary syndrome [75 76] Theseclinical studies support our simulation results indicating thatthe balance of enzymatic activity of CYP17H and HSD3B2determines the shift in steroidogenic output to mineralocor-ticoids glucocorticoids or androgens
43 Methodologies of Quantitative Mechanistic Analysis forDrug Discovery According to our results obtained using themathematical model of steroidogenesis in NCI-H295R cellssuch as sensitivity analysis comprehensive analysis based onsystems biology is available to quantitatively estimate themechanism of action of steroidogenic disrupting compoundsfrom differential profiling of adrenal steroid hormonesbecause dynamic patterns of steroid hormones in adrenalsteroidogenesis pathway are highly complex Our proposedmethod of quantitative mechanistic analysis of steroidogenicinhibitors was able to predict known action sites in theadrenal steroidogenesis pathway at only one time point (72 hafter drug exposure) Moreover according to the resultsof sensitivity analysis (Figure 6) 119881max of all steroidogenicenzymes was more sensitive than the 119870
119898 because the
intracellular concentrations of steroid hormones were almostmaintained at sufficiently high levels compared with 119870
119898
values of steroidogenic enzymesThese results suggested thatestimation of the mechanism of action of drugs is moreeffective and detectable when using the influences of 119881maxas the searching parameters such as our proposed methodOur data showed that the proposed method based on asystems biology model is a very powerful tool for exploratoryscreening of steroidogenic disrupting compounds
RCGA as a solver of parameter estimation problemsin systems biology has been applied to biological network
identification of gene regulatory networks and metabolicpathways and optimization of biological processes usingexperimentally observed time-course data [77ndash82] In thisstudy RCGA was useful to estimate the mechanism ofaction of novel pharmaceutical drug candidates for adrenalsteroidogenesis as a new application of RCGA in systemsbiology We had two issues when applying RCGA to thequantitative mechanistic analysis of drug actions Theseissues were the vast calculation cost and multimodality ofquasi-optimum solutions in solving the optimization prob-lem because the mathematical model in systems biologyconsists of many equations and parameters A proposedoptimization strategy using RCGA based on REXJGG wasa highly stable and efficient calculation method for a betterquasi-optimum solution than the unimodal normal distri-bution crossover (UNDX)minimum generation gap (MGG)method that is well applied in the engineering field In addi-tion we expanded the RCGA optimization program basedon REXJGG to a hybrid method of GA and then applieda local search as recommended by Harada and Kobayashi[28 83] A final optimal solution was obtained with a goodconvergence property Because these problems are generalin systems biology studies we suggest that the proposedhybrid method based on REXJGG is a very useful tool forquantitative mechanistic analysis of novel pharmaceuticaldrugs not limited to steroidogenic disrupting compounds
5 Conclusions
Thenovelmathematicalmodel of adrenal steroidogenesis wasconstructed in this study including cholesterol transport anddistribution the C
21-steroid hormone pathway steroid trans-
port and cell proliferation which could reproduce adrenalsteroidogenesis in NCI-H295R cells According to the resultsof dynamic sensitivity analysis using the newmodel HSD3B2plays the most important role in the metabolic balance ofadrenal steroidogenesis in NCI-H295R cells Moreover toquantitatively estimate mechanisms of action of adrenal toxiccompounds we analyzed differential metabolic profiles of12 steroid hormones at 3 days after exposure to 11 adrenaltoxic compounds by using the new mathematical modeland a hybrid optimization method of the RCGA and a localsearch (nonlinear least squares) We could estimate whichsteroidogenic enzymes were affected by these compoundsusing the hybrid optimization method Vasculotoxic agentswere estimated to have no effect according to the resultsobtained by our method In terms of adrenal steroidogenicinhibitors the predicted action sites were approximatelymatched to the target enzymes as reported in the literatureThus our computer-aided method based on a systems biol-ogy approach may be useful to analyze the mechanism ofaction of endocrine-disrupting compounds and to assess thehuman adrenal toxicity of novel pharmaceutical drugs basedon steroid hormone profiling
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Journal of Toxicology 17
Authorsrsquo Contribution
Ryuta Saito Natsuko Terasaki and Naohisa Tsutsui designedthe study Naohisa Tsutsui and Naoya Masutomi supervisedthe experiments Natsuko Terasaki performed all in vitroexperiments and collected dataMakotoYamazaki performedall measurements of steroids Ryuta Saito conceived the insilico strategies analyzed data developed all programs andthe mathematical model and performed all simulation anal-yses with Masahiro Okamoto Ryuta Saito Natsuko TerasakiMakoto Yamazaki NaoyaMasutomi andMasahiroOkamotowrote the paper
Acknowledgments
The authors would like to thank Dr Michael S Breenand Dr Miyuki Breen for helpful discussions concerningprevious mathematical models of adrenal steroidogenesisThis work was partially supported by the Grant-in-Aid forScientific Research on Innovative Areas ldquoSynthetic Biologyrdquo(No 23119001 (MO)) from Ministry of Education CultureSports Science and Technology in Japan
References
[1] P W Harvey D J Everett and C J Springall ldquoAdrenal toxi-cology a strategy for assessment of functional toxicity to theadrenal cortex and steroidogenesisrdquo Journal of Applied Toxi-cology vol 27 no 2 pp 103ndash115 2007
[2] W E Ribelin ldquoThe effects of drugs and chemicals upon thestructure of the adrenal glandrdquo Fundamental and Applied Toxi-cology vol 4 no 1 pp 105ndash119 1984
[3] H D Colby and P A Longhurst ldquoToxicology of the adrenalglandrdquo in Endocrine Toxicology C K Atterwill and J D FlackEds pp 243ndash281 CambridgeUniversity Press Cambridge UK1992
[4] H D Colby ldquoIn vitro assessment of adrenocortical toxicityrdquoJournal of Pharmacological and Toxicological Methods vol 32no 1 pp 1ndash6 1994
[5] T J Rosol J T Yarrington J Latendresse and C C CapenldquoAdrenal gland structure function and mechanisms of toxic-ityrdquo Toxicologic Pathology vol 29 no 1 pp 41ndash48 2001
[6] P W Harvey and D J Everett ldquoThe adrenal cortex and ster-oidogenesis as cellular and molecular targets for toxicity crit-ical omissions from regulatory endocrine disrupter screeningstrategies for human healthrdquo Journal of Applied Toxicology vol23 no 2 pp 81ndash87 2003
[7] A F Gazdar H K Oie C H Shackleton et al ldquoEstablishmentand characterization of a human adrenocortical carcinoma cellline that expresses multiple pathways of steroid biosynthesisrdquoCancer Research vol 50 no 17 pp 5488ndash5496 1990
[8] W E Rainey I M Bird C Sawetawan et al ldquoRegulation ofhuman adrenal carcinoma cell (NCI-H295) production of C19steroidsrdquo The Journal of Clinical Endocrinology amp Metabolismvol 77 no 3 pp 731ndash737 1993
[9] B Staels D W Hum andW L Miller ldquoRegulation of steroido-genesis in NCI-H295 cells a cellular model of the human fetaladrenalrdquo Molecular Endocrinology vol 7 no 3 pp 423ndash4331993
[10] M Hecker J L Newsted M B Murphy et al ldquoHuman adreno-carcinoma (H295R) cells for rapid in vitro determination of
effects on steroidogenesis hormone productionrdquo Toxicologyand Applied Pharmacology vol 217 no 1 pp 114ndash124 2006
[11] A Oskarsson E Ulleras K E Plant J P Hinson and P SGoldfarb ldquoSteroidogenic gene expression in H295R cells andthe human adrenal gland adrenotoxic effects of lindane invitrordquo Journal of Applied Toxicology vol 26 no 6 pp 484ndash4922006
[12] UMuller-Vieira M Angotti and RW Hartman ldquoThe adreno-cortical tumor cell line NCI-H295R as an in vitro screeningsystem for the evaluation of CYP11B2 (aldosterone synthase)and CYP11B1 (steroid-11120573-hydroxylase) inhibitorsrdquoThe Journalof Steroid Biochemistry and Molecular Biology vol 96 no 3-4pp 259ndash270 2005
[13] K Imagawa S Okayama M Takaoka et al ldquoInhibitory effectof efonidipine on aldosterone synthesis and secretion in humanadrenocarcinoma (H295R) cellsrdquo Journal of CardiovascularPharmacology vol 47 no 1 pp 133ndash138 2006
[14] M Voets UMuller-Vieira S Marchais-Oberwinkler and RWHartmann ldquoSynthesis of amidinohydrazones and evaluation oftheir inhibitory effect towards aldosterone synthase (CYP11B2)and the formation of selected steroidsrdquo Archiv der Pharmazievol 337 no 7 pp 411ndash416 2004
[15] F K Nielsen C H Hansen J A Fey et al ldquoH295R cells as amodel for steroidogenic disruption a broader perspective usingsimultaneous chemical analysis of 7 key steroid hormonesrdquoToxicology in Vitro vol 26 no 2 pp 343ndash350 2012
[16] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cortisolproduction and on activities of steroidogenic enzymes inhuman adrenocortical H295R cellsrdquo Journal of Steroid Biochem-istry and Molecular Biology vol 80 no 3 pp 355ndash363 2002
[17] R F Canton J T Sanderson S Nijmeijer A Bergman R JLetcher and M van den Berg ldquoIn vitro effects of brominatedflame retardants and metabolites on CYP17 catalytic activity anovel mechanism of actionrdquo Toxicology and Applied Pharma-cology vol 216 no 2 pp 274ndash281 2006
[18] K Hilscherova P D Jones T Gracia et al ldquoAssessment ofthe effects of chemicals on the expression of ten steroidogenicgenes in the H295R cell line using real-time PCRrdquo ToxicologicalSciences vol 81 no 1 pp 78ndash89 2004
[19] T Gracia K Hilscherova P D Jones et al ldquoThe H295R systemfor evaluation of endocrine-disrupting effectsrdquo Ecotoxicologyand Environmental Safety vol 65 no 3 pp 293ndash305 2006
[20] X Zhang J L NewstedM Hecker E B Higley P D Jones andJ P Giesy ldquoClassification of chemicals based on concentration-dependent toxicological data using ToxClustrdquo EnvironmentalScience and Technology vol 43 no 10 pp 3926ndash3932 2009
[21] M S Breen M Breen N Terasaki M Yamazaki and R BConolly ldquoComputational model of steroidogenesis in humanH295R cells to predict biochemical response to endocrine-active chemicalsmodel development formetyraponerdquoEnviron-mental Health Perspectives vol 118 no 2 pp 265ndash272 2010
[22] M Breen M S Breen N Terasaki M Yamazaki A LLloyd and R B Conolly ldquoMechanistic computational model ofsteroidogenesis in H295R cells role of oxysterols and cell pro-liferation to improve predictability of biochemical response toendocrine active chemical-metyraponerdquo Toxicological Sciencesvol 123 no 1 pp 80ndash93 2011
[23] F B Kraemer ldquoAdrenal cholesterol utilizationrdquo Molecular andCellular Endocrinology vol 265-266 pp 42ndash45 2007
[24] W L Miller ldquoSteroidogenic acute regulatory protein (StAR)a novel mitochondrial cholesterol transporterrdquo Biochimica et
18 Journal of Toxicology
Biophysica Acta (BBA)mdashMolecular and Cell Biology of Lipidsvol 1771 no 6 pp 663ndash676 2007
[25] S Szabo and I T Lippe ldquoAdrenal gland chemically inducedstructural and functional changes in the cortexrdquo ToxicologicPathology vol 17 no 2 pp 317ndash329 1989
[26] C Chen-Pan I-J Pan Y Yamamoto T Sakogawa J Yamadaand YHayashi ldquoPrompt recovery of damaged adrenalmedullaeinduced by salinomycinrdquoToxicologic Pathology vol 27 no 5 pp563ndash572 1999
[27] R J Santen and R I Misbin ldquoAminoglutethimide review ofpharmacology and clinical userdquo Pharmacotherapy vol 1 no 2pp 95ndash120 1981
[28] M J Carella N V Dimitrov V V Gossain L Srivastava andD R Rovner ldquoAdrenal effects of low-dose aminoglutethimidewhen used alone in postmenopausal women with advancedbreast cancerrdquoMetabolism vol 43 no 6 pp 723ndash727 1994
[29] M K Johansson J T Sanderson and B-O Lund ldquoEffects of3-MeSO2-DDE and some CYP inhibitors on glucocorticoidsteroidogenesis in the H295R human adrenocortical carcinomacell linerdquo Toxicology in Vitro vol 16 no 2 pp 113ndash121 2002
[30] Y Touitou A Bogdan J C Legrand and P Desgrez ldquoAmino-glutethimide and glutethimide effects on 18-hydroxycorti-costerone biosynthesis by human and sheep adrenals in vitrordquoActa Endocrinologica vol 80 no 3 pp 517ndash526 1975
[31] M M Hart and J A Straw ldquoStudies on the site of action ofop1015840-DDD in the dog adrenal cortex 1 Inhibition of ACTH-mediated pregnenolone synthesisrdquo Steroids vol 17 no 1ndash5 pp559ndash574 1971
[32] M Ojima M Saitoh N Itoh Y Kusano S Fukuchi and HNaganuma ldquoThe effects of op1015840-DDD on adrenal steroidogen-esis and hepatic steroid metabolismrdquo Nippon Naibunpi Gakkaizasshi vol 61 no 3 pp 168ndash178 1985
[33] L F Canosa and N R Ceballos ldquoEffects of different steroid-biosynthesis inhibitors on the testicular steroidogenesis of thetoad Bufo arenarumrdquo Journal of Comparative Physiology B vol171 no 6 pp 519ndash526 2001
[34] D C Kossor S Kominami S Takemori and H D ColbyldquoRole of the steroid 17 alpha-hydroxylase in spironolactone-mediated destruction of adrenal cytochrome P-450rdquoMolecularPharmacology vol 40 no 2 pp 321ndash325 1991
[35] S C Cheng K Suzuki W Sadee and B W Harding ldquoEffectsof spironolactone canrenone and canrenoate K on cytochromeP450 and 11120573 and 18 hydroxylation in bovine and humanadrenal corticalmitochondriardquo Endocrinology vol 99 no 4 pp1097ndash1106 1976
[36] K Yanagibashi M Haniu J E Shively W H Shen and P HallldquoThe synthesis of aldosterone by the adrenal cortex Two zones(fasciculata and glomerulosa) possess one enzyme for 11120573- 18-hydroxylation and aldehyde synthesisrdquoThe Journal of BiologicalChemistry vol 261 no 8 pp 3556ndash3562 1986
[37] K Morishita H Okumura N Ito and N Takahashi ldquoPrimaryculture system of adrenocortical cells from dogs to evaluatedirect effects of chemicals on steroidogenesisrdquo Toxicology vol165 no 2-3 pp 171ndash178 2001
[38] S W J Lamberts E G Bons H A Bruining and F H de JongldquoDifferential effects of the imidazole derivatives etomidateketoconazole and miconazole and of metyrapone on the secre-tion of cortisol and its precursors by human adrenocorticalcellsrdquo Journal of Pharmacology and Experimental Therapeuticsvol 240 no 1 pp 259ndash264 1987
[39] L Yin S Lucas F Maurer U Kazmaier Q Hu and RW Hartmann ldquoNovel imidazol-1-ylmethyl substituted 1256-tetrahydropyrrolo[321-ij]quinolin-4-ones as potent and selec-tive CYP11B1 inhibitors for the treatment of Cushingrsquos syn-dromerdquo Journal ofMedicinal Chemistry vol 55 no 14 pp 6629ndash6633 2012
[40] M DiMattina N Maronian H Ashby D L Loriaux andB D Albertson ldquoKetoconazole inhibits multiple steroidogenicenzymes involved in androgen biosynthesis in the humanovaryrdquo Fertility and Sterility vol 49 no 1 pp 62ndash65 1988
[41] M Ayub and M J Levell ldquoInhibition of human adrenal steroi-dogenic enzymes in vitro by imidazole drugs including keto-conazolerdquo Journal of Steroid Biochemistry vol 32 no 4 pp 515ndash524 1989
[42] D Engelhardt M M Weber T Miksch F Abedinpour andC Jaspers ldquoThe influence of ketoconazole on human adrenalsteroidogenesis incubation studies with tissue slicesrdquo ClinicalEndocrinology vol 35 no 2 pp 163ndash168 1991
[43] S Bhasin S Sikka T Fielder et al ldquoHormonal effects of keto-conazole in vivo in the male rat mechanism of actionrdquo Endo-crinology vol 118 no 3 pp 1229ndash1232 1986
[44] D M Zisterer and D C Williams ldquoCalmidazolium and otherimidazole compounds affect steroidogenesis in Y1 cells lack ofinvolvement of the peripheral-type benzodiazepine receptorrdquoJournal of Steroid Biochemistry and Molecular Biology vol 60no 3-4 pp 189ndash195 1997
[45] L P Walsh C N Kuratko and D M Stocco ldquoEconazoleand miconazole inhibit steroidogenesis and disrupt steroido-genic acute regulatory (StAR) protein expression post-trans-criptionallyrdquo Journal of Steroid Biochemistry and Molecular Bio-logy vol 75 no 4-5 pp 229ndash236 2000
[46] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cor-tisol production and on activities of steroidogenic enzymesin human adrenocortical H295R cellsrdquo The Journal of SteroidBiochemistry and Molecular Biology vol 80 no 3 pp 355ndash3632002
[47] D P Dempsher D S Gann and R D Phair ldquoA mechanisticmodel of ACTH-stimulated cortisol secretionrdquo American Jour-nal of Physiology vol 246 no 4 part 2 pp R587ndashR596 1984
[48] N Sarai S Matsuoka and A Noma ldquosimBio a Java package forthe development of detailed cell modelsrdquo Progress in Biophysicsand Molecular Biology vol 90 no 1ndash3 pp 360ndash377 2006
[49] K Levenberg ldquoA method for the solution of certain non-linearproblems in least squaresrdquo Quarterly of Applied Mathematicsvol 2 pp 164ndash168 1944
[50] D W Marquardt ldquoAn algorithm for least-squares estimation ofnonlinear parametersrdquo SIAM Journal on Applied Mathematicsvol 11 no 2 pp 431ndash441 1963
[51] P E Gill and W Murray ldquoAlgorithms for the solution of thenonlinear least-squares problemrdquo SIAM Journal on NumericalAnalysis vol 15 no 5 pp 977ndash992 1978
[52] Y Akimoto R Hasada J Sakuma I Ono and S KobayashildquoGeneration alternation model for real-coded GA using multi-parent proposal and evaluation of just generation gap (JGG)rdquoin Proceedings of the 19th SICE Symposium on DecentralizedAutonomous Systems pp 341ndash346 Tokyo Japan January 2007
[53] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoProposal and evaluation of adaptive real-coded crossoverAREXrdquo Transactions of the Japanese Society for Artificial Intel-ligence vol 24 no 6 pp 446ndash458 2009
Journal of Toxicology 19
[54] S Kobayashi ldquoThe frontiers of real-coded genetic algorithmsrdquoJournal of Japanese Society for Artificial Intelligence vol 24 no1 pp 147ndash162 2009
[55] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoAnalysis of the behavior of MGG and JGG as a selection modelfor real-coded genetic algorithmsrdquo Transactions of the JapaneseSociety for Artificial Intelligence vol 25 no 2 pp 281ndash289 2010
[56] A Struthers H Krum and G H Williams ldquoA comparisonof the aldosterone-blocking agents eplerenone and spironolac-tonerdquo Clinical Cardiology vol 31 no 4 pp 153ndash158 2008
[57] K Yamasaki M Sawaki S Noda et al ldquoComparison of theHershberger assay and androgen receptor binding assay oftwelve chemicalsrdquo Toxicology vol 195 no 2-3 pp 177ndash1862004
[58] C Eil and S K Edelson ldquoThe use of human skin fibroblasts toobtain potency estimates of drug binding to androgen recept-orsrdquo The Journal of Clinical Endocrinology amp Metabolism vol59 no 1 pp 51ndash55 1984
[59] P Ye T Yamashita D M Pollock H Sasano and W ERainey ldquoContrasting effects of eplerenone and spironolac-tone on adrenal cell steroidogenesisrdquo Hormone and MetabolicResearch vol 41 no 1 pp 35ndash39 2009
[60] B D Albertson K L Frederick N C Maronian et al ldquoTheeffect of ketoconazole on steroidogenesis I Leydig cell enzymeactivity in vitrordquo Research Communications in Chemical Patho-logy and Pharmacology vol 61 no 1 pp 17ndash26 1988
[61] B D Albertson N C Maronian K L Frederick et al ldquoTheeffect of ketoconazole on steroidogenesis II Adrenocorticalenzyme activity in vitrordquoResearch Communications in ChemicalPathology and Pharmacology vol 61 no 1 pp 27ndash34 1988
[62] DM Stocco andB J Clark ldquoRegulation of the acute productionof steroids in steroidogenic cellsrdquo Endocrine Reviews vol 17 no3 pp 221ndash244 1996
[63] L K Christenson and J F Strauss III ldquoSteroidogenic acute regu-latory protein (StAR) and the intramitochondrial translocationof cholesterolrdquo Biochimica et Biophysica Acta (BBA)mdashMolecularand Cell Biology of Lipids vol 1529 no 1ndash3 pp 175ndash187 2000
[64] E R Simpson ldquoCholesterol side-chain cleavage cytochromeP450 and the control of steroidogenesisrdquo Molecular and Cel-lular Endocrinology vol 13 no 3 pp 213ndash227 1979
[65] H A LaVoie and S R King ldquoTranscriptional regulation of ster-oidogenic genes STARD1 CYP11A1 and HSD3Brdquo ExperimentalBiology and Medicine vol 234 no 8 pp 880ndash907 2009
[66] F Raucci A DrsquoAniello and M M Di Fiore ldquoStimulation ofandrogen production by d-aspartate through the enhancementof StAR P450scc and 3120573-HSD mRNA levels in vivo rat testisand in culture of immature rat Leydig cellsrdquo Steroids vol 84pp 103ndash110 2014
[67] M Doi Y Takahashi R Komatsu et al ldquoSalt-sensitive hyper-tension in circadian clock-deficient Cry-null mice involves dys-regulated adrenal Hsd3b6rdquo Nature Medicine vol 16 no 1 pp67ndash74 2010
[68] H Okamura M Doi Y Yamaguchi and J-M Fustin ldquoHyper-tension due to loss of clock novel insight from the molecu-lar analysis of Cry1Cry2-deleted micerdquo Current HypertensionReports vol 13 no 2 pp 103ndash108 2011
[69] M Doi F Satoh T Maekawa et al ldquoIsoform-specific mono-clonal antibodies against 3120573-hydroxysteroid dehydrogenaseisomerase family provide markers for subclassification ofhuman primary aldosteronismrdquo The Journal of Clinical Endo-crinology amp Metabolism vol 99 no 2 pp E257ndashE262 2014
[70] S Konosu-Fukaya Y Nakamura F Satoh et al ldquo3120573-hydroxy-steroid dehydrogenase isoforms in human aldosterone-produc-ing adenomardquo Molecular and Cellular Endocrinology vol 408pp 205ndash212 2015
[71] G C Verwoert J Hofland N Amin et al ldquoExpression and genevariation studies deny association of humanHSD3B1 gene withaldosterone production or blood pressurerdquoAmerican Journal ofHypertension vol 28 no 1 pp 113ndash120 2015
[72] D Hough K Storbeck SW P Cloete A C Swart and P SwartldquoRelative contribution of P450c17 towards the acute cortisolresponse lessons from sheep and goatsrdquoMolecular and CellularEndocrinology vol 408 pp 107ndash113 2015
[73] I Sakuma S Suematsu Y Matsuzawa et al ldquoCharacterizationof steroidogenic enzyme expression in aldosterone-producingadenoma a comparison with various human adrenal tumorsrdquoEndocrine Journal vol 60 no 3 pp 329ndash336 2013
[74] A Tong A Jia S Yan Y Zhang Y Xie and G Liu ldquoEctopiccortisol-producing adrenocortical adenoma in the renal hilumhistopathological features and steroidogenic enzyme profilerdquoInternational Journal of Clinical and Experimental Pathologyvol 7 no 7 pp 4415ndash4421 2014
[75] A Hirsch D Hahn P Kempna et al ldquoMetformin inhibitshuman androgen production by regulating steroidogenicenzymes HSD3B2 and CYP17A1 and complex I activity of therespiratory chainrdquo Endocrinology vol 153 no 9 pp 4354ndash43662012
[76] B Xu L Gao Y Cui et al ldquoSET protein up-regulated test-osterone production in the cultured preantral folliclesrdquo Repro-ductive Biology and Endocrinology vol 11 article 9 2013
[77] S Kikuchi D Tominaga M Arita M Takahashi and MTomita ldquoDynamic modeling of genetic networks using geneticalgorithm and S-systemrdquo Bioinformatics vol 19 no 5 pp 643ndash650 2003
[78] N Shikata YMaki M Nakatsui et al ldquoDetermining importantregulatory relations of amino acids from dynamic networkanalysis of plasma amino acidsrdquo Amino Acids vol 38 no 1 pp179ndash187 2010
[79] M Nakatsui T Ueda Y Maki I Ono and M OkamotoldquoMethod for inferring and extracting reliable genetic interac-tions from time-series profile of gene expressionrdquoMathematicalBiosciences vol 215 no 1 pp 105ndash114 2008
[80] A Komori Y Maki M Nakatsui I Ono and M OkamotoldquoEfficient numerical optimization algorithm based on new real-coded genetic algorithm AREX + JGG and application to theinverse problem in systems biologyrdquo Applied Mathematics vol3 no 10 pp 1463ndash1470 2012
[81] J A Roubos G van Straten and A J B van Boxtel ldquoAn evolu-tionary strategy for fed-batch bioreactor optimization conceptsand performancerdquo Journal of Biotechnology vol 67 no 2-3 pp173ndash187 1999
[82] D Sarkar and J M Modak ldquoGenetic algorithms with filters foroptimal control problems in fed-batch bioreactorsrdquo Bioprocessand Biosystems Engineering vol 26 no 5 pp 295ndash306 2004
[83] K Harada K Ikeda J Sakuma I Ono and S KobayashildquoHybridization of genetic algorithm with local search in mul-tiobjective function optimization recommendation of GA thenLSrdquoTransactions of the Japanese Society for Artificial Intelligencevol 21 no 6 pp 482ndash492 2006
Submit your manuscripts athttpwwwhindawicom
PainResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
ToxinsJournal of
VaccinesJournal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AntibioticsInternational Journal of
ToxicologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
StrokeResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Drug DeliveryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Pharmacological Sciences
Tropical MedicineJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AddictionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Emergency Medicine InternationalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Autoimmune Diseases
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anesthesiology Research and Practice
ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Pharmaceutics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
MEDIATORSINFLAMMATION
of
Journal of Toxicology 13
protein expression andor transport activity without affectingtotal steroid synthesis or CYP11A1 and HSD3B2 enzymeexpression or activities have been reported in (BU)
2cAMP-
stimulated MA-10 Leydig tumor cells treated with MC bymeasuring PROG production [45] Therefore the effect ofMC on the initial reaction in adrenal steroidogenesis fromcholesterol should be different according to the cell type andstimulation condition Inhibition of CYP21A2 and CYP11B1by MC has been reported as decreases in the consumptionof PROG and DCORTICO respectively [41] Inhibition ofCYP11B1 was estimated by the action of MC in this studybut that of CYP21A2 was not detected In the previousexperimental report inhibitory sites by MCmight have beenreflected by inhibition of CYP17H activity because CYP21A2activity was measured as a decrease in labeled PROG
338 DZ The mechanism of action of DZ in adrenalsteroidogenesis was estimated by inhibition of CYP11A1HSD3B2 CYP21A2 CYP11B1 and HSD17B3 (estimated inhi-bitions at 100 120583M were 586 941 965 872 and981 resp) (Figure 5(k)) DZ has been reported to inhibitHSD3B2 and CYP21A2 [46] The results of HSD3B2 andCYP21A2 were consistent with the previous report Howeverinhibitions have not been reported for CYP11A1 CYP11B1and HSD17B3These estimated effects of DZ on CYP11B1 andHSD17B3 were unclear because the concentrations of ALDOand TESTO were less than the lower limit of quantificationat 100 120583M In addition these enzymes act downstream of thestrong action points of DZ such as HSD3B2 and CYP21A2
34 The Simulations and the Systematic Model Analysis
341 Dynamic Sensitivity Analysis of Adrenal SteroidogenesisTo comprehensively understand the dynamics of adrenalsteroidogenesis dynamic sensitivities were calculated forsteroid concentrations secreted by NCI-H295R cells usingour constructed mathematical model of steroidogenesis Theresults of dynamic sensitivity analysis at 72 h of duration and6 h of interval time are presented as a heat-map in Figure 6
The top 10 parameters of the total area under the curveof dynamic sensitivity for cholesterol and the 14 steroids inculture medium were 119881HSD3B2
maxA 119881CYP11A1max 119881CYP21A2maxA 119896loc119891 119896loc119887
119870CYP11A1119898
119881CYP17HmaxA 119896Cholesterol Transport 119896acc119891 and 119881CYP21A2maxB in
order from the top Cholesterol uptake (119896Cholesterol Transport)StAR protein (119896loc
119891) and CYP11A1 (119881CYP11A1max ) which are
determining factors of the capacity for steroidogenesis pro-moted the production of mineralocorticoids (DCORTICOCORTICO and ALDO) and restrained the synthesis of glu-cocorticoids (DCORT and CORT) and sex steroids (DIONETESTO and E1) because of the accumulation of interme-diate molecules in steroidogenesis (PREG HPREG DHEAPROG and HPROG) only by self-activation The dynamicpatterns of the intermediate molecules in steroidogenesiswere mainly dependent on the activity of CYP17H andHSD3B2 with PREG as the substrate of these enzymes inwhich the dynamic sensitivities of 119881CYP17HmaxA for HPREG andHPROG and 119881HSD3B2
maxA for PROG HPROG and DCORTICO
reversed the direction of sensitivity at 49ndash66 h after stimu-lation The dynamic sensitivities of the maximum activitiesof HSD3B2 for PREG (119881HSD3B2
maxA ) and CYP21A2 for PROG(119881CYP21A2maxA ) were related to all steroids at 72 h Almost allmodel parameters had positive sensitivity for downstreamsteroids in the adrenal steroidogenic pathway and negativesensitivity for direct-binding steroids as substrates of thesteroidogenic enzyme The sensitivity of 119881max in all steroido-genic enzymes was relatively higher than 119870
119898for the same
steroid substrate
342 Simulation of the Metabolic Balance of AdrenalSteroidogenesis Pathway To clearly show the property of themetabolic shift between mineralocorticoid and glucocorti-coid biosynthesis we performed two-dimensional parameterscanning of the enzymatic activities of CYP17H and HSD3B2(Figure 7) NCI-H295R cells lost the ability to produce allsteroid hormones when enzymatic activities of CYP17H andHSD3B2 were changed by over 60 and 30 respectivelyActivation ofCYP17HandorHSD3B2 induced themetabolicshift that enhanced the glucocorticoid biosynthesis and devi-ated from the mineralocorticoid biosynthesis On the otherhand inhibition of CYP17H andor HSD3B2 induced themetabolic shift that enhanced the mineralocorticoid biosyn-thesis and deviated from the glucocorticoid biosynthesisMoreover the enzymatic activity of HSD3B2 regulated themetabolic balance of sex steroids and the precursors onadrenal steroidogenesis of NCI-H295R cells E1 TESTO andDIONE were produced by NCI-H295R cells when activatingthe enzymatic activity of HSD3B2 Conversely E2 andDHEAwere produced by NCI-H295R cells when suppressing theenzymatic activity of HSD3B2 The biosynthesis of down-stream steroids in adrenal steroidogenesis pathway such asmineralocorticoids and glucocorticoids was almost com-pletely terminated when the enzymatic activity of HSD3B2was decreased by over 80
4 Discussion
41 Importance of 3120573-HSDActivity in Adrenal SteroidogenesisOur systematic analysis using the mathematical model ofadrenal steroidogenesis in NCI-H295R cells revealed thatthe enzymatic activity of 3120573-HSD controls the dynamics ofadrenal steroidogenesis The activity of the StAR proteincontrols the net capacity of steroidogenesis in steroidogeniccells which is the transport of cholesterol from the outer toinner mitochondrial membranes Both the expression levelsof StAR protein and mRNA are rapidly elevated in responseto stimulation by tropic hormones such as ACTH [62 63]Another important factor in adrenal steroidogenesisis is thecholesterol side chain cleavage enzyme CYP11A1 the firstrate-limiting and hormonally regulated step in the synthesisof all steroid hormones which is conversion of cholesterolto pregnenolone in mitochondria [64] According to ourresults of global sensitivity analysis (Supplementary Table 1and Figure 6(d)) in addition to CYP11A1 and StAR proteins3120573-HSD was one of the key regulators in adrenal steroido-genesis of NCI-H295R cells And also this result suggeststhat a significant regulatory mechanism in steroidogenesis
14 Journal of Toxicology
60544830 4236 66 7218 246 120
Time (hours)
0
1
2
3C
once
ntra
tion
(a)
60544830 4236 66 7218 246 120
Time (hours)
minus05
0
05
minus1
1
Sens
itivi
ty
(b)
0
6
12
18
24
30
36
42
48
54
60
66
72
Sens
itivi
ty fo
r med
ium
conc
entr
atio
ns
of ch
oles
tero
l or s
tero
id h
orm
ones
Dur
atio
n tim
e (ho
urs)
Model parameters
+10
+05
00
minus05
minus10
(c)
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
Parameter number in the mathematical model of steroidogenesis in NCI-H295R cells
CHOL
PREG
HPREG
DHEA
PROG
HPROG
DIONE
DCORTICO
DCORT
CORTICO
CORT
ALDO
TESTO
E1
E2
Chol
este
rol o
r ste
roid
hor
mon
es
0
05
10
15
20
minus05
minus10
minus15
minus20
(d)
Figure 6 Heat-map of the global dynamic sensitivity analysis of adrenal steroid concentrations produced by NCI-H295R cells The globaldynamic sensitivity analysis is a powerful tool to comprehensively understand the dependencies of themodel parameters in themathematicalmodel of the biological complex system Dynamic sensitivities of model parameters in the mathematical model of adrenal steroidogenesisin NCI-H295R cells were calculated for all steroid concentrations in the culture medium every 6 h until 72 h after stimulation To clarify theview of heat-map of global dynamic sensitivity analysis imaginary data of the dynamics of steroid concentrations in the original model (blueline) and perturbed model for sensitivity analysis (red line) were prepared (a) Using this imaginary data the calculated dynamic sensitivities(b) and the visualized dynamic sensitivity as one block of the heat-map (c) were shown respectively By the same method that explainedusing imaginary data the large-scale data of the global dynamic sensitivity analysis on the mathematical model of adrenal steroidogenesisin NCI-H295R cells was comprehensively visualized as a big graph of heat-map (d) Parameter numbers in the horizontal axis are (1)119896Choresterol Transport (2) 119896CEH (3) 119896MTR
119891 (4) 119896MTR
119887 (5) 119896acc
119891 (6) 119896acc
119887 (7) 119896loc
119891 (8) 119896loc
119887 (9) 119896Oxysterol Synthesis (10)119870
CYP11A1119898
(11) 119881CYP11A1max (12)119870CYP17H119898A
(13) 119870CYP17H119898B (14) 119881CYP17HmaxA (15) 119881CYP17HmaxB (16) 119881CYP17HmaxB (17) 119870CYP17L
119898B (18) 119881CYP17LmaxA (19) 119881CYP17LmaxB (20) 119870HSD3B2119898A (21) 119870HSD3B2
119898B (22) 119870HSD3B2119898C (23)
119881HSD3B2maxA (24) 119881HSD3B2
maxB (25) 119881HSD3B2maxB (26) 119870CYP21A2
119898A (27) 119870CYP21A2119898B (28) 119881CYP21A2maxA (29) 119881CYP21A2maxB (30) 119870CYP11B1
119898A (31) 119870CYP11B1119898B (32) 119881CYP11B1maxA
(33) 119881CYP11B1maxB (34) 119896CYP11B2 (35) 119870HSD17B3119898A (36) 119870HSD17B3
119898B (37) 119881HSD17B3maxA (38) 119881HSD17B3
maxB (39) 119870CYP19A1119898A (40) 119870CYP19A1
119898B (41) 119881CYP19A1maxA and (42)119881
CYP19A1maxB PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-
hydroxyprogesteroneDIONE androstenedioneDCORTICO 11-deoxycorticosteroneDCORT 11-deoxycortisol CORTICO corticosteroneCORT cortisol ALDO aldosterone and TESTO testosterone
pathway is very reasonable StAR CYP11A1 and 3120573-HSD(isoforms 1 or 2 in humans) proteins generally respond to thesame hormones that stimulate steroid production throughcommon pathways such as cAMP signaling in adrenal glandsand testes [65 66] Moreover our data also support recentexperimental evidence from clinical and in vivo studiessuggesting that the enzymatic activity of 3120573-HSD plays
an important role in the regulation of mineralocorticoidsynthesis in adrenal steroidogenesis and contributes to hyper-tension caused by abnormal overproduction of aldosterone[67ndash70] Circadian clock-deficient Cry-null mice show salt-sensitive hypertension due to abnormally high synthesis ofaldosterone which is caused by constitutively high expressionof HSD3B6mRNA and protein in the adrenal cortex [67 68]
Journal of Toxicology 15
All steroid producing cells
Mineralocorticoid producing cells Glucocorticoid producing cells
Mineralocorticoid and glucocorticoid producing cells
Terminated steroidogenesis
Sex steroids and precursor sex steroidsDIONE TESTO E1DEHA E2
NCI-H295Rcells at controlcondition
Normalized activities of HSD3B2
Nor
mal
ized
activ
ities
of C
YP17
H
2
18
16
14
12
1
08
06
04
02
00 05 1 15 2
Figure 7 Metabolic categories of the steroidogenic cells determined by the balance of HSD3B2 and CYP17H activitiesThe two-dimensionalparameter scanning analysis by the perturbation of two focused parameters clarifies the interaction and the relationship between the twoparameters in the complex system Functional cellular categories of steroidogenic cells were defined by the levels of mineralocorticoid(ALDO) glucocorticoids (DCORTandCORT) and androgens (DHEAandDIONE) at 72 h after stimulation Enzymatic activities ofHSD3B2andCYP17were normalized by standard values of the simulationmodel inNCI-H295R cells Scanning ranges ofHSD3B2 andCYP17 activitieswere 0ndash200 each 10 Green regions are all 14 steroids produced by NCI-H295R cells Red blue and purple regions are mineralocorticoidproducing cells glucocorticoids-producing cells and both corticoid-producing cells respectively Yellow regions are steroidogenesis of NCI-H295R cells terminated upstream of the adrenal steroidogenesis pathway
Recent clinical observations have revealed predominantexpression of HSD3B2 mRNA and protein in tumor cellsof aldosterone-producing adenoma (APA) and HSD3B1mRNA significantly correlated with CYP11B2 mRNA levelsand plasma aldosterone concentrations in APA patients [6970] However the relationship is unclear and disputed ina small-scale clinical study indicating that genetic variationin HSD3B1 affects blood pressure and hypertension in APApatients [71] The results of our simulation study suggest that3120573-HSD protein (human genes are HSD3B1 and HSD3B2) isone of the determination factors for the dynamic propertyof adrenal steroidogenesis Our results support the clinicalevidence of Doi and colleagues [69] and we believe that theHSD3B1 enzyme has a promising potential as novel drugtarget for endocrine hypertension
42 Metabolic Shift of Adrenal Steroidogenesis and the Contri-butions of HSD3B2 and CYP17H The metabolic propertiesof adrenal steroidogenesis in NCI-H295R cells were revealedby dynamic sensitivity analysis using themathematicalmodel(Figure 6) Mineralocorticoids such as DCORTICO COR-TICO and ALDO and intermediate steroids upstream of
the adrenal steroidogenesis pathway such as PREG HPREGDHEA PROG and HPROG were accelerated by reactionsof cholesterol import (119896Cholesterol Transport) StAR protein (119896MTR
119891
and 119896loc119891) and CYP11A1 (119881CYP11A1max ) On the other hand glu-
cocorticoids such as DCORT and CORT and sex hormonessuch as DIONE TESTO and E1 were suppressed by thesemodel parametersTherefore enhancement of the net adrenalsteroidogenesis capacity which supplies PREG precursor tothe pathway causes a production shift from glucocorticoidsto mineralocorticoids by substrate inhibitions of CYP17HHSD3B2 and CYP21A2 caused by accumulation of initialintermediate steroids such as PREG and PROG Sensitivi-ties of CYP17H (119881CYP17HmaxA ) and HSD3B2 (119881HSD3B2
maxA ) for theproducts were dynamically changed and these parametersdetermined the metabolic balance of downstream steroidsin the adrenal steroidogenesis pathway According to theseresults of dynamic sensitivity analysis of StAR CYP11A1CYP17H and HSD3B2 we suggest that the enhancementof CYP17H and HSD3B2 activity during ACTH stimulationwas important to shift the steroidogenic output away fromALDO biosynthesis towards CORT biosynthesis as wellas adrenal androgen production This suggestion partially
16 Journal of Toxicology
supports a comparative animal study in which molecularand cellular variations in CYP17H activity dramatically affectacute cortisol production resulting in distinct physiologicaland behavioral responses [72]
Results of two-dimensional parameter scanning of theenzymatic activities of CYP17H and HSD3B2 quantitativelyshowed the detail of the metabolic relationship betweenmin-eralocorticoid and glucocorticoid biosynthesis (Figure 7)Particularly the results showed that the balance of theseenzymatic activities was very important for the typicalfunction of NCI-H295R cells namely the ability to produceall steroid hormones NCI-H295R cells lost this functionwhen enzymatic activities of CYP17H and HSD3B2 werechanged by over 60 and 30 respectively In addition theybecame mineralocorticoid (ALDO) secreting cells when theenzymatic activity of CYP17H or HSD3B2 was inhibited byover 50 or glucocorticoid (DCORT and CORT) secretingcells when these enzymes were activated by over 50 Inparticular this analysis also showed that HSD3B2 was a keyplayer in the adrenal steroidogenesis of NCI-H295R cellsbecause HSD3B2 inhibition by over 80 almost completelyinhibited the biosynthesis of downstream steroids The ratioof CYP17A1 to HSD3B2 mRNA expression levels has beenrelated to several endocrine diseases with a low level in APAs[73] and high level in cortisol-producing adenomas [74]Furthermore the expression levels or enzymatic activitiesof CYP17A1 and HSD3B1 have been related to androgenproduction in polycystic ovary syndrome [75 76] Theseclinical studies support our simulation results indicating thatthe balance of enzymatic activity of CYP17H and HSD3B2determines the shift in steroidogenic output to mineralocor-ticoids glucocorticoids or androgens
43 Methodologies of Quantitative Mechanistic Analysis forDrug Discovery According to our results obtained using themathematical model of steroidogenesis in NCI-H295R cellssuch as sensitivity analysis comprehensive analysis based onsystems biology is available to quantitatively estimate themechanism of action of steroidogenic disrupting compoundsfrom differential profiling of adrenal steroid hormonesbecause dynamic patterns of steroid hormones in adrenalsteroidogenesis pathway are highly complex Our proposedmethod of quantitative mechanistic analysis of steroidogenicinhibitors was able to predict known action sites in theadrenal steroidogenesis pathway at only one time point (72 hafter drug exposure) Moreover according to the resultsof sensitivity analysis (Figure 6) 119881max of all steroidogenicenzymes was more sensitive than the 119870
119898 because the
intracellular concentrations of steroid hormones were almostmaintained at sufficiently high levels compared with 119870
119898
values of steroidogenic enzymesThese results suggested thatestimation of the mechanism of action of drugs is moreeffective and detectable when using the influences of 119881maxas the searching parameters such as our proposed methodOur data showed that the proposed method based on asystems biology model is a very powerful tool for exploratoryscreening of steroidogenic disrupting compounds
RCGA as a solver of parameter estimation problemsin systems biology has been applied to biological network
identification of gene regulatory networks and metabolicpathways and optimization of biological processes usingexperimentally observed time-course data [77ndash82] In thisstudy RCGA was useful to estimate the mechanism ofaction of novel pharmaceutical drug candidates for adrenalsteroidogenesis as a new application of RCGA in systemsbiology We had two issues when applying RCGA to thequantitative mechanistic analysis of drug actions Theseissues were the vast calculation cost and multimodality ofquasi-optimum solutions in solving the optimization prob-lem because the mathematical model in systems biologyconsists of many equations and parameters A proposedoptimization strategy using RCGA based on REXJGG wasa highly stable and efficient calculation method for a betterquasi-optimum solution than the unimodal normal distri-bution crossover (UNDX)minimum generation gap (MGG)method that is well applied in the engineering field In addi-tion we expanded the RCGA optimization program basedon REXJGG to a hybrid method of GA and then applieda local search as recommended by Harada and Kobayashi[28 83] A final optimal solution was obtained with a goodconvergence property Because these problems are generalin systems biology studies we suggest that the proposedhybrid method based on REXJGG is a very useful tool forquantitative mechanistic analysis of novel pharmaceuticaldrugs not limited to steroidogenic disrupting compounds
5 Conclusions
Thenovelmathematicalmodel of adrenal steroidogenesis wasconstructed in this study including cholesterol transport anddistribution the C
21-steroid hormone pathway steroid trans-
port and cell proliferation which could reproduce adrenalsteroidogenesis in NCI-H295R cells According to the resultsof dynamic sensitivity analysis using the newmodel HSD3B2plays the most important role in the metabolic balance ofadrenal steroidogenesis in NCI-H295R cells Moreover toquantitatively estimate mechanisms of action of adrenal toxiccompounds we analyzed differential metabolic profiles of12 steroid hormones at 3 days after exposure to 11 adrenaltoxic compounds by using the new mathematical modeland a hybrid optimization method of the RCGA and a localsearch (nonlinear least squares) We could estimate whichsteroidogenic enzymes were affected by these compoundsusing the hybrid optimization method Vasculotoxic agentswere estimated to have no effect according to the resultsobtained by our method In terms of adrenal steroidogenicinhibitors the predicted action sites were approximatelymatched to the target enzymes as reported in the literatureThus our computer-aided method based on a systems biol-ogy approach may be useful to analyze the mechanism ofaction of endocrine-disrupting compounds and to assess thehuman adrenal toxicity of novel pharmaceutical drugs basedon steroid hormone profiling
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Journal of Toxicology 17
Authorsrsquo Contribution
Ryuta Saito Natsuko Terasaki and Naohisa Tsutsui designedthe study Naohisa Tsutsui and Naoya Masutomi supervisedthe experiments Natsuko Terasaki performed all in vitroexperiments and collected dataMakotoYamazaki performedall measurements of steroids Ryuta Saito conceived the insilico strategies analyzed data developed all programs andthe mathematical model and performed all simulation anal-yses with Masahiro Okamoto Ryuta Saito Natsuko TerasakiMakoto Yamazaki NaoyaMasutomi andMasahiroOkamotowrote the paper
Acknowledgments
The authors would like to thank Dr Michael S Breenand Dr Miyuki Breen for helpful discussions concerningprevious mathematical models of adrenal steroidogenesisThis work was partially supported by the Grant-in-Aid forScientific Research on Innovative Areas ldquoSynthetic Biologyrdquo(No 23119001 (MO)) from Ministry of Education CultureSports Science and Technology in Japan
References
[1] P W Harvey D J Everett and C J Springall ldquoAdrenal toxi-cology a strategy for assessment of functional toxicity to theadrenal cortex and steroidogenesisrdquo Journal of Applied Toxi-cology vol 27 no 2 pp 103ndash115 2007
[2] W E Ribelin ldquoThe effects of drugs and chemicals upon thestructure of the adrenal glandrdquo Fundamental and Applied Toxi-cology vol 4 no 1 pp 105ndash119 1984
[3] H D Colby and P A Longhurst ldquoToxicology of the adrenalglandrdquo in Endocrine Toxicology C K Atterwill and J D FlackEds pp 243ndash281 CambridgeUniversity Press Cambridge UK1992
[4] H D Colby ldquoIn vitro assessment of adrenocortical toxicityrdquoJournal of Pharmacological and Toxicological Methods vol 32no 1 pp 1ndash6 1994
[5] T J Rosol J T Yarrington J Latendresse and C C CapenldquoAdrenal gland structure function and mechanisms of toxic-ityrdquo Toxicologic Pathology vol 29 no 1 pp 41ndash48 2001
[6] P W Harvey and D J Everett ldquoThe adrenal cortex and ster-oidogenesis as cellular and molecular targets for toxicity crit-ical omissions from regulatory endocrine disrupter screeningstrategies for human healthrdquo Journal of Applied Toxicology vol23 no 2 pp 81ndash87 2003
[7] A F Gazdar H K Oie C H Shackleton et al ldquoEstablishmentand characterization of a human adrenocortical carcinoma cellline that expresses multiple pathways of steroid biosynthesisrdquoCancer Research vol 50 no 17 pp 5488ndash5496 1990
[8] W E Rainey I M Bird C Sawetawan et al ldquoRegulation ofhuman adrenal carcinoma cell (NCI-H295) production of C19steroidsrdquo The Journal of Clinical Endocrinology amp Metabolismvol 77 no 3 pp 731ndash737 1993
[9] B Staels D W Hum andW L Miller ldquoRegulation of steroido-genesis in NCI-H295 cells a cellular model of the human fetaladrenalrdquo Molecular Endocrinology vol 7 no 3 pp 423ndash4331993
[10] M Hecker J L Newsted M B Murphy et al ldquoHuman adreno-carcinoma (H295R) cells for rapid in vitro determination of
effects on steroidogenesis hormone productionrdquo Toxicologyand Applied Pharmacology vol 217 no 1 pp 114ndash124 2006
[11] A Oskarsson E Ulleras K E Plant J P Hinson and P SGoldfarb ldquoSteroidogenic gene expression in H295R cells andthe human adrenal gland adrenotoxic effects of lindane invitrordquo Journal of Applied Toxicology vol 26 no 6 pp 484ndash4922006
[12] UMuller-Vieira M Angotti and RW Hartman ldquoThe adreno-cortical tumor cell line NCI-H295R as an in vitro screeningsystem for the evaluation of CYP11B2 (aldosterone synthase)and CYP11B1 (steroid-11120573-hydroxylase) inhibitorsrdquoThe Journalof Steroid Biochemistry and Molecular Biology vol 96 no 3-4pp 259ndash270 2005
[13] K Imagawa S Okayama M Takaoka et al ldquoInhibitory effectof efonidipine on aldosterone synthesis and secretion in humanadrenocarcinoma (H295R) cellsrdquo Journal of CardiovascularPharmacology vol 47 no 1 pp 133ndash138 2006
[14] M Voets UMuller-Vieira S Marchais-Oberwinkler and RWHartmann ldquoSynthesis of amidinohydrazones and evaluation oftheir inhibitory effect towards aldosterone synthase (CYP11B2)and the formation of selected steroidsrdquo Archiv der Pharmazievol 337 no 7 pp 411ndash416 2004
[15] F K Nielsen C H Hansen J A Fey et al ldquoH295R cells as amodel for steroidogenic disruption a broader perspective usingsimultaneous chemical analysis of 7 key steroid hormonesrdquoToxicology in Vitro vol 26 no 2 pp 343ndash350 2012
[16] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cortisolproduction and on activities of steroidogenic enzymes inhuman adrenocortical H295R cellsrdquo Journal of Steroid Biochem-istry and Molecular Biology vol 80 no 3 pp 355ndash363 2002
[17] R F Canton J T Sanderson S Nijmeijer A Bergman R JLetcher and M van den Berg ldquoIn vitro effects of brominatedflame retardants and metabolites on CYP17 catalytic activity anovel mechanism of actionrdquo Toxicology and Applied Pharma-cology vol 216 no 2 pp 274ndash281 2006
[18] K Hilscherova P D Jones T Gracia et al ldquoAssessment ofthe effects of chemicals on the expression of ten steroidogenicgenes in the H295R cell line using real-time PCRrdquo ToxicologicalSciences vol 81 no 1 pp 78ndash89 2004
[19] T Gracia K Hilscherova P D Jones et al ldquoThe H295R systemfor evaluation of endocrine-disrupting effectsrdquo Ecotoxicologyand Environmental Safety vol 65 no 3 pp 293ndash305 2006
[20] X Zhang J L NewstedM Hecker E B Higley P D Jones andJ P Giesy ldquoClassification of chemicals based on concentration-dependent toxicological data using ToxClustrdquo EnvironmentalScience and Technology vol 43 no 10 pp 3926ndash3932 2009
[21] M S Breen M Breen N Terasaki M Yamazaki and R BConolly ldquoComputational model of steroidogenesis in humanH295R cells to predict biochemical response to endocrine-active chemicalsmodel development formetyraponerdquoEnviron-mental Health Perspectives vol 118 no 2 pp 265ndash272 2010
[22] M Breen M S Breen N Terasaki M Yamazaki A LLloyd and R B Conolly ldquoMechanistic computational model ofsteroidogenesis in H295R cells role of oxysterols and cell pro-liferation to improve predictability of biochemical response toendocrine active chemical-metyraponerdquo Toxicological Sciencesvol 123 no 1 pp 80ndash93 2011
[23] F B Kraemer ldquoAdrenal cholesterol utilizationrdquo Molecular andCellular Endocrinology vol 265-266 pp 42ndash45 2007
[24] W L Miller ldquoSteroidogenic acute regulatory protein (StAR)a novel mitochondrial cholesterol transporterrdquo Biochimica et
18 Journal of Toxicology
Biophysica Acta (BBA)mdashMolecular and Cell Biology of Lipidsvol 1771 no 6 pp 663ndash676 2007
[25] S Szabo and I T Lippe ldquoAdrenal gland chemically inducedstructural and functional changes in the cortexrdquo ToxicologicPathology vol 17 no 2 pp 317ndash329 1989
[26] C Chen-Pan I-J Pan Y Yamamoto T Sakogawa J Yamadaand YHayashi ldquoPrompt recovery of damaged adrenalmedullaeinduced by salinomycinrdquoToxicologic Pathology vol 27 no 5 pp563ndash572 1999
[27] R J Santen and R I Misbin ldquoAminoglutethimide review ofpharmacology and clinical userdquo Pharmacotherapy vol 1 no 2pp 95ndash120 1981
[28] M J Carella N V Dimitrov V V Gossain L Srivastava andD R Rovner ldquoAdrenal effects of low-dose aminoglutethimidewhen used alone in postmenopausal women with advancedbreast cancerrdquoMetabolism vol 43 no 6 pp 723ndash727 1994
[29] M K Johansson J T Sanderson and B-O Lund ldquoEffects of3-MeSO2-DDE and some CYP inhibitors on glucocorticoidsteroidogenesis in the H295R human adrenocortical carcinomacell linerdquo Toxicology in Vitro vol 16 no 2 pp 113ndash121 2002
[30] Y Touitou A Bogdan J C Legrand and P Desgrez ldquoAmino-glutethimide and glutethimide effects on 18-hydroxycorti-costerone biosynthesis by human and sheep adrenals in vitrordquoActa Endocrinologica vol 80 no 3 pp 517ndash526 1975
[31] M M Hart and J A Straw ldquoStudies on the site of action ofop1015840-DDD in the dog adrenal cortex 1 Inhibition of ACTH-mediated pregnenolone synthesisrdquo Steroids vol 17 no 1ndash5 pp559ndash574 1971
[32] M Ojima M Saitoh N Itoh Y Kusano S Fukuchi and HNaganuma ldquoThe effects of op1015840-DDD on adrenal steroidogen-esis and hepatic steroid metabolismrdquo Nippon Naibunpi Gakkaizasshi vol 61 no 3 pp 168ndash178 1985
[33] L F Canosa and N R Ceballos ldquoEffects of different steroid-biosynthesis inhibitors on the testicular steroidogenesis of thetoad Bufo arenarumrdquo Journal of Comparative Physiology B vol171 no 6 pp 519ndash526 2001
[34] D C Kossor S Kominami S Takemori and H D ColbyldquoRole of the steroid 17 alpha-hydroxylase in spironolactone-mediated destruction of adrenal cytochrome P-450rdquoMolecularPharmacology vol 40 no 2 pp 321ndash325 1991
[35] S C Cheng K Suzuki W Sadee and B W Harding ldquoEffectsof spironolactone canrenone and canrenoate K on cytochromeP450 and 11120573 and 18 hydroxylation in bovine and humanadrenal corticalmitochondriardquo Endocrinology vol 99 no 4 pp1097ndash1106 1976
[36] K Yanagibashi M Haniu J E Shively W H Shen and P HallldquoThe synthesis of aldosterone by the adrenal cortex Two zones(fasciculata and glomerulosa) possess one enzyme for 11120573- 18-hydroxylation and aldehyde synthesisrdquoThe Journal of BiologicalChemistry vol 261 no 8 pp 3556ndash3562 1986
[37] K Morishita H Okumura N Ito and N Takahashi ldquoPrimaryculture system of adrenocortical cells from dogs to evaluatedirect effects of chemicals on steroidogenesisrdquo Toxicology vol165 no 2-3 pp 171ndash178 2001
[38] S W J Lamberts E G Bons H A Bruining and F H de JongldquoDifferential effects of the imidazole derivatives etomidateketoconazole and miconazole and of metyrapone on the secre-tion of cortisol and its precursors by human adrenocorticalcellsrdquo Journal of Pharmacology and Experimental Therapeuticsvol 240 no 1 pp 259ndash264 1987
[39] L Yin S Lucas F Maurer U Kazmaier Q Hu and RW Hartmann ldquoNovel imidazol-1-ylmethyl substituted 1256-tetrahydropyrrolo[321-ij]quinolin-4-ones as potent and selec-tive CYP11B1 inhibitors for the treatment of Cushingrsquos syn-dromerdquo Journal ofMedicinal Chemistry vol 55 no 14 pp 6629ndash6633 2012
[40] M DiMattina N Maronian H Ashby D L Loriaux andB D Albertson ldquoKetoconazole inhibits multiple steroidogenicenzymes involved in androgen biosynthesis in the humanovaryrdquo Fertility and Sterility vol 49 no 1 pp 62ndash65 1988
[41] M Ayub and M J Levell ldquoInhibition of human adrenal steroi-dogenic enzymes in vitro by imidazole drugs including keto-conazolerdquo Journal of Steroid Biochemistry vol 32 no 4 pp 515ndash524 1989
[42] D Engelhardt M M Weber T Miksch F Abedinpour andC Jaspers ldquoThe influence of ketoconazole on human adrenalsteroidogenesis incubation studies with tissue slicesrdquo ClinicalEndocrinology vol 35 no 2 pp 163ndash168 1991
[43] S Bhasin S Sikka T Fielder et al ldquoHormonal effects of keto-conazole in vivo in the male rat mechanism of actionrdquo Endo-crinology vol 118 no 3 pp 1229ndash1232 1986
[44] D M Zisterer and D C Williams ldquoCalmidazolium and otherimidazole compounds affect steroidogenesis in Y1 cells lack ofinvolvement of the peripheral-type benzodiazepine receptorrdquoJournal of Steroid Biochemistry and Molecular Biology vol 60no 3-4 pp 189ndash195 1997
[45] L P Walsh C N Kuratko and D M Stocco ldquoEconazoleand miconazole inhibit steroidogenesis and disrupt steroido-genic acute regulatory (StAR) protein expression post-trans-criptionallyrdquo Journal of Steroid Biochemistry and Molecular Bio-logy vol 75 no 4-5 pp 229ndash236 2000
[46] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cor-tisol production and on activities of steroidogenic enzymesin human adrenocortical H295R cellsrdquo The Journal of SteroidBiochemistry and Molecular Biology vol 80 no 3 pp 355ndash3632002
[47] D P Dempsher D S Gann and R D Phair ldquoA mechanisticmodel of ACTH-stimulated cortisol secretionrdquo American Jour-nal of Physiology vol 246 no 4 part 2 pp R587ndashR596 1984
[48] N Sarai S Matsuoka and A Noma ldquosimBio a Java package forthe development of detailed cell modelsrdquo Progress in Biophysicsand Molecular Biology vol 90 no 1ndash3 pp 360ndash377 2006
[49] K Levenberg ldquoA method for the solution of certain non-linearproblems in least squaresrdquo Quarterly of Applied Mathematicsvol 2 pp 164ndash168 1944
[50] D W Marquardt ldquoAn algorithm for least-squares estimation ofnonlinear parametersrdquo SIAM Journal on Applied Mathematicsvol 11 no 2 pp 431ndash441 1963
[51] P E Gill and W Murray ldquoAlgorithms for the solution of thenonlinear least-squares problemrdquo SIAM Journal on NumericalAnalysis vol 15 no 5 pp 977ndash992 1978
[52] Y Akimoto R Hasada J Sakuma I Ono and S KobayashildquoGeneration alternation model for real-coded GA using multi-parent proposal and evaluation of just generation gap (JGG)rdquoin Proceedings of the 19th SICE Symposium on DecentralizedAutonomous Systems pp 341ndash346 Tokyo Japan January 2007
[53] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoProposal and evaluation of adaptive real-coded crossoverAREXrdquo Transactions of the Japanese Society for Artificial Intel-ligence vol 24 no 6 pp 446ndash458 2009
Journal of Toxicology 19
[54] S Kobayashi ldquoThe frontiers of real-coded genetic algorithmsrdquoJournal of Japanese Society for Artificial Intelligence vol 24 no1 pp 147ndash162 2009
[55] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoAnalysis of the behavior of MGG and JGG as a selection modelfor real-coded genetic algorithmsrdquo Transactions of the JapaneseSociety for Artificial Intelligence vol 25 no 2 pp 281ndash289 2010
[56] A Struthers H Krum and G H Williams ldquoA comparisonof the aldosterone-blocking agents eplerenone and spironolac-tonerdquo Clinical Cardiology vol 31 no 4 pp 153ndash158 2008
[57] K Yamasaki M Sawaki S Noda et al ldquoComparison of theHershberger assay and androgen receptor binding assay oftwelve chemicalsrdquo Toxicology vol 195 no 2-3 pp 177ndash1862004
[58] C Eil and S K Edelson ldquoThe use of human skin fibroblasts toobtain potency estimates of drug binding to androgen recept-orsrdquo The Journal of Clinical Endocrinology amp Metabolism vol59 no 1 pp 51ndash55 1984
[59] P Ye T Yamashita D M Pollock H Sasano and W ERainey ldquoContrasting effects of eplerenone and spironolac-tone on adrenal cell steroidogenesisrdquo Hormone and MetabolicResearch vol 41 no 1 pp 35ndash39 2009
[60] B D Albertson K L Frederick N C Maronian et al ldquoTheeffect of ketoconazole on steroidogenesis I Leydig cell enzymeactivity in vitrordquo Research Communications in Chemical Patho-logy and Pharmacology vol 61 no 1 pp 17ndash26 1988
[61] B D Albertson N C Maronian K L Frederick et al ldquoTheeffect of ketoconazole on steroidogenesis II Adrenocorticalenzyme activity in vitrordquoResearch Communications in ChemicalPathology and Pharmacology vol 61 no 1 pp 27ndash34 1988
[62] DM Stocco andB J Clark ldquoRegulation of the acute productionof steroids in steroidogenic cellsrdquo Endocrine Reviews vol 17 no3 pp 221ndash244 1996
[63] L K Christenson and J F Strauss III ldquoSteroidogenic acute regu-latory protein (StAR) and the intramitochondrial translocationof cholesterolrdquo Biochimica et Biophysica Acta (BBA)mdashMolecularand Cell Biology of Lipids vol 1529 no 1ndash3 pp 175ndash187 2000
[64] E R Simpson ldquoCholesterol side-chain cleavage cytochromeP450 and the control of steroidogenesisrdquo Molecular and Cel-lular Endocrinology vol 13 no 3 pp 213ndash227 1979
[65] H A LaVoie and S R King ldquoTranscriptional regulation of ster-oidogenic genes STARD1 CYP11A1 and HSD3Brdquo ExperimentalBiology and Medicine vol 234 no 8 pp 880ndash907 2009
[66] F Raucci A DrsquoAniello and M M Di Fiore ldquoStimulation ofandrogen production by d-aspartate through the enhancementof StAR P450scc and 3120573-HSD mRNA levels in vivo rat testisand in culture of immature rat Leydig cellsrdquo Steroids vol 84pp 103ndash110 2014
[67] M Doi Y Takahashi R Komatsu et al ldquoSalt-sensitive hyper-tension in circadian clock-deficient Cry-null mice involves dys-regulated adrenal Hsd3b6rdquo Nature Medicine vol 16 no 1 pp67ndash74 2010
[68] H Okamura M Doi Y Yamaguchi and J-M Fustin ldquoHyper-tension due to loss of clock novel insight from the molecu-lar analysis of Cry1Cry2-deleted micerdquo Current HypertensionReports vol 13 no 2 pp 103ndash108 2011
[69] M Doi F Satoh T Maekawa et al ldquoIsoform-specific mono-clonal antibodies against 3120573-hydroxysteroid dehydrogenaseisomerase family provide markers for subclassification ofhuman primary aldosteronismrdquo The Journal of Clinical Endo-crinology amp Metabolism vol 99 no 2 pp E257ndashE262 2014
[70] S Konosu-Fukaya Y Nakamura F Satoh et al ldquo3120573-hydroxy-steroid dehydrogenase isoforms in human aldosterone-produc-ing adenomardquo Molecular and Cellular Endocrinology vol 408pp 205ndash212 2015
[71] G C Verwoert J Hofland N Amin et al ldquoExpression and genevariation studies deny association of humanHSD3B1 gene withaldosterone production or blood pressurerdquoAmerican Journal ofHypertension vol 28 no 1 pp 113ndash120 2015
[72] D Hough K Storbeck SW P Cloete A C Swart and P SwartldquoRelative contribution of P450c17 towards the acute cortisolresponse lessons from sheep and goatsrdquoMolecular and CellularEndocrinology vol 408 pp 107ndash113 2015
[73] I Sakuma S Suematsu Y Matsuzawa et al ldquoCharacterizationof steroidogenic enzyme expression in aldosterone-producingadenoma a comparison with various human adrenal tumorsrdquoEndocrine Journal vol 60 no 3 pp 329ndash336 2013
[74] A Tong A Jia S Yan Y Zhang Y Xie and G Liu ldquoEctopiccortisol-producing adrenocortical adenoma in the renal hilumhistopathological features and steroidogenic enzyme profilerdquoInternational Journal of Clinical and Experimental Pathologyvol 7 no 7 pp 4415ndash4421 2014
[75] A Hirsch D Hahn P Kempna et al ldquoMetformin inhibitshuman androgen production by regulating steroidogenicenzymes HSD3B2 and CYP17A1 and complex I activity of therespiratory chainrdquo Endocrinology vol 153 no 9 pp 4354ndash43662012
[76] B Xu L Gao Y Cui et al ldquoSET protein up-regulated test-osterone production in the cultured preantral folliclesrdquo Repro-ductive Biology and Endocrinology vol 11 article 9 2013
[77] S Kikuchi D Tominaga M Arita M Takahashi and MTomita ldquoDynamic modeling of genetic networks using geneticalgorithm and S-systemrdquo Bioinformatics vol 19 no 5 pp 643ndash650 2003
[78] N Shikata YMaki M Nakatsui et al ldquoDetermining importantregulatory relations of amino acids from dynamic networkanalysis of plasma amino acidsrdquo Amino Acids vol 38 no 1 pp179ndash187 2010
[79] M Nakatsui T Ueda Y Maki I Ono and M OkamotoldquoMethod for inferring and extracting reliable genetic interac-tions from time-series profile of gene expressionrdquoMathematicalBiosciences vol 215 no 1 pp 105ndash114 2008
[80] A Komori Y Maki M Nakatsui I Ono and M OkamotoldquoEfficient numerical optimization algorithm based on new real-coded genetic algorithm AREX + JGG and application to theinverse problem in systems biologyrdquo Applied Mathematics vol3 no 10 pp 1463ndash1470 2012
[81] J A Roubos G van Straten and A J B van Boxtel ldquoAn evolu-tionary strategy for fed-batch bioreactor optimization conceptsand performancerdquo Journal of Biotechnology vol 67 no 2-3 pp173ndash187 1999
[82] D Sarkar and J M Modak ldquoGenetic algorithms with filters foroptimal control problems in fed-batch bioreactorsrdquo Bioprocessand Biosystems Engineering vol 26 no 5 pp 295ndash306 2004
[83] K Harada K Ikeda J Sakuma I Ono and S KobayashildquoHybridization of genetic algorithm with local search in mul-tiobjective function optimization recommendation of GA thenLSrdquoTransactions of the Japanese Society for Artificial Intelligencevol 21 no 6 pp 482ndash492 2006
Submit your manuscripts athttpwwwhindawicom
PainResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
ToxinsJournal of
VaccinesJournal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AntibioticsInternational Journal of
ToxicologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
StrokeResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Drug DeliveryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Pharmacological Sciences
Tropical MedicineJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AddictionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Emergency Medicine InternationalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Autoimmune Diseases
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anesthesiology Research and Practice
ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Pharmaceutics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
MEDIATORSINFLAMMATION
of
14 Journal of Toxicology
60544830 4236 66 7218 246 120
Time (hours)
0
1
2
3C
once
ntra
tion
(a)
60544830 4236 66 7218 246 120
Time (hours)
minus05
0
05
minus1
1
Sens
itivi
ty
(b)
0
6
12
18
24
30
36
42
48
54
60
66
72
Sens
itivi
ty fo
r med
ium
conc
entr
atio
ns
of ch
oles
tero
l or s
tero
id h
orm
ones
Dur
atio
n tim
e (ho
urs)
Model parameters
+10
+05
00
minus05
minus10
(c)
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
Parameter number in the mathematical model of steroidogenesis in NCI-H295R cells
CHOL
PREG
HPREG
DHEA
PROG
HPROG
DIONE
DCORTICO
DCORT
CORTICO
CORT
ALDO
TESTO
E1
E2
Chol
este
rol o
r ste
roid
hor
mon
es
0
05
10
15
20
minus05
minus10
minus15
minus20
(d)
Figure 6 Heat-map of the global dynamic sensitivity analysis of adrenal steroid concentrations produced by NCI-H295R cells The globaldynamic sensitivity analysis is a powerful tool to comprehensively understand the dependencies of themodel parameters in themathematicalmodel of the biological complex system Dynamic sensitivities of model parameters in the mathematical model of adrenal steroidogenesisin NCI-H295R cells were calculated for all steroid concentrations in the culture medium every 6 h until 72 h after stimulation To clarify theview of heat-map of global dynamic sensitivity analysis imaginary data of the dynamics of steroid concentrations in the original model (blueline) and perturbed model for sensitivity analysis (red line) were prepared (a) Using this imaginary data the calculated dynamic sensitivities(b) and the visualized dynamic sensitivity as one block of the heat-map (c) were shown respectively By the same method that explainedusing imaginary data the large-scale data of the global dynamic sensitivity analysis on the mathematical model of adrenal steroidogenesisin NCI-H295R cells was comprehensively visualized as a big graph of heat-map (d) Parameter numbers in the horizontal axis are (1)119896Choresterol Transport (2) 119896CEH (3) 119896MTR
119891 (4) 119896MTR
119887 (5) 119896acc
119891 (6) 119896acc
119887 (7) 119896loc
119891 (8) 119896loc
119887 (9) 119896Oxysterol Synthesis (10)119870
CYP11A1119898
(11) 119881CYP11A1max (12)119870CYP17H119898A
(13) 119870CYP17H119898B (14) 119881CYP17HmaxA (15) 119881CYP17HmaxB (16) 119881CYP17HmaxB (17) 119870CYP17L
119898B (18) 119881CYP17LmaxA (19) 119881CYP17LmaxB (20) 119870HSD3B2119898A (21) 119870HSD3B2
119898B (22) 119870HSD3B2119898C (23)
119881HSD3B2maxA (24) 119881HSD3B2
maxB (25) 119881HSD3B2maxB (26) 119870CYP21A2
119898A (27) 119870CYP21A2119898B (28) 119881CYP21A2maxA (29) 119881CYP21A2maxB (30) 119870CYP11B1
119898A (31) 119870CYP11B1119898B (32) 119881CYP11B1maxA
(33) 119881CYP11B1maxB (34) 119896CYP11B2 (35) 119870HSD17B3119898A (36) 119870HSD17B3
119898B (37) 119881HSD17B3maxA (38) 119881HSD17B3
maxB (39) 119870CYP19A1119898A (40) 119870CYP19A1
119898B (41) 119881CYP19A1maxA and (42)119881
CYP19A1maxB PREG pregnenolone HPREG 17120572-hydroxypregnenolone DHEA dehydroepiandrosterone PROG progesterone HPROG 17120572-
hydroxyprogesteroneDIONE androstenedioneDCORTICO 11-deoxycorticosteroneDCORT 11-deoxycortisol CORTICO corticosteroneCORT cortisol ALDO aldosterone and TESTO testosterone
pathway is very reasonable StAR CYP11A1 and 3120573-HSD(isoforms 1 or 2 in humans) proteins generally respond to thesame hormones that stimulate steroid production throughcommon pathways such as cAMP signaling in adrenal glandsand testes [65 66] Moreover our data also support recentexperimental evidence from clinical and in vivo studiessuggesting that the enzymatic activity of 3120573-HSD plays
an important role in the regulation of mineralocorticoidsynthesis in adrenal steroidogenesis and contributes to hyper-tension caused by abnormal overproduction of aldosterone[67ndash70] Circadian clock-deficient Cry-null mice show salt-sensitive hypertension due to abnormally high synthesis ofaldosterone which is caused by constitutively high expressionof HSD3B6mRNA and protein in the adrenal cortex [67 68]
Journal of Toxicology 15
All steroid producing cells
Mineralocorticoid producing cells Glucocorticoid producing cells
Mineralocorticoid and glucocorticoid producing cells
Terminated steroidogenesis
Sex steroids and precursor sex steroidsDIONE TESTO E1DEHA E2
NCI-H295Rcells at controlcondition
Normalized activities of HSD3B2
Nor
mal
ized
activ
ities
of C
YP17
H
2
18
16
14
12
1
08
06
04
02
00 05 1 15 2
Figure 7 Metabolic categories of the steroidogenic cells determined by the balance of HSD3B2 and CYP17H activitiesThe two-dimensionalparameter scanning analysis by the perturbation of two focused parameters clarifies the interaction and the relationship between the twoparameters in the complex system Functional cellular categories of steroidogenic cells were defined by the levels of mineralocorticoid(ALDO) glucocorticoids (DCORTandCORT) and androgens (DHEAandDIONE) at 72 h after stimulation Enzymatic activities ofHSD3B2andCYP17were normalized by standard values of the simulationmodel inNCI-H295R cells Scanning ranges ofHSD3B2 andCYP17 activitieswere 0ndash200 each 10 Green regions are all 14 steroids produced by NCI-H295R cells Red blue and purple regions are mineralocorticoidproducing cells glucocorticoids-producing cells and both corticoid-producing cells respectively Yellow regions are steroidogenesis of NCI-H295R cells terminated upstream of the adrenal steroidogenesis pathway
Recent clinical observations have revealed predominantexpression of HSD3B2 mRNA and protein in tumor cellsof aldosterone-producing adenoma (APA) and HSD3B1mRNA significantly correlated with CYP11B2 mRNA levelsand plasma aldosterone concentrations in APA patients [6970] However the relationship is unclear and disputed ina small-scale clinical study indicating that genetic variationin HSD3B1 affects blood pressure and hypertension in APApatients [71] The results of our simulation study suggest that3120573-HSD protein (human genes are HSD3B1 and HSD3B2) isone of the determination factors for the dynamic propertyof adrenal steroidogenesis Our results support the clinicalevidence of Doi and colleagues [69] and we believe that theHSD3B1 enzyme has a promising potential as novel drugtarget for endocrine hypertension
42 Metabolic Shift of Adrenal Steroidogenesis and the Contri-butions of HSD3B2 and CYP17H The metabolic propertiesof adrenal steroidogenesis in NCI-H295R cells were revealedby dynamic sensitivity analysis using themathematicalmodel(Figure 6) Mineralocorticoids such as DCORTICO COR-TICO and ALDO and intermediate steroids upstream of
the adrenal steroidogenesis pathway such as PREG HPREGDHEA PROG and HPROG were accelerated by reactionsof cholesterol import (119896Cholesterol Transport) StAR protein (119896MTR
119891
and 119896loc119891) and CYP11A1 (119881CYP11A1max ) On the other hand glu-
cocorticoids such as DCORT and CORT and sex hormonessuch as DIONE TESTO and E1 were suppressed by thesemodel parametersTherefore enhancement of the net adrenalsteroidogenesis capacity which supplies PREG precursor tothe pathway causes a production shift from glucocorticoidsto mineralocorticoids by substrate inhibitions of CYP17HHSD3B2 and CYP21A2 caused by accumulation of initialintermediate steroids such as PREG and PROG Sensitivi-ties of CYP17H (119881CYP17HmaxA ) and HSD3B2 (119881HSD3B2
maxA ) for theproducts were dynamically changed and these parametersdetermined the metabolic balance of downstream steroidsin the adrenal steroidogenesis pathway According to theseresults of dynamic sensitivity analysis of StAR CYP11A1CYP17H and HSD3B2 we suggest that the enhancementof CYP17H and HSD3B2 activity during ACTH stimulationwas important to shift the steroidogenic output away fromALDO biosynthesis towards CORT biosynthesis as wellas adrenal androgen production This suggestion partially
16 Journal of Toxicology
supports a comparative animal study in which molecularand cellular variations in CYP17H activity dramatically affectacute cortisol production resulting in distinct physiologicaland behavioral responses [72]
Results of two-dimensional parameter scanning of theenzymatic activities of CYP17H and HSD3B2 quantitativelyshowed the detail of the metabolic relationship betweenmin-eralocorticoid and glucocorticoid biosynthesis (Figure 7)Particularly the results showed that the balance of theseenzymatic activities was very important for the typicalfunction of NCI-H295R cells namely the ability to produceall steroid hormones NCI-H295R cells lost this functionwhen enzymatic activities of CYP17H and HSD3B2 werechanged by over 60 and 30 respectively In addition theybecame mineralocorticoid (ALDO) secreting cells when theenzymatic activity of CYP17H or HSD3B2 was inhibited byover 50 or glucocorticoid (DCORT and CORT) secretingcells when these enzymes were activated by over 50 Inparticular this analysis also showed that HSD3B2 was a keyplayer in the adrenal steroidogenesis of NCI-H295R cellsbecause HSD3B2 inhibition by over 80 almost completelyinhibited the biosynthesis of downstream steroids The ratioof CYP17A1 to HSD3B2 mRNA expression levels has beenrelated to several endocrine diseases with a low level in APAs[73] and high level in cortisol-producing adenomas [74]Furthermore the expression levels or enzymatic activitiesof CYP17A1 and HSD3B1 have been related to androgenproduction in polycystic ovary syndrome [75 76] Theseclinical studies support our simulation results indicating thatthe balance of enzymatic activity of CYP17H and HSD3B2determines the shift in steroidogenic output to mineralocor-ticoids glucocorticoids or androgens
43 Methodologies of Quantitative Mechanistic Analysis forDrug Discovery According to our results obtained using themathematical model of steroidogenesis in NCI-H295R cellssuch as sensitivity analysis comprehensive analysis based onsystems biology is available to quantitatively estimate themechanism of action of steroidogenic disrupting compoundsfrom differential profiling of adrenal steroid hormonesbecause dynamic patterns of steroid hormones in adrenalsteroidogenesis pathway are highly complex Our proposedmethod of quantitative mechanistic analysis of steroidogenicinhibitors was able to predict known action sites in theadrenal steroidogenesis pathway at only one time point (72 hafter drug exposure) Moreover according to the resultsof sensitivity analysis (Figure 6) 119881max of all steroidogenicenzymes was more sensitive than the 119870
119898 because the
intracellular concentrations of steroid hormones were almostmaintained at sufficiently high levels compared with 119870
119898
values of steroidogenic enzymesThese results suggested thatestimation of the mechanism of action of drugs is moreeffective and detectable when using the influences of 119881maxas the searching parameters such as our proposed methodOur data showed that the proposed method based on asystems biology model is a very powerful tool for exploratoryscreening of steroidogenic disrupting compounds
RCGA as a solver of parameter estimation problemsin systems biology has been applied to biological network
identification of gene regulatory networks and metabolicpathways and optimization of biological processes usingexperimentally observed time-course data [77ndash82] In thisstudy RCGA was useful to estimate the mechanism ofaction of novel pharmaceutical drug candidates for adrenalsteroidogenesis as a new application of RCGA in systemsbiology We had two issues when applying RCGA to thequantitative mechanistic analysis of drug actions Theseissues were the vast calculation cost and multimodality ofquasi-optimum solutions in solving the optimization prob-lem because the mathematical model in systems biologyconsists of many equations and parameters A proposedoptimization strategy using RCGA based on REXJGG wasa highly stable and efficient calculation method for a betterquasi-optimum solution than the unimodal normal distri-bution crossover (UNDX)minimum generation gap (MGG)method that is well applied in the engineering field In addi-tion we expanded the RCGA optimization program basedon REXJGG to a hybrid method of GA and then applieda local search as recommended by Harada and Kobayashi[28 83] A final optimal solution was obtained with a goodconvergence property Because these problems are generalin systems biology studies we suggest that the proposedhybrid method based on REXJGG is a very useful tool forquantitative mechanistic analysis of novel pharmaceuticaldrugs not limited to steroidogenic disrupting compounds
5 Conclusions
Thenovelmathematicalmodel of adrenal steroidogenesis wasconstructed in this study including cholesterol transport anddistribution the C
21-steroid hormone pathway steroid trans-
port and cell proliferation which could reproduce adrenalsteroidogenesis in NCI-H295R cells According to the resultsof dynamic sensitivity analysis using the newmodel HSD3B2plays the most important role in the metabolic balance ofadrenal steroidogenesis in NCI-H295R cells Moreover toquantitatively estimate mechanisms of action of adrenal toxiccompounds we analyzed differential metabolic profiles of12 steroid hormones at 3 days after exposure to 11 adrenaltoxic compounds by using the new mathematical modeland a hybrid optimization method of the RCGA and a localsearch (nonlinear least squares) We could estimate whichsteroidogenic enzymes were affected by these compoundsusing the hybrid optimization method Vasculotoxic agentswere estimated to have no effect according to the resultsobtained by our method In terms of adrenal steroidogenicinhibitors the predicted action sites were approximatelymatched to the target enzymes as reported in the literatureThus our computer-aided method based on a systems biol-ogy approach may be useful to analyze the mechanism ofaction of endocrine-disrupting compounds and to assess thehuman adrenal toxicity of novel pharmaceutical drugs basedon steroid hormone profiling
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Journal of Toxicology 17
Authorsrsquo Contribution
Ryuta Saito Natsuko Terasaki and Naohisa Tsutsui designedthe study Naohisa Tsutsui and Naoya Masutomi supervisedthe experiments Natsuko Terasaki performed all in vitroexperiments and collected dataMakotoYamazaki performedall measurements of steroids Ryuta Saito conceived the insilico strategies analyzed data developed all programs andthe mathematical model and performed all simulation anal-yses with Masahiro Okamoto Ryuta Saito Natsuko TerasakiMakoto Yamazaki NaoyaMasutomi andMasahiroOkamotowrote the paper
Acknowledgments
The authors would like to thank Dr Michael S Breenand Dr Miyuki Breen for helpful discussions concerningprevious mathematical models of adrenal steroidogenesisThis work was partially supported by the Grant-in-Aid forScientific Research on Innovative Areas ldquoSynthetic Biologyrdquo(No 23119001 (MO)) from Ministry of Education CultureSports Science and Technology in Japan
References
[1] P W Harvey D J Everett and C J Springall ldquoAdrenal toxi-cology a strategy for assessment of functional toxicity to theadrenal cortex and steroidogenesisrdquo Journal of Applied Toxi-cology vol 27 no 2 pp 103ndash115 2007
[2] W E Ribelin ldquoThe effects of drugs and chemicals upon thestructure of the adrenal glandrdquo Fundamental and Applied Toxi-cology vol 4 no 1 pp 105ndash119 1984
[3] H D Colby and P A Longhurst ldquoToxicology of the adrenalglandrdquo in Endocrine Toxicology C K Atterwill and J D FlackEds pp 243ndash281 CambridgeUniversity Press Cambridge UK1992
[4] H D Colby ldquoIn vitro assessment of adrenocortical toxicityrdquoJournal of Pharmacological and Toxicological Methods vol 32no 1 pp 1ndash6 1994
[5] T J Rosol J T Yarrington J Latendresse and C C CapenldquoAdrenal gland structure function and mechanisms of toxic-ityrdquo Toxicologic Pathology vol 29 no 1 pp 41ndash48 2001
[6] P W Harvey and D J Everett ldquoThe adrenal cortex and ster-oidogenesis as cellular and molecular targets for toxicity crit-ical omissions from regulatory endocrine disrupter screeningstrategies for human healthrdquo Journal of Applied Toxicology vol23 no 2 pp 81ndash87 2003
[7] A F Gazdar H K Oie C H Shackleton et al ldquoEstablishmentand characterization of a human adrenocortical carcinoma cellline that expresses multiple pathways of steroid biosynthesisrdquoCancer Research vol 50 no 17 pp 5488ndash5496 1990
[8] W E Rainey I M Bird C Sawetawan et al ldquoRegulation ofhuman adrenal carcinoma cell (NCI-H295) production of C19steroidsrdquo The Journal of Clinical Endocrinology amp Metabolismvol 77 no 3 pp 731ndash737 1993
[9] B Staels D W Hum andW L Miller ldquoRegulation of steroido-genesis in NCI-H295 cells a cellular model of the human fetaladrenalrdquo Molecular Endocrinology vol 7 no 3 pp 423ndash4331993
[10] M Hecker J L Newsted M B Murphy et al ldquoHuman adreno-carcinoma (H295R) cells for rapid in vitro determination of
effects on steroidogenesis hormone productionrdquo Toxicologyand Applied Pharmacology vol 217 no 1 pp 114ndash124 2006
[11] A Oskarsson E Ulleras K E Plant J P Hinson and P SGoldfarb ldquoSteroidogenic gene expression in H295R cells andthe human adrenal gland adrenotoxic effects of lindane invitrordquo Journal of Applied Toxicology vol 26 no 6 pp 484ndash4922006
[12] UMuller-Vieira M Angotti and RW Hartman ldquoThe adreno-cortical tumor cell line NCI-H295R as an in vitro screeningsystem for the evaluation of CYP11B2 (aldosterone synthase)and CYP11B1 (steroid-11120573-hydroxylase) inhibitorsrdquoThe Journalof Steroid Biochemistry and Molecular Biology vol 96 no 3-4pp 259ndash270 2005
[13] K Imagawa S Okayama M Takaoka et al ldquoInhibitory effectof efonidipine on aldosterone synthesis and secretion in humanadrenocarcinoma (H295R) cellsrdquo Journal of CardiovascularPharmacology vol 47 no 1 pp 133ndash138 2006
[14] M Voets UMuller-Vieira S Marchais-Oberwinkler and RWHartmann ldquoSynthesis of amidinohydrazones and evaluation oftheir inhibitory effect towards aldosterone synthase (CYP11B2)and the formation of selected steroidsrdquo Archiv der Pharmazievol 337 no 7 pp 411ndash416 2004
[15] F K Nielsen C H Hansen J A Fey et al ldquoH295R cells as amodel for steroidogenic disruption a broader perspective usingsimultaneous chemical analysis of 7 key steroid hormonesrdquoToxicology in Vitro vol 26 no 2 pp 343ndash350 2012
[16] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cortisolproduction and on activities of steroidogenic enzymes inhuman adrenocortical H295R cellsrdquo Journal of Steroid Biochem-istry and Molecular Biology vol 80 no 3 pp 355ndash363 2002
[17] R F Canton J T Sanderson S Nijmeijer A Bergman R JLetcher and M van den Berg ldquoIn vitro effects of brominatedflame retardants and metabolites on CYP17 catalytic activity anovel mechanism of actionrdquo Toxicology and Applied Pharma-cology vol 216 no 2 pp 274ndash281 2006
[18] K Hilscherova P D Jones T Gracia et al ldquoAssessment ofthe effects of chemicals on the expression of ten steroidogenicgenes in the H295R cell line using real-time PCRrdquo ToxicologicalSciences vol 81 no 1 pp 78ndash89 2004
[19] T Gracia K Hilscherova P D Jones et al ldquoThe H295R systemfor evaluation of endocrine-disrupting effectsrdquo Ecotoxicologyand Environmental Safety vol 65 no 3 pp 293ndash305 2006
[20] X Zhang J L NewstedM Hecker E B Higley P D Jones andJ P Giesy ldquoClassification of chemicals based on concentration-dependent toxicological data using ToxClustrdquo EnvironmentalScience and Technology vol 43 no 10 pp 3926ndash3932 2009
[21] M S Breen M Breen N Terasaki M Yamazaki and R BConolly ldquoComputational model of steroidogenesis in humanH295R cells to predict biochemical response to endocrine-active chemicalsmodel development formetyraponerdquoEnviron-mental Health Perspectives vol 118 no 2 pp 265ndash272 2010
[22] M Breen M S Breen N Terasaki M Yamazaki A LLloyd and R B Conolly ldquoMechanistic computational model ofsteroidogenesis in H295R cells role of oxysterols and cell pro-liferation to improve predictability of biochemical response toendocrine active chemical-metyraponerdquo Toxicological Sciencesvol 123 no 1 pp 80ndash93 2011
[23] F B Kraemer ldquoAdrenal cholesterol utilizationrdquo Molecular andCellular Endocrinology vol 265-266 pp 42ndash45 2007
[24] W L Miller ldquoSteroidogenic acute regulatory protein (StAR)a novel mitochondrial cholesterol transporterrdquo Biochimica et
18 Journal of Toxicology
Biophysica Acta (BBA)mdashMolecular and Cell Biology of Lipidsvol 1771 no 6 pp 663ndash676 2007
[25] S Szabo and I T Lippe ldquoAdrenal gland chemically inducedstructural and functional changes in the cortexrdquo ToxicologicPathology vol 17 no 2 pp 317ndash329 1989
[26] C Chen-Pan I-J Pan Y Yamamoto T Sakogawa J Yamadaand YHayashi ldquoPrompt recovery of damaged adrenalmedullaeinduced by salinomycinrdquoToxicologic Pathology vol 27 no 5 pp563ndash572 1999
[27] R J Santen and R I Misbin ldquoAminoglutethimide review ofpharmacology and clinical userdquo Pharmacotherapy vol 1 no 2pp 95ndash120 1981
[28] M J Carella N V Dimitrov V V Gossain L Srivastava andD R Rovner ldquoAdrenal effects of low-dose aminoglutethimidewhen used alone in postmenopausal women with advancedbreast cancerrdquoMetabolism vol 43 no 6 pp 723ndash727 1994
[29] M K Johansson J T Sanderson and B-O Lund ldquoEffects of3-MeSO2-DDE and some CYP inhibitors on glucocorticoidsteroidogenesis in the H295R human adrenocortical carcinomacell linerdquo Toxicology in Vitro vol 16 no 2 pp 113ndash121 2002
[30] Y Touitou A Bogdan J C Legrand and P Desgrez ldquoAmino-glutethimide and glutethimide effects on 18-hydroxycorti-costerone biosynthesis by human and sheep adrenals in vitrordquoActa Endocrinologica vol 80 no 3 pp 517ndash526 1975
[31] M M Hart and J A Straw ldquoStudies on the site of action ofop1015840-DDD in the dog adrenal cortex 1 Inhibition of ACTH-mediated pregnenolone synthesisrdquo Steroids vol 17 no 1ndash5 pp559ndash574 1971
[32] M Ojima M Saitoh N Itoh Y Kusano S Fukuchi and HNaganuma ldquoThe effects of op1015840-DDD on adrenal steroidogen-esis and hepatic steroid metabolismrdquo Nippon Naibunpi Gakkaizasshi vol 61 no 3 pp 168ndash178 1985
[33] L F Canosa and N R Ceballos ldquoEffects of different steroid-biosynthesis inhibitors on the testicular steroidogenesis of thetoad Bufo arenarumrdquo Journal of Comparative Physiology B vol171 no 6 pp 519ndash526 2001
[34] D C Kossor S Kominami S Takemori and H D ColbyldquoRole of the steroid 17 alpha-hydroxylase in spironolactone-mediated destruction of adrenal cytochrome P-450rdquoMolecularPharmacology vol 40 no 2 pp 321ndash325 1991
[35] S C Cheng K Suzuki W Sadee and B W Harding ldquoEffectsof spironolactone canrenone and canrenoate K on cytochromeP450 and 11120573 and 18 hydroxylation in bovine and humanadrenal corticalmitochondriardquo Endocrinology vol 99 no 4 pp1097ndash1106 1976
[36] K Yanagibashi M Haniu J E Shively W H Shen and P HallldquoThe synthesis of aldosterone by the adrenal cortex Two zones(fasciculata and glomerulosa) possess one enzyme for 11120573- 18-hydroxylation and aldehyde synthesisrdquoThe Journal of BiologicalChemistry vol 261 no 8 pp 3556ndash3562 1986
[37] K Morishita H Okumura N Ito and N Takahashi ldquoPrimaryculture system of adrenocortical cells from dogs to evaluatedirect effects of chemicals on steroidogenesisrdquo Toxicology vol165 no 2-3 pp 171ndash178 2001
[38] S W J Lamberts E G Bons H A Bruining and F H de JongldquoDifferential effects of the imidazole derivatives etomidateketoconazole and miconazole and of metyrapone on the secre-tion of cortisol and its precursors by human adrenocorticalcellsrdquo Journal of Pharmacology and Experimental Therapeuticsvol 240 no 1 pp 259ndash264 1987
[39] L Yin S Lucas F Maurer U Kazmaier Q Hu and RW Hartmann ldquoNovel imidazol-1-ylmethyl substituted 1256-tetrahydropyrrolo[321-ij]quinolin-4-ones as potent and selec-tive CYP11B1 inhibitors for the treatment of Cushingrsquos syn-dromerdquo Journal ofMedicinal Chemistry vol 55 no 14 pp 6629ndash6633 2012
[40] M DiMattina N Maronian H Ashby D L Loriaux andB D Albertson ldquoKetoconazole inhibits multiple steroidogenicenzymes involved in androgen biosynthesis in the humanovaryrdquo Fertility and Sterility vol 49 no 1 pp 62ndash65 1988
[41] M Ayub and M J Levell ldquoInhibition of human adrenal steroi-dogenic enzymes in vitro by imidazole drugs including keto-conazolerdquo Journal of Steroid Biochemistry vol 32 no 4 pp 515ndash524 1989
[42] D Engelhardt M M Weber T Miksch F Abedinpour andC Jaspers ldquoThe influence of ketoconazole on human adrenalsteroidogenesis incubation studies with tissue slicesrdquo ClinicalEndocrinology vol 35 no 2 pp 163ndash168 1991
[43] S Bhasin S Sikka T Fielder et al ldquoHormonal effects of keto-conazole in vivo in the male rat mechanism of actionrdquo Endo-crinology vol 118 no 3 pp 1229ndash1232 1986
[44] D M Zisterer and D C Williams ldquoCalmidazolium and otherimidazole compounds affect steroidogenesis in Y1 cells lack ofinvolvement of the peripheral-type benzodiazepine receptorrdquoJournal of Steroid Biochemistry and Molecular Biology vol 60no 3-4 pp 189ndash195 1997
[45] L P Walsh C N Kuratko and D M Stocco ldquoEconazoleand miconazole inhibit steroidogenesis and disrupt steroido-genic acute regulatory (StAR) protein expression post-trans-criptionallyrdquo Journal of Steroid Biochemistry and Molecular Bio-logy vol 75 no 4-5 pp 229ndash236 2000
[46] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cor-tisol production and on activities of steroidogenic enzymesin human adrenocortical H295R cellsrdquo The Journal of SteroidBiochemistry and Molecular Biology vol 80 no 3 pp 355ndash3632002
[47] D P Dempsher D S Gann and R D Phair ldquoA mechanisticmodel of ACTH-stimulated cortisol secretionrdquo American Jour-nal of Physiology vol 246 no 4 part 2 pp R587ndashR596 1984
[48] N Sarai S Matsuoka and A Noma ldquosimBio a Java package forthe development of detailed cell modelsrdquo Progress in Biophysicsand Molecular Biology vol 90 no 1ndash3 pp 360ndash377 2006
[49] K Levenberg ldquoA method for the solution of certain non-linearproblems in least squaresrdquo Quarterly of Applied Mathematicsvol 2 pp 164ndash168 1944
[50] D W Marquardt ldquoAn algorithm for least-squares estimation ofnonlinear parametersrdquo SIAM Journal on Applied Mathematicsvol 11 no 2 pp 431ndash441 1963
[51] P E Gill and W Murray ldquoAlgorithms for the solution of thenonlinear least-squares problemrdquo SIAM Journal on NumericalAnalysis vol 15 no 5 pp 977ndash992 1978
[52] Y Akimoto R Hasada J Sakuma I Ono and S KobayashildquoGeneration alternation model for real-coded GA using multi-parent proposal and evaluation of just generation gap (JGG)rdquoin Proceedings of the 19th SICE Symposium on DecentralizedAutonomous Systems pp 341ndash346 Tokyo Japan January 2007
[53] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoProposal and evaluation of adaptive real-coded crossoverAREXrdquo Transactions of the Japanese Society for Artificial Intel-ligence vol 24 no 6 pp 446ndash458 2009
Journal of Toxicology 19
[54] S Kobayashi ldquoThe frontiers of real-coded genetic algorithmsrdquoJournal of Japanese Society for Artificial Intelligence vol 24 no1 pp 147ndash162 2009
[55] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoAnalysis of the behavior of MGG and JGG as a selection modelfor real-coded genetic algorithmsrdquo Transactions of the JapaneseSociety for Artificial Intelligence vol 25 no 2 pp 281ndash289 2010
[56] A Struthers H Krum and G H Williams ldquoA comparisonof the aldosterone-blocking agents eplerenone and spironolac-tonerdquo Clinical Cardiology vol 31 no 4 pp 153ndash158 2008
[57] K Yamasaki M Sawaki S Noda et al ldquoComparison of theHershberger assay and androgen receptor binding assay oftwelve chemicalsrdquo Toxicology vol 195 no 2-3 pp 177ndash1862004
[58] C Eil and S K Edelson ldquoThe use of human skin fibroblasts toobtain potency estimates of drug binding to androgen recept-orsrdquo The Journal of Clinical Endocrinology amp Metabolism vol59 no 1 pp 51ndash55 1984
[59] P Ye T Yamashita D M Pollock H Sasano and W ERainey ldquoContrasting effects of eplerenone and spironolac-tone on adrenal cell steroidogenesisrdquo Hormone and MetabolicResearch vol 41 no 1 pp 35ndash39 2009
[60] B D Albertson K L Frederick N C Maronian et al ldquoTheeffect of ketoconazole on steroidogenesis I Leydig cell enzymeactivity in vitrordquo Research Communications in Chemical Patho-logy and Pharmacology vol 61 no 1 pp 17ndash26 1988
[61] B D Albertson N C Maronian K L Frederick et al ldquoTheeffect of ketoconazole on steroidogenesis II Adrenocorticalenzyme activity in vitrordquoResearch Communications in ChemicalPathology and Pharmacology vol 61 no 1 pp 27ndash34 1988
[62] DM Stocco andB J Clark ldquoRegulation of the acute productionof steroids in steroidogenic cellsrdquo Endocrine Reviews vol 17 no3 pp 221ndash244 1996
[63] L K Christenson and J F Strauss III ldquoSteroidogenic acute regu-latory protein (StAR) and the intramitochondrial translocationof cholesterolrdquo Biochimica et Biophysica Acta (BBA)mdashMolecularand Cell Biology of Lipids vol 1529 no 1ndash3 pp 175ndash187 2000
[64] E R Simpson ldquoCholesterol side-chain cleavage cytochromeP450 and the control of steroidogenesisrdquo Molecular and Cel-lular Endocrinology vol 13 no 3 pp 213ndash227 1979
[65] H A LaVoie and S R King ldquoTranscriptional regulation of ster-oidogenic genes STARD1 CYP11A1 and HSD3Brdquo ExperimentalBiology and Medicine vol 234 no 8 pp 880ndash907 2009
[66] F Raucci A DrsquoAniello and M M Di Fiore ldquoStimulation ofandrogen production by d-aspartate through the enhancementof StAR P450scc and 3120573-HSD mRNA levels in vivo rat testisand in culture of immature rat Leydig cellsrdquo Steroids vol 84pp 103ndash110 2014
[67] M Doi Y Takahashi R Komatsu et al ldquoSalt-sensitive hyper-tension in circadian clock-deficient Cry-null mice involves dys-regulated adrenal Hsd3b6rdquo Nature Medicine vol 16 no 1 pp67ndash74 2010
[68] H Okamura M Doi Y Yamaguchi and J-M Fustin ldquoHyper-tension due to loss of clock novel insight from the molecu-lar analysis of Cry1Cry2-deleted micerdquo Current HypertensionReports vol 13 no 2 pp 103ndash108 2011
[69] M Doi F Satoh T Maekawa et al ldquoIsoform-specific mono-clonal antibodies against 3120573-hydroxysteroid dehydrogenaseisomerase family provide markers for subclassification ofhuman primary aldosteronismrdquo The Journal of Clinical Endo-crinology amp Metabolism vol 99 no 2 pp E257ndashE262 2014
[70] S Konosu-Fukaya Y Nakamura F Satoh et al ldquo3120573-hydroxy-steroid dehydrogenase isoforms in human aldosterone-produc-ing adenomardquo Molecular and Cellular Endocrinology vol 408pp 205ndash212 2015
[71] G C Verwoert J Hofland N Amin et al ldquoExpression and genevariation studies deny association of humanHSD3B1 gene withaldosterone production or blood pressurerdquoAmerican Journal ofHypertension vol 28 no 1 pp 113ndash120 2015
[72] D Hough K Storbeck SW P Cloete A C Swart and P SwartldquoRelative contribution of P450c17 towards the acute cortisolresponse lessons from sheep and goatsrdquoMolecular and CellularEndocrinology vol 408 pp 107ndash113 2015
[73] I Sakuma S Suematsu Y Matsuzawa et al ldquoCharacterizationof steroidogenic enzyme expression in aldosterone-producingadenoma a comparison with various human adrenal tumorsrdquoEndocrine Journal vol 60 no 3 pp 329ndash336 2013
[74] A Tong A Jia S Yan Y Zhang Y Xie and G Liu ldquoEctopiccortisol-producing adrenocortical adenoma in the renal hilumhistopathological features and steroidogenic enzyme profilerdquoInternational Journal of Clinical and Experimental Pathologyvol 7 no 7 pp 4415ndash4421 2014
[75] A Hirsch D Hahn P Kempna et al ldquoMetformin inhibitshuman androgen production by regulating steroidogenicenzymes HSD3B2 and CYP17A1 and complex I activity of therespiratory chainrdquo Endocrinology vol 153 no 9 pp 4354ndash43662012
[76] B Xu L Gao Y Cui et al ldquoSET protein up-regulated test-osterone production in the cultured preantral folliclesrdquo Repro-ductive Biology and Endocrinology vol 11 article 9 2013
[77] S Kikuchi D Tominaga M Arita M Takahashi and MTomita ldquoDynamic modeling of genetic networks using geneticalgorithm and S-systemrdquo Bioinformatics vol 19 no 5 pp 643ndash650 2003
[78] N Shikata YMaki M Nakatsui et al ldquoDetermining importantregulatory relations of amino acids from dynamic networkanalysis of plasma amino acidsrdquo Amino Acids vol 38 no 1 pp179ndash187 2010
[79] M Nakatsui T Ueda Y Maki I Ono and M OkamotoldquoMethod for inferring and extracting reliable genetic interac-tions from time-series profile of gene expressionrdquoMathematicalBiosciences vol 215 no 1 pp 105ndash114 2008
[80] A Komori Y Maki M Nakatsui I Ono and M OkamotoldquoEfficient numerical optimization algorithm based on new real-coded genetic algorithm AREX + JGG and application to theinverse problem in systems biologyrdquo Applied Mathematics vol3 no 10 pp 1463ndash1470 2012
[81] J A Roubos G van Straten and A J B van Boxtel ldquoAn evolu-tionary strategy for fed-batch bioreactor optimization conceptsand performancerdquo Journal of Biotechnology vol 67 no 2-3 pp173ndash187 1999
[82] D Sarkar and J M Modak ldquoGenetic algorithms with filters foroptimal control problems in fed-batch bioreactorsrdquo Bioprocessand Biosystems Engineering vol 26 no 5 pp 295ndash306 2004
[83] K Harada K Ikeda J Sakuma I Ono and S KobayashildquoHybridization of genetic algorithm with local search in mul-tiobjective function optimization recommendation of GA thenLSrdquoTransactions of the Japanese Society for Artificial Intelligencevol 21 no 6 pp 482ndash492 2006
Submit your manuscripts athttpwwwhindawicom
PainResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
ToxinsJournal of
VaccinesJournal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AntibioticsInternational Journal of
ToxicologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
StrokeResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Drug DeliveryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Pharmacological Sciences
Tropical MedicineJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AddictionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Emergency Medicine InternationalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Autoimmune Diseases
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anesthesiology Research and Practice
ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Pharmaceutics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
MEDIATORSINFLAMMATION
of
Journal of Toxicology 15
All steroid producing cells
Mineralocorticoid producing cells Glucocorticoid producing cells
Mineralocorticoid and glucocorticoid producing cells
Terminated steroidogenesis
Sex steroids and precursor sex steroidsDIONE TESTO E1DEHA E2
NCI-H295Rcells at controlcondition
Normalized activities of HSD3B2
Nor
mal
ized
activ
ities
of C
YP17
H
2
18
16
14
12
1
08
06
04
02
00 05 1 15 2
Figure 7 Metabolic categories of the steroidogenic cells determined by the balance of HSD3B2 and CYP17H activitiesThe two-dimensionalparameter scanning analysis by the perturbation of two focused parameters clarifies the interaction and the relationship between the twoparameters in the complex system Functional cellular categories of steroidogenic cells were defined by the levels of mineralocorticoid(ALDO) glucocorticoids (DCORTandCORT) and androgens (DHEAandDIONE) at 72 h after stimulation Enzymatic activities ofHSD3B2andCYP17were normalized by standard values of the simulationmodel inNCI-H295R cells Scanning ranges ofHSD3B2 andCYP17 activitieswere 0ndash200 each 10 Green regions are all 14 steroids produced by NCI-H295R cells Red blue and purple regions are mineralocorticoidproducing cells glucocorticoids-producing cells and both corticoid-producing cells respectively Yellow regions are steroidogenesis of NCI-H295R cells terminated upstream of the adrenal steroidogenesis pathway
Recent clinical observations have revealed predominantexpression of HSD3B2 mRNA and protein in tumor cellsof aldosterone-producing adenoma (APA) and HSD3B1mRNA significantly correlated with CYP11B2 mRNA levelsand plasma aldosterone concentrations in APA patients [6970] However the relationship is unclear and disputed ina small-scale clinical study indicating that genetic variationin HSD3B1 affects blood pressure and hypertension in APApatients [71] The results of our simulation study suggest that3120573-HSD protein (human genes are HSD3B1 and HSD3B2) isone of the determination factors for the dynamic propertyof adrenal steroidogenesis Our results support the clinicalevidence of Doi and colleagues [69] and we believe that theHSD3B1 enzyme has a promising potential as novel drugtarget for endocrine hypertension
42 Metabolic Shift of Adrenal Steroidogenesis and the Contri-butions of HSD3B2 and CYP17H The metabolic propertiesof adrenal steroidogenesis in NCI-H295R cells were revealedby dynamic sensitivity analysis using themathematicalmodel(Figure 6) Mineralocorticoids such as DCORTICO COR-TICO and ALDO and intermediate steroids upstream of
the adrenal steroidogenesis pathway such as PREG HPREGDHEA PROG and HPROG were accelerated by reactionsof cholesterol import (119896Cholesterol Transport) StAR protein (119896MTR
119891
and 119896loc119891) and CYP11A1 (119881CYP11A1max ) On the other hand glu-
cocorticoids such as DCORT and CORT and sex hormonessuch as DIONE TESTO and E1 were suppressed by thesemodel parametersTherefore enhancement of the net adrenalsteroidogenesis capacity which supplies PREG precursor tothe pathway causes a production shift from glucocorticoidsto mineralocorticoids by substrate inhibitions of CYP17HHSD3B2 and CYP21A2 caused by accumulation of initialintermediate steroids such as PREG and PROG Sensitivi-ties of CYP17H (119881CYP17HmaxA ) and HSD3B2 (119881HSD3B2
maxA ) for theproducts were dynamically changed and these parametersdetermined the metabolic balance of downstream steroidsin the adrenal steroidogenesis pathway According to theseresults of dynamic sensitivity analysis of StAR CYP11A1CYP17H and HSD3B2 we suggest that the enhancementof CYP17H and HSD3B2 activity during ACTH stimulationwas important to shift the steroidogenic output away fromALDO biosynthesis towards CORT biosynthesis as wellas adrenal androgen production This suggestion partially
16 Journal of Toxicology
supports a comparative animal study in which molecularand cellular variations in CYP17H activity dramatically affectacute cortisol production resulting in distinct physiologicaland behavioral responses [72]
Results of two-dimensional parameter scanning of theenzymatic activities of CYP17H and HSD3B2 quantitativelyshowed the detail of the metabolic relationship betweenmin-eralocorticoid and glucocorticoid biosynthesis (Figure 7)Particularly the results showed that the balance of theseenzymatic activities was very important for the typicalfunction of NCI-H295R cells namely the ability to produceall steroid hormones NCI-H295R cells lost this functionwhen enzymatic activities of CYP17H and HSD3B2 werechanged by over 60 and 30 respectively In addition theybecame mineralocorticoid (ALDO) secreting cells when theenzymatic activity of CYP17H or HSD3B2 was inhibited byover 50 or glucocorticoid (DCORT and CORT) secretingcells when these enzymes were activated by over 50 Inparticular this analysis also showed that HSD3B2 was a keyplayer in the adrenal steroidogenesis of NCI-H295R cellsbecause HSD3B2 inhibition by over 80 almost completelyinhibited the biosynthesis of downstream steroids The ratioof CYP17A1 to HSD3B2 mRNA expression levels has beenrelated to several endocrine diseases with a low level in APAs[73] and high level in cortisol-producing adenomas [74]Furthermore the expression levels or enzymatic activitiesof CYP17A1 and HSD3B1 have been related to androgenproduction in polycystic ovary syndrome [75 76] Theseclinical studies support our simulation results indicating thatthe balance of enzymatic activity of CYP17H and HSD3B2determines the shift in steroidogenic output to mineralocor-ticoids glucocorticoids or androgens
43 Methodologies of Quantitative Mechanistic Analysis forDrug Discovery According to our results obtained using themathematical model of steroidogenesis in NCI-H295R cellssuch as sensitivity analysis comprehensive analysis based onsystems biology is available to quantitatively estimate themechanism of action of steroidogenic disrupting compoundsfrom differential profiling of adrenal steroid hormonesbecause dynamic patterns of steroid hormones in adrenalsteroidogenesis pathway are highly complex Our proposedmethod of quantitative mechanistic analysis of steroidogenicinhibitors was able to predict known action sites in theadrenal steroidogenesis pathway at only one time point (72 hafter drug exposure) Moreover according to the resultsof sensitivity analysis (Figure 6) 119881max of all steroidogenicenzymes was more sensitive than the 119870
119898 because the
intracellular concentrations of steroid hormones were almostmaintained at sufficiently high levels compared with 119870
119898
values of steroidogenic enzymesThese results suggested thatestimation of the mechanism of action of drugs is moreeffective and detectable when using the influences of 119881maxas the searching parameters such as our proposed methodOur data showed that the proposed method based on asystems biology model is a very powerful tool for exploratoryscreening of steroidogenic disrupting compounds
RCGA as a solver of parameter estimation problemsin systems biology has been applied to biological network
identification of gene regulatory networks and metabolicpathways and optimization of biological processes usingexperimentally observed time-course data [77ndash82] In thisstudy RCGA was useful to estimate the mechanism ofaction of novel pharmaceutical drug candidates for adrenalsteroidogenesis as a new application of RCGA in systemsbiology We had two issues when applying RCGA to thequantitative mechanistic analysis of drug actions Theseissues were the vast calculation cost and multimodality ofquasi-optimum solutions in solving the optimization prob-lem because the mathematical model in systems biologyconsists of many equations and parameters A proposedoptimization strategy using RCGA based on REXJGG wasa highly stable and efficient calculation method for a betterquasi-optimum solution than the unimodal normal distri-bution crossover (UNDX)minimum generation gap (MGG)method that is well applied in the engineering field In addi-tion we expanded the RCGA optimization program basedon REXJGG to a hybrid method of GA and then applieda local search as recommended by Harada and Kobayashi[28 83] A final optimal solution was obtained with a goodconvergence property Because these problems are generalin systems biology studies we suggest that the proposedhybrid method based on REXJGG is a very useful tool forquantitative mechanistic analysis of novel pharmaceuticaldrugs not limited to steroidogenic disrupting compounds
5 Conclusions
Thenovelmathematicalmodel of adrenal steroidogenesis wasconstructed in this study including cholesterol transport anddistribution the C
21-steroid hormone pathway steroid trans-
port and cell proliferation which could reproduce adrenalsteroidogenesis in NCI-H295R cells According to the resultsof dynamic sensitivity analysis using the newmodel HSD3B2plays the most important role in the metabolic balance ofadrenal steroidogenesis in NCI-H295R cells Moreover toquantitatively estimate mechanisms of action of adrenal toxiccompounds we analyzed differential metabolic profiles of12 steroid hormones at 3 days after exposure to 11 adrenaltoxic compounds by using the new mathematical modeland a hybrid optimization method of the RCGA and a localsearch (nonlinear least squares) We could estimate whichsteroidogenic enzymes were affected by these compoundsusing the hybrid optimization method Vasculotoxic agentswere estimated to have no effect according to the resultsobtained by our method In terms of adrenal steroidogenicinhibitors the predicted action sites were approximatelymatched to the target enzymes as reported in the literatureThus our computer-aided method based on a systems biol-ogy approach may be useful to analyze the mechanism ofaction of endocrine-disrupting compounds and to assess thehuman adrenal toxicity of novel pharmaceutical drugs basedon steroid hormone profiling
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Journal of Toxicology 17
Authorsrsquo Contribution
Ryuta Saito Natsuko Terasaki and Naohisa Tsutsui designedthe study Naohisa Tsutsui and Naoya Masutomi supervisedthe experiments Natsuko Terasaki performed all in vitroexperiments and collected dataMakotoYamazaki performedall measurements of steroids Ryuta Saito conceived the insilico strategies analyzed data developed all programs andthe mathematical model and performed all simulation anal-yses with Masahiro Okamoto Ryuta Saito Natsuko TerasakiMakoto Yamazaki NaoyaMasutomi andMasahiroOkamotowrote the paper
Acknowledgments
The authors would like to thank Dr Michael S Breenand Dr Miyuki Breen for helpful discussions concerningprevious mathematical models of adrenal steroidogenesisThis work was partially supported by the Grant-in-Aid forScientific Research on Innovative Areas ldquoSynthetic Biologyrdquo(No 23119001 (MO)) from Ministry of Education CultureSports Science and Technology in Japan
References
[1] P W Harvey D J Everett and C J Springall ldquoAdrenal toxi-cology a strategy for assessment of functional toxicity to theadrenal cortex and steroidogenesisrdquo Journal of Applied Toxi-cology vol 27 no 2 pp 103ndash115 2007
[2] W E Ribelin ldquoThe effects of drugs and chemicals upon thestructure of the adrenal glandrdquo Fundamental and Applied Toxi-cology vol 4 no 1 pp 105ndash119 1984
[3] H D Colby and P A Longhurst ldquoToxicology of the adrenalglandrdquo in Endocrine Toxicology C K Atterwill and J D FlackEds pp 243ndash281 CambridgeUniversity Press Cambridge UK1992
[4] H D Colby ldquoIn vitro assessment of adrenocortical toxicityrdquoJournal of Pharmacological and Toxicological Methods vol 32no 1 pp 1ndash6 1994
[5] T J Rosol J T Yarrington J Latendresse and C C CapenldquoAdrenal gland structure function and mechanisms of toxic-ityrdquo Toxicologic Pathology vol 29 no 1 pp 41ndash48 2001
[6] P W Harvey and D J Everett ldquoThe adrenal cortex and ster-oidogenesis as cellular and molecular targets for toxicity crit-ical omissions from regulatory endocrine disrupter screeningstrategies for human healthrdquo Journal of Applied Toxicology vol23 no 2 pp 81ndash87 2003
[7] A F Gazdar H K Oie C H Shackleton et al ldquoEstablishmentand characterization of a human adrenocortical carcinoma cellline that expresses multiple pathways of steroid biosynthesisrdquoCancer Research vol 50 no 17 pp 5488ndash5496 1990
[8] W E Rainey I M Bird C Sawetawan et al ldquoRegulation ofhuman adrenal carcinoma cell (NCI-H295) production of C19steroidsrdquo The Journal of Clinical Endocrinology amp Metabolismvol 77 no 3 pp 731ndash737 1993
[9] B Staels D W Hum andW L Miller ldquoRegulation of steroido-genesis in NCI-H295 cells a cellular model of the human fetaladrenalrdquo Molecular Endocrinology vol 7 no 3 pp 423ndash4331993
[10] M Hecker J L Newsted M B Murphy et al ldquoHuman adreno-carcinoma (H295R) cells for rapid in vitro determination of
effects on steroidogenesis hormone productionrdquo Toxicologyand Applied Pharmacology vol 217 no 1 pp 114ndash124 2006
[11] A Oskarsson E Ulleras K E Plant J P Hinson and P SGoldfarb ldquoSteroidogenic gene expression in H295R cells andthe human adrenal gland adrenotoxic effects of lindane invitrordquo Journal of Applied Toxicology vol 26 no 6 pp 484ndash4922006
[12] UMuller-Vieira M Angotti and RW Hartman ldquoThe adreno-cortical tumor cell line NCI-H295R as an in vitro screeningsystem for the evaluation of CYP11B2 (aldosterone synthase)and CYP11B1 (steroid-11120573-hydroxylase) inhibitorsrdquoThe Journalof Steroid Biochemistry and Molecular Biology vol 96 no 3-4pp 259ndash270 2005
[13] K Imagawa S Okayama M Takaoka et al ldquoInhibitory effectof efonidipine on aldosterone synthesis and secretion in humanadrenocarcinoma (H295R) cellsrdquo Journal of CardiovascularPharmacology vol 47 no 1 pp 133ndash138 2006
[14] M Voets UMuller-Vieira S Marchais-Oberwinkler and RWHartmann ldquoSynthesis of amidinohydrazones and evaluation oftheir inhibitory effect towards aldosterone synthase (CYP11B2)and the formation of selected steroidsrdquo Archiv der Pharmazievol 337 no 7 pp 411ndash416 2004
[15] F K Nielsen C H Hansen J A Fey et al ldquoH295R cells as amodel for steroidogenic disruption a broader perspective usingsimultaneous chemical analysis of 7 key steroid hormonesrdquoToxicology in Vitro vol 26 no 2 pp 343ndash350 2012
[16] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cortisolproduction and on activities of steroidogenic enzymes inhuman adrenocortical H295R cellsrdquo Journal of Steroid Biochem-istry and Molecular Biology vol 80 no 3 pp 355ndash363 2002
[17] R F Canton J T Sanderson S Nijmeijer A Bergman R JLetcher and M van den Berg ldquoIn vitro effects of brominatedflame retardants and metabolites on CYP17 catalytic activity anovel mechanism of actionrdquo Toxicology and Applied Pharma-cology vol 216 no 2 pp 274ndash281 2006
[18] K Hilscherova P D Jones T Gracia et al ldquoAssessment ofthe effects of chemicals on the expression of ten steroidogenicgenes in the H295R cell line using real-time PCRrdquo ToxicologicalSciences vol 81 no 1 pp 78ndash89 2004
[19] T Gracia K Hilscherova P D Jones et al ldquoThe H295R systemfor evaluation of endocrine-disrupting effectsrdquo Ecotoxicologyand Environmental Safety vol 65 no 3 pp 293ndash305 2006
[20] X Zhang J L NewstedM Hecker E B Higley P D Jones andJ P Giesy ldquoClassification of chemicals based on concentration-dependent toxicological data using ToxClustrdquo EnvironmentalScience and Technology vol 43 no 10 pp 3926ndash3932 2009
[21] M S Breen M Breen N Terasaki M Yamazaki and R BConolly ldquoComputational model of steroidogenesis in humanH295R cells to predict biochemical response to endocrine-active chemicalsmodel development formetyraponerdquoEnviron-mental Health Perspectives vol 118 no 2 pp 265ndash272 2010
[22] M Breen M S Breen N Terasaki M Yamazaki A LLloyd and R B Conolly ldquoMechanistic computational model ofsteroidogenesis in H295R cells role of oxysterols and cell pro-liferation to improve predictability of biochemical response toendocrine active chemical-metyraponerdquo Toxicological Sciencesvol 123 no 1 pp 80ndash93 2011
[23] F B Kraemer ldquoAdrenal cholesterol utilizationrdquo Molecular andCellular Endocrinology vol 265-266 pp 42ndash45 2007
[24] W L Miller ldquoSteroidogenic acute regulatory protein (StAR)a novel mitochondrial cholesterol transporterrdquo Biochimica et
18 Journal of Toxicology
Biophysica Acta (BBA)mdashMolecular and Cell Biology of Lipidsvol 1771 no 6 pp 663ndash676 2007
[25] S Szabo and I T Lippe ldquoAdrenal gland chemically inducedstructural and functional changes in the cortexrdquo ToxicologicPathology vol 17 no 2 pp 317ndash329 1989
[26] C Chen-Pan I-J Pan Y Yamamoto T Sakogawa J Yamadaand YHayashi ldquoPrompt recovery of damaged adrenalmedullaeinduced by salinomycinrdquoToxicologic Pathology vol 27 no 5 pp563ndash572 1999
[27] R J Santen and R I Misbin ldquoAminoglutethimide review ofpharmacology and clinical userdquo Pharmacotherapy vol 1 no 2pp 95ndash120 1981
[28] M J Carella N V Dimitrov V V Gossain L Srivastava andD R Rovner ldquoAdrenal effects of low-dose aminoglutethimidewhen used alone in postmenopausal women with advancedbreast cancerrdquoMetabolism vol 43 no 6 pp 723ndash727 1994
[29] M K Johansson J T Sanderson and B-O Lund ldquoEffects of3-MeSO2-DDE and some CYP inhibitors on glucocorticoidsteroidogenesis in the H295R human adrenocortical carcinomacell linerdquo Toxicology in Vitro vol 16 no 2 pp 113ndash121 2002
[30] Y Touitou A Bogdan J C Legrand and P Desgrez ldquoAmino-glutethimide and glutethimide effects on 18-hydroxycorti-costerone biosynthesis by human and sheep adrenals in vitrordquoActa Endocrinologica vol 80 no 3 pp 517ndash526 1975
[31] M M Hart and J A Straw ldquoStudies on the site of action ofop1015840-DDD in the dog adrenal cortex 1 Inhibition of ACTH-mediated pregnenolone synthesisrdquo Steroids vol 17 no 1ndash5 pp559ndash574 1971
[32] M Ojima M Saitoh N Itoh Y Kusano S Fukuchi and HNaganuma ldquoThe effects of op1015840-DDD on adrenal steroidogen-esis and hepatic steroid metabolismrdquo Nippon Naibunpi Gakkaizasshi vol 61 no 3 pp 168ndash178 1985
[33] L F Canosa and N R Ceballos ldquoEffects of different steroid-biosynthesis inhibitors on the testicular steroidogenesis of thetoad Bufo arenarumrdquo Journal of Comparative Physiology B vol171 no 6 pp 519ndash526 2001
[34] D C Kossor S Kominami S Takemori and H D ColbyldquoRole of the steroid 17 alpha-hydroxylase in spironolactone-mediated destruction of adrenal cytochrome P-450rdquoMolecularPharmacology vol 40 no 2 pp 321ndash325 1991
[35] S C Cheng K Suzuki W Sadee and B W Harding ldquoEffectsof spironolactone canrenone and canrenoate K on cytochromeP450 and 11120573 and 18 hydroxylation in bovine and humanadrenal corticalmitochondriardquo Endocrinology vol 99 no 4 pp1097ndash1106 1976
[36] K Yanagibashi M Haniu J E Shively W H Shen and P HallldquoThe synthesis of aldosterone by the adrenal cortex Two zones(fasciculata and glomerulosa) possess one enzyme for 11120573- 18-hydroxylation and aldehyde synthesisrdquoThe Journal of BiologicalChemistry vol 261 no 8 pp 3556ndash3562 1986
[37] K Morishita H Okumura N Ito and N Takahashi ldquoPrimaryculture system of adrenocortical cells from dogs to evaluatedirect effects of chemicals on steroidogenesisrdquo Toxicology vol165 no 2-3 pp 171ndash178 2001
[38] S W J Lamberts E G Bons H A Bruining and F H de JongldquoDifferential effects of the imidazole derivatives etomidateketoconazole and miconazole and of metyrapone on the secre-tion of cortisol and its precursors by human adrenocorticalcellsrdquo Journal of Pharmacology and Experimental Therapeuticsvol 240 no 1 pp 259ndash264 1987
[39] L Yin S Lucas F Maurer U Kazmaier Q Hu and RW Hartmann ldquoNovel imidazol-1-ylmethyl substituted 1256-tetrahydropyrrolo[321-ij]quinolin-4-ones as potent and selec-tive CYP11B1 inhibitors for the treatment of Cushingrsquos syn-dromerdquo Journal ofMedicinal Chemistry vol 55 no 14 pp 6629ndash6633 2012
[40] M DiMattina N Maronian H Ashby D L Loriaux andB D Albertson ldquoKetoconazole inhibits multiple steroidogenicenzymes involved in androgen biosynthesis in the humanovaryrdquo Fertility and Sterility vol 49 no 1 pp 62ndash65 1988
[41] M Ayub and M J Levell ldquoInhibition of human adrenal steroi-dogenic enzymes in vitro by imidazole drugs including keto-conazolerdquo Journal of Steroid Biochemistry vol 32 no 4 pp 515ndash524 1989
[42] D Engelhardt M M Weber T Miksch F Abedinpour andC Jaspers ldquoThe influence of ketoconazole on human adrenalsteroidogenesis incubation studies with tissue slicesrdquo ClinicalEndocrinology vol 35 no 2 pp 163ndash168 1991
[43] S Bhasin S Sikka T Fielder et al ldquoHormonal effects of keto-conazole in vivo in the male rat mechanism of actionrdquo Endo-crinology vol 118 no 3 pp 1229ndash1232 1986
[44] D M Zisterer and D C Williams ldquoCalmidazolium and otherimidazole compounds affect steroidogenesis in Y1 cells lack ofinvolvement of the peripheral-type benzodiazepine receptorrdquoJournal of Steroid Biochemistry and Molecular Biology vol 60no 3-4 pp 189ndash195 1997
[45] L P Walsh C N Kuratko and D M Stocco ldquoEconazoleand miconazole inhibit steroidogenesis and disrupt steroido-genic acute regulatory (StAR) protein expression post-trans-criptionallyrdquo Journal of Steroid Biochemistry and Molecular Bio-logy vol 75 no 4-5 pp 229ndash236 2000
[46] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cor-tisol production and on activities of steroidogenic enzymesin human adrenocortical H295R cellsrdquo The Journal of SteroidBiochemistry and Molecular Biology vol 80 no 3 pp 355ndash3632002
[47] D P Dempsher D S Gann and R D Phair ldquoA mechanisticmodel of ACTH-stimulated cortisol secretionrdquo American Jour-nal of Physiology vol 246 no 4 part 2 pp R587ndashR596 1984
[48] N Sarai S Matsuoka and A Noma ldquosimBio a Java package forthe development of detailed cell modelsrdquo Progress in Biophysicsand Molecular Biology vol 90 no 1ndash3 pp 360ndash377 2006
[49] K Levenberg ldquoA method for the solution of certain non-linearproblems in least squaresrdquo Quarterly of Applied Mathematicsvol 2 pp 164ndash168 1944
[50] D W Marquardt ldquoAn algorithm for least-squares estimation ofnonlinear parametersrdquo SIAM Journal on Applied Mathematicsvol 11 no 2 pp 431ndash441 1963
[51] P E Gill and W Murray ldquoAlgorithms for the solution of thenonlinear least-squares problemrdquo SIAM Journal on NumericalAnalysis vol 15 no 5 pp 977ndash992 1978
[52] Y Akimoto R Hasada J Sakuma I Ono and S KobayashildquoGeneration alternation model for real-coded GA using multi-parent proposal and evaluation of just generation gap (JGG)rdquoin Proceedings of the 19th SICE Symposium on DecentralizedAutonomous Systems pp 341ndash346 Tokyo Japan January 2007
[53] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoProposal and evaluation of adaptive real-coded crossoverAREXrdquo Transactions of the Japanese Society for Artificial Intel-ligence vol 24 no 6 pp 446ndash458 2009
Journal of Toxicology 19
[54] S Kobayashi ldquoThe frontiers of real-coded genetic algorithmsrdquoJournal of Japanese Society for Artificial Intelligence vol 24 no1 pp 147ndash162 2009
[55] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoAnalysis of the behavior of MGG and JGG as a selection modelfor real-coded genetic algorithmsrdquo Transactions of the JapaneseSociety for Artificial Intelligence vol 25 no 2 pp 281ndash289 2010
[56] A Struthers H Krum and G H Williams ldquoA comparisonof the aldosterone-blocking agents eplerenone and spironolac-tonerdquo Clinical Cardiology vol 31 no 4 pp 153ndash158 2008
[57] K Yamasaki M Sawaki S Noda et al ldquoComparison of theHershberger assay and androgen receptor binding assay oftwelve chemicalsrdquo Toxicology vol 195 no 2-3 pp 177ndash1862004
[58] C Eil and S K Edelson ldquoThe use of human skin fibroblasts toobtain potency estimates of drug binding to androgen recept-orsrdquo The Journal of Clinical Endocrinology amp Metabolism vol59 no 1 pp 51ndash55 1984
[59] P Ye T Yamashita D M Pollock H Sasano and W ERainey ldquoContrasting effects of eplerenone and spironolac-tone on adrenal cell steroidogenesisrdquo Hormone and MetabolicResearch vol 41 no 1 pp 35ndash39 2009
[60] B D Albertson K L Frederick N C Maronian et al ldquoTheeffect of ketoconazole on steroidogenesis I Leydig cell enzymeactivity in vitrordquo Research Communications in Chemical Patho-logy and Pharmacology vol 61 no 1 pp 17ndash26 1988
[61] B D Albertson N C Maronian K L Frederick et al ldquoTheeffect of ketoconazole on steroidogenesis II Adrenocorticalenzyme activity in vitrordquoResearch Communications in ChemicalPathology and Pharmacology vol 61 no 1 pp 27ndash34 1988
[62] DM Stocco andB J Clark ldquoRegulation of the acute productionof steroids in steroidogenic cellsrdquo Endocrine Reviews vol 17 no3 pp 221ndash244 1996
[63] L K Christenson and J F Strauss III ldquoSteroidogenic acute regu-latory protein (StAR) and the intramitochondrial translocationof cholesterolrdquo Biochimica et Biophysica Acta (BBA)mdashMolecularand Cell Biology of Lipids vol 1529 no 1ndash3 pp 175ndash187 2000
[64] E R Simpson ldquoCholesterol side-chain cleavage cytochromeP450 and the control of steroidogenesisrdquo Molecular and Cel-lular Endocrinology vol 13 no 3 pp 213ndash227 1979
[65] H A LaVoie and S R King ldquoTranscriptional regulation of ster-oidogenic genes STARD1 CYP11A1 and HSD3Brdquo ExperimentalBiology and Medicine vol 234 no 8 pp 880ndash907 2009
[66] F Raucci A DrsquoAniello and M M Di Fiore ldquoStimulation ofandrogen production by d-aspartate through the enhancementof StAR P450scc and 3120573-HSD mRNA levels in vivo rat testisand in culture of immature rat Leydig cellsrdquo Steroids vol 84pp 103ndash110 2014
[67] M Doi Y Takahashi R Komatsu et al ldquoSalt-sensitive hyper-tension in circadian clock-deficient Cry-null mice involves dys-regulated adrenal Hsd3b6rdquo Nature Medicine vol 16 no 1 pp67ndash74 2010
[68] H Okamura M Doi Y Yamaguchi and J-M Fustin ldquoHyper-tension due to loss of clock novel insight from the molecu-lar analysis of Cry1Cry2-deleted micerdquo Current HypertensionReports vol 13 no 2 pp 103ndash108 2011
[69] M Doi F Satoh T Maekawa et al ldquoIsoform-specific mono-clonal antibodies against 3120573-hydroxysteroid dehydrogenaseisomerase family provide markers for subclassification ofhuman primary aldosteronismrdquo The Journal of Clinical Endo-crinology amp Metabolism vol 99 no 2 pp E257ndashE262 2014
[70] S Konosu-Fukaya Y Nakamura F Satoh et al ldquo3120573-hydroxy-steroid dehydrogenase isoforms in human aldosterone-produc-ing adenomardquo Molecular and Cellular Endocrinology vol 408pp 205ndash212 2015
[71] G C Verwoert J Hofland N Amin et al ldquoExpression and genevariation studies deny association of humanHSD3B1 gene withaldosterone production or blood pressurerdquoAmerican Journal ofHypertension vol 28 no 1 pp 113ndash120 2015
[72] D Hough K Storbeck SW P Cloete A C Swart and P SwartldquoRelative contribution of P450c17 towards the acute cortisolresponse lessons from sheep and goatsrdquoMolecular and CellularEndocrinology vol 408 pp 107ndash113 2015
[73] I Sakuma S Suematsu Y Matsuzawa et al ldquoCharacterizationof steroidogenic enzyme expression in aldosterone-producingadenoma a comparison with various human adrenal tumorsrdquoEndocrine Journal vol 60 no 3 pp 329ndash336 2013
[74] A Tong A Jia S Yan Y Zhang Y Xie and G Liu ldquoEctopiccortisol-producing adrenocortical adenoma in the renal hilumhistopathological features and steroidogenic enzyme profilerdquoInternational Journal of Clinical and Experimental Pathologyvol 7 no 7 pp 4415ndash4421 2014
[75] A Hirsch D Hahn P Kempna et al ldquoMetformin inhibitshuman androgen production by regulating steroidogenicenzymes HSD3B2 and CYP17A1 and complex I activity of therespiratory chainrdquo Endocrinology vol 153 no 9 pp 4354ndash43662012
[76] B Xu L Gao Y Cui et al ldquoSET protein up-regulated test-osterone production in the cultured preantral folliclesrdquo Repro-ductive Biology and Endocrinology vol 11 article 9 2013
[77] S Kikuchi D Tominaga M Arita M Takahashi and MTomita ldquoDynamic modeling of genetic networks using geneticalgorithm and S-systemrdquo Bioinformatics vol 19 no 5 pp 643ndash650 2003
[78] N Shikata YMaki M Nakatsui et al ldquoDetermining importantregulatory relations of amino acids from dynamic networkanalysis of plasma amino acidsrdquo Amino Acids vol 38 no 1 pp179ndash187 2010
[79] M Nakatsui T Ueda Y Maki I Ono and M OkamotoldquoMethod for inferring and extracting reliable genetic interac-tions from time-series profile of gene expressionrdquoMathematicalBiosciences vol 215 no 1 pp 105ndash114 2008
[80] A Komori Y Maki M Nakatsui I Ono and M OkamotoldquoEfficient numerical optimization algorithm based on new real-coded genetic algorithm AREX + JGG and application to theinverse problem in systems biologyrdquo Applied Mathematics vol3 no 10 pp 1463ndash1470 2012
[81] J A Roubos G van Straten and A J B van Boxtel ldquoAn evolu-tionary strategy for fed-batch bioreactor optimization conceptsand performancerdquo Journal of Biotechnology vol 67 no 2-3 pp173ndash187 1999
[82] D Sarkar and J M Modak ldquoGenetic algorithms with filters foroptimal control problems in fed-batch bioreactorsrdquo Bioprocessand Biosystems Engineering vol 26 no 5 pp 295ndash306 2004
[83] K Harada K Ikeda J Sakuma I Ono and S KobayashildquoHybridization of genetic algorithm with local search in mul-tiobjective function optimization recommendation of GA thenLSrdquoTransactions of the Japanese Society for Artificial Intelligencevol 21 no 6 pp 482ndash492 2006
Submit your manuscripts athttpwwwhindawicom
PainResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
ToxinsJournal of
VaccinesJournal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AntibioticsInternational Journal of
ToxicologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
StrokeResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Drug DeliveryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Pharmacological Sciences
Tropical MedicineJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AddictionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Emergency Medicine InternationalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Autoimmune Diseases
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anesthesiology Research and Practice
ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Pharmaceutics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
MEDIATORSINFLAMMATION
of
16 Journal of Toxicology
supports a comparative animal study in which molecularand cellular variations in CYP17H activity dramatically affectacute cortisol production resulting in distinct physiologicaland behavioral responses [72]
Results of two-dimensional parameter scanning of theenzymatic activities of CYP17H and HSD3B2 quantitativelyshowed the detail of the metabolic relationship betweenmin-eralocorticoid and glucocorticoid biosynthesis (Figure 7)Particularly the results showed that the balance of theseenzymatic activities was very important for the typicalfunction of NCI-H295R cells namely the ability to produceall steroid hormones NCI-H295R cells lost this functionwhen enzymatic activities of CYP17H and HSD3B2 werechanged by over 60 and 30 respectively In addition theybecame mineralocorticoid (ALDO) secreting cells when theenzymatic activity of CYP17H or HSD3B2 was inhibited byover 50 or glucocorticoid (DCORT and CORT) secretingcells when these enzymes were activated by over 50 Inparticular this analysis also showed that HSD3B2 was a keyplayer in the adrenal steroidogenesis of NCI-H295R cellsbecause HSD3B2 inhibition by over 80 almost completelyinhibited the biosynthesis of downstream steroids The ratioof CYP17A1 to HSD3B2 mRNA expression levels has beenrelated to several endocrine diseases with a low level in APAs[73] and high level in cortisol-producing adenomas [74]Furthermore the expression levels or enzymatic activitiesof CYP17A1 and HSD3B1 have been related to androgenproduction in polycystic ovary syndrome [75 76] Theseclinical studies support our simulation results indicating thatthe balance of enzymatic activity of CYP17H and HSD3B2determines the shift in steroidogenic output to mineralocor-ticoids glucocorticoids or androgens
43 Methodologies of Quantitative Mechanistic Analysis forDrug Discovery According to our results obtained using themathematical model of steroidogenesis in NCI-H295R cellssuch as sensitivity analysis comprehensive analysis based onsystems biology is available to quantitatively estimate themechanism of action of steroidogenic disrupting compoundsfrom differential profiling of adrenal steroid hormonesbecause dynamic patterns of steroid hormones in adrenalsteroidogenesis pathway are highly complex Our proposedmethod of quantitative mechanistic analysis of steroidogenicinhibitors was able to predict known action sites in theadrenal steroidogenesis pathway at only one time point (72 hafter drug exposure) Moreover according to the resultsof sensitivity analysis (Figure 6) 119881max of all steroidogenicenzymes was more sensitive than the 119870
119898 because the
intracellular concentrations of steroid hormones were almostmaintained at sufficiently high levels compared with 119870
119898
values of steroidogenic enzymesThese results suggested thatestimation of the mechanism of action of drugs is moreeffective and detectable when using the influences of 119881maxas the searching parameters such as our proposed methodOur data showed that the proposed method based on asystems biology model is a very powerful tool for exploratoryscreening of steroidogenic disrupting compounds
RCGA as a solver of parameter estimation problemsin systems biology has been applied to biological network
identification of gene regulatory networks and metabolicpathways and optimization of biological processes usingexperimentally observed time-course data [77ndash82] In thisstudy RCGA was useful to estimate the mechanism ofaction of novel pharmaceutical drug candidates for adrenalsteroidogenesis as a new application of RCGA in systemsbiology We had two issues when applying RCGA to thequantitative mechanistic analysis of drug actions Theseissues were the vast calculation cost and multimodality ofquasi-optimum solutions in solving the optimization prob-lem because the mathematical model in systems biologyconsists of many equations and parameters A proposedoptimization strategy using RCGA based on REXJGG wasa highly stable and efficient calculation method for a betterquasi-optimum solution than the unimodal normal distri-bution crossover (UNDX)minimum generation gap (MGG)method that is well applied in the engineering field In addi-tion we expanded the RCGA optimization program basedon REXJGG to a hybrid method of GA and then applieda local search as recommended by Harada and Kobayashi[28 83] A final optimal solution was obtained with a goodconvergence property Because these problems are generalin systems biology studies we suggest that the proposedhybrid method based on REXJGG is a very useful tool forquantitative mechanistic analysis of novel pharmaceuticaldrugs not limited to steroidogenic disrupting compounds
5 Conclusions
Thenovelmathematicalmodel of adrenal steroidogenesis wasconstructed in this study including cholesterol transport anddistribution the C
21-steroid hormone pathway steroid trans-
port and cell proliferation which could reproduce adrenalsteroidogenesis in NCI-H295R cells According to the resultsof dynamic sensitivity analysis using the newmodel HSD3B2plays the most important role in the metabolic balance ofadrenal steroidogenesis in NCI-H295R cells Moreover toquantitatively estimate mechanisms of action of adrenal toxiccompounds we analyzed differential metabolic profiles of12 steroid hormones at 3 days after exposure to 11 adrenaltoxic compounds by using the new mathematical modeland a hybrid optimization method of the RCGA and a localsearch (nonlinear least squares) We could estimate whichsteroidogenic enzymes were affected by these compoundsusing the hybrid optimization method Vasculotoxic agentswere estimated to have no effect according to the resultsobtained by our method In terms of adrenal steroidogenicinhibitors the predicted action sites were approximatelymatched to the target enzymes as reported in the literatureThus our computer-aided method based on a systems biol-ogy approach may be useful to analyze the mechanism ofaction of endocrine-disrupting compounds and to assess thehuman adrenal toxicity of novel pharmaceutical drugs basedon steroid hormone profiling
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
Journal of Toxicology 17
Authorsrsquo Contribution
Ryuta Saito Natsuko Terasaki and Naohisa Tsutsui designedthe study Naohisa Tsutsui and Naoya Masutomi supervisedthe experiments Natsuko Terasaki performed all in vitroexperiments and collected dataMakotoYamazaki performedall measurements of steroids Ryuta Saito conceived the insilico strategies analyzed data developed all programs andthe mathematical model and performed all simulation anal-yses with Masahiro Okamoto Ryuta Saito Natsuko TerasakiMakoto Yamazaki NaoyaMasutomi andMasahiroOkamotowrote the paper
Acknowledgments
The authors would like to thank Dr Michael S Breenand Dr Miyuki Breen for helpful discussions concerningprevious mathematical models of adrenal steroidogenesisThis work was partially supported by the Grant-in-Aid forScientific Research on Innovative Areas ldquoSynthetic Biologyrdquo(No 23119001 (MO)) from Ministry of Education CultureSports Science and Technology in Japan
References
[1] P W Harvey D J Everett and C J Springall ldquoAdrenal toxi-cology a strategy for assessment of functional toxicity to theadrenal cortex and steroidogenesisrdquo Journal of Applied Toxi-cology vol 27 no 2 pp 103ndash115 2007
[2] W E Ribelin ldquoThe effects of drugs and chemicals upon thestructure of the adrenal glandrdquo Fundamental and Applied Toxi-cology vol 4 no 1 pp 105ndash119 1984
[3] H D Colby and P A Longhurst ldquoToxicology of the adrenalglandrdquo in Endocrine Toxicology C K Atterwill and J D FlackEds pp 243ndash281 CambridgeUniversity Press Cambridge UK1992
[4] H D Colby ldquoIn vitro assessment of adrenocortical toxicityrdquoJournal of Pharmacological and Toxicological Methods vol 32no 1 pp 1ndash6 1994
[5] T J Rosol J T Yarrington J Latendresse and C C CapenldquoAdrenal gland structure function and mechanisms of toxic-ityrdquo Toxicologic Pathology vol 29 no 1 pp 41ndash48 2001
[6] P W Harvey and D J Everett ldquoThe adrenal cortex and ster-oidogenesis as cellular and molecular targets for toxicity crit-ical omissions from regulatory endocrine disrupter screeningstrategies for human healthrdquo Journal of Applied Toxicology vol23 no 2 pp 81ndash87 2003
[7] A F Gazdar H K Oie C H Shackleton et al ldquoEstablishmentand characterization of a human adrenocortical carcinoma cellline that expresses multiple pathways of steroid biosynthesisrdquoCancer Research vol 50 no 17 pp 5488ndash5496 1990
[8] W E Rainey I M Bird C Sawetawan et al ldquoRegulation ofhuman adrenal carcinoma cell (NCI-H295) production of C19steroidsrdquo The Journal of Clinical Endocrinology amp Metabolismvol 77 no 3 pp 731ndash737 1993
[9] B Staels D W Hum andW L Miller ldquoRegulation of steroido-genesis in NCI-H295 cells a cellular model of the human fetaladrenalrdquo Molecular Endocrinology vol 7 no 3 pp 423ndash4331993
[10] M Hecker J L Newsted M B Murphy et al ldquoHuman adreno-carcinoma (H295R) cells for rapid in vitro determination of
effects on steroidogenesis hormone productionrdquo Toxicologyand Applied Pharmacology vol 217 no 1 pp 114ndash124 2006
[11] A Oskarsson E Ulleras K E Plant J P Hinson and P SGoldfarb ldquoSteroidogenic gene expression in H295R cells andthe human adrenal gland adrenotoxic effects of lindane invitrordquo Journal of Applied Toxicology vol 26 no 6 pp 484ndash4922006
[12] UMuller-Vieira M Angotti and RW Hartman ldquoThe adreno-cortical tumor cell line NCI-H295R as an in vitro screeningsystem for the evaluation of CYP11B2 (aldosterone synthase)and CYP11B1 (steroid-11120573-hydroxylase) inhibitorsrdquoThe Journalof Steroid Biochemistry and Molecular Biology vol 96 no 3-4pp 259ndash270 2005
[13] K Imagawa S Okayama M Takaoka et al ldquoInhibitory effectof efonidipine on aldosterone synthesis and secretion in humanadrenocarcinoma (H295R) cellsrdquo Journal of CardiovascularPharmacology vol 47 no 1 pp 133ndash138 2006
[14] M Voets UMuller-Vieira S Marchais-Oberwinkler and RWHartmann ldquoSynthesis of amidinohydrazones and evaluation oftheir inhibitory effect towards aldosterone synthase (CYP11B2)and the formation of selected steroidsrdquo Archiv der Pharmazievol 337 no 7 pp 411ndash416 2004
[15] F K Nielsen C H Hansen J A Fey et al ldquoH295R cells as amodel for steroidogenic disruption a broader perspective usingsimultaneous chemical analysis of 7 key steroid hormonesrdquoToxicology in Vitro vol 26 no 2 pp 343ndash350 2012
[16] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cortisolproduction and on activities of steroidogenic enzymes inhuman adrenocortical H295R cellsrdquo Journal of Steroid Biochem-istry and Molecular Biology vol 80 no 3 pp 355ndash363 2002
[17] R F Canton J T Sanderson S Nijmeijer A Bergman R JLetcher and M van den Berg ldquoIn vitro effects of brominatedflame retardants and metabolites on CYP17 catalytic activity anovel mechanism of actionrdquo Toxicology and Applied Pharma-cology vol 216 no 2 pp 274ndash281 2006
[18] K Hilscherova P D Jones T Gracia et al ldquoAssessment ofthe effects of chemicals on the expression of ten steroidogenicgenes in the H295R cell line using real-time PCRrdquo ToxicologicalSciences vol 81 no 1 pp 78ndash89 2004
[19] T Gracia K Hilscherova P D Jones et al ldquoThe H295R systemfor evaluation of endocrine-disrupting effectsrdquo Ecotoxicologyand Environmental Safety vol 65 no 3 pp 293ndash305 2006
[20] X Zhang J L NewstedM Hecker E B Higley P D Jones andJ P Giesy ldquoClassification of chemicals based on concentration-dependent toxicological data using ToxClustrdquo EnvironmentalScience and Technology vol 43 no 10 pp 3926ndash3932 2009
[21] M S Breen M Breen N Terasaki M Yamazaki and R BConolly ldquoComputational model of steroidogenesis in humanH295R cells to predict biochemical response to endocrine-active chemicalsmodel development formetyraponerdquoEnviron-mental Health Perspectives vol 118 no 2 pp 265ndash272 2010
[22] M Breen M S Breen N Terasaki M Yamazaki A LLloyd and R B Conolly ldquoMechanistic computational model ofsteroidogenesis in H295R cells role of oxysterols and cell pro-liferation to improve predictability of biochemical response toendocrine active chemical-metyraponerdquo Toxicological Sciencesvol 123 no 1 pp 80ndash93 2011
[23] F B Kraemer ldquoAdrenal cholesterol utilizationrdquo Molecular andCellular Endocrinology vol 265-266 pp 42ndash45 2007
[24] W L Miller ldquoSteroidogenic acute regulatory protein (StAR)a novel mitochondrial cholesterol transporterrdquo Biochimica et
18 Journal of Toxicology
Biophysica Acta (BBA)mdashMolecular and Cell Biology of Lipidsvol 1771 no 6 pp 663ndash676 2007
[25] S Szabo and I T Lippe ldquoAdrenal gland chemically inducedstructural and functional changes in the cortexrdquo ToxicologicPathology vol 17 no 2 pp 317ndash329 1989
[26] C Chen-Pan I-J Pan Y Yamamoto T Sakogawa J Yamadaand YHayashi ldquoPrompt recovery of damaged adrenalmedullaeinduced by salinomycinrdquoToxicologic Pathology vol 27 no 5 pp563ndash572 1999
[27] R J Santen and R I Misbin ldquoAminoglutethimide review ofpharmacology and clinical userdquo Pharmacotherapy vol 1 no 2pp 95ndash120 1981
[28] M J Carella N V Dimitrov V V Gossain L Srivastava andD R Rovner ldquoAdrenal effects of low-dose aminoglutethimidewhen used alone in postmenopausal women with advancedbreast cancerrdquoMetabolism vol 43 no 6 pp 723ndash727 1994
[29] M K Johansson J T Sanderson and B-O Lund ldquoEffects of3-MeSO2-DDE and some CYP inhibitors on glucocorticoidsteroidogenesis in the H295R human adrenocortical carcinomacell linerdquo Toxicology in Vitro vol 16 no 2 pp 113ndash121 2002
[30] Y Touitou A Bogdan J C Legrand and P Desgrez ldquoAmino-glutethimide and glutethimide effects on 18-hydroxycorti-costerone biosynthesis by human and sheep adrenals in vitrordquoActa Endocrinologica vol 80 no 3 pp 517ndash526 1975
[31] M M Hart and J A Straw ldquoStudies on the site of action ofop1015840-DDD in the dog adrenal cortex 1 Inhibition of ACTH-mediated pregnenolone synthesisrdquo Steroids vol 17 no 1ndash5 pp559ndash574 1971
[32] M Ojima M Saitoh N Itoh Y Kusano S Fukuchi and HNaganuma ldquoThe effects of op1015840-DDD on adrenal steroidogen-esis and hepatic steroid metabolismrdquo Nippon Naibunpi Gakkaizasshi vol 61 no 3 pp 168ndash178 1985
[33] L F Canosa and N R Ceballos ldquoEffects of different steroid-biosynthesis inhibitors on the testicular steroidogenesis of thetoad Bufo arenarumrdquo Journal of Comparative Physiology B vol171 no 6 pp 519ndash526 2001
[34] D C Kossor S Kominami S Takemori and H D ColbyldquoRole of the steroid 17 alpha-hydroxylase in spironolactone-mediated destruction of adrenal cytochrome P-450rdquoMolecularPharmacology vol 40 no 2 pp 321ndash325 1991
[35] S C Cheng K Suzuki W Sadee and B W Harding ldquoEffectsof spironolactone canrenone and canrenoate K on cytochromeP450 and 11120573 and 18 hydroxylation in bovine and humanadrenal corticalmitochondriardquo Endocrinology vol 99 no 4 pp1097ndash1106 1976
[36] K Yanagibashi M Haniu J E Shively W H Shen and P HallldquoThe synthesis of aldosterone by the adrenal cortex Two zones(fasciculata and glomerulosa) possess one enzyme for 11120573- 18-hydroxylation and aldehyde synthesisrdquoThe Journal of BiologicalChemistry vol 261 no 8 pp 3556ndash3562 1986
[37] K Morishita H Okumura N Ito and N Takahashi ldquoPrimaryculture system of adrenocortical cells from dogs to evaluatedirect effects of chemicals on steroidogenesisrdquo Toxicology vol165 no 2-3 pp 171ndash178 2001
[38] S W J Lamberts E G Bons H A Bruining and F H de JongldquoDifferential effects of the imidazole derivatives etomidateketoconazole and miconazole and of metyrapone on the secre-tion of cortisol and its precursors by human adrenocorticalcellsrdquo Journal of Pharmacology and Experimental Therapeuticsvol 240 no 1 pp 259ndash264 1987
[39] L Yin S Lucas F Maurer U Kazmaier Q Hu and RW Hartmann ldquoNovel imidazol-1-ylmethyl substituted 1256-tetrahydropyrrolo[321-ij]quinolin-4-ones as potent and selec-tive CYP11B1 inhibitors for the treatment of Cushingrsquos syn-dromerdquo Journal ofMedicinal Chemistry vol 55 no 14 pp 6629ndash6633 2012
[40] M DiMattina N Maronian H Ashby D L Loriaux andB D Albertson ldquoKetoconazole inhibits multiple steroidogenicenzymes involved in androgen biosynthesis in the humanovaryrdquo Fertility and Sterility vol 49 no 1 pp 62ndash65 1988
[41] M Ayub and M J Levell ldquoInhibition of human adrenal steroi-dogenic enzymes in vitro by imidazole drugs including keto-conazolerdquo Journal of Steroid Biochemistry vol 32 no 4 pp 515ndash524 1989
[42] D Engelhardt M M Weber T Miksch F Abedinpour andC Jaspers ldquoThe influence of ketoconazole on human adrenalsteroidogenesis incubation studies with tissue slicesrdquo ClinicalEndocrinology vol 35 no 2 pp 163ndash168 1991
[43] S Bhasin S Sikka T Fielder et al ldquoHormonal effects of keto-conazole in vivo in the male rat mechanism of actionrdquo Endo-crinology vol 118 no 3 pp 1229ndash1232 1986
[44] D M Zisterer and D C Williams ldquoCalmidazolium and otherimidazole compounds affect steroidogenesis in Y1 cells lack ofinvolvement of the peripheral-type benzodiazepine receptorrdquoJournal of Steroid Biochemistry and Molecular Biology vol 60no 3-4 pp 189ndash195 1997
[45] L P Walsh C N Kuratko and D M Stocco ldquoEconazoleand miconazole inhibit steroidogenesis and disrupt steroido-genic acute regulatory (StAR) protein expression post-trans-criptionallyrdquo Journal of Steroid Biochemistry and Molecular Bio-logy vol 75 no 4-5 pp 229ndash236 2000
[46] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cor-tisol production and on activities of steroidogenic enzymesin human adrenocortical H295R cellsrdquo The Journal of SteroidBiochemistry and Molecular Biology vol 80 no 3 pp 355ndash3632002
[47] D P Dempsher D S Gann and R D Phair ldquoA mechanisticmodel of ACTH-stimulated cortisol secretionrdquo American Jour-nal of Physiology vol 246 no 4 part 2 pp R587ndashR596 1984
[48] N Sarai S Matsuoka and A Noma ldquosimBio a Java package forthe development of detailed cell modelsrdquo Progress in Biophysicsand Molecular Biology vol 90 no 1ndash3 pp 360ndash377 2006
[49] K Levenberg ldquoA method for the solution of certain non-linearproblems in least squaresrdquo Quarterly of Applied Mathematicsvol 2 pp 164ndash168 1944
[50] D W Marquardt ldquoAn algorithm for least-squares estimation ofnonlinear parametersrdquo SIAM Journal on Applied Mathematicsvol 11 no 2 pp 431ndash441 1963
[51] P E Gill and W Murray ldquoAlgorithms for the solution of thenonlinear least-squares problemrdquo SIAM Journal on NumericalAnalysis vol 15 no 5 pp 977ndash992 1978
[52] Y Akimoto R Hasada J Sakuma I Ono and S KobayashildquoGeneration alternation model for real-coded GA using multi-parent proposal and evaluation of just generation gap (JGG)rdquoin Proceedings of the 19th SICE Symposium on DecentralizedAutonomous Systems pp 341ndash346 Tokyo Japan January 2007
[53] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoProposal and evaluation of adaptive real-coded crossoverAREXrdquo Transactions of the Japanese Society for Artificial Intel-ligence vol 24 no 6 pp 446ndash458 2009
Journal of Toxicology 19
[54] S Kobayashi ldquoThe frontiers of real-coded genetic algorithmsrdquoJournal of Japanese Society for Artificial Intelligence vol 24 no1 pp 147ndash162 2009
[55] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoAnalysis of the behavior of MGG and JGG as a selection modelfor real-coded genetic algorithmsrdquo Transactions of the JapaneseSociety for Artificial Intelligence vol 25 no 2 pp 281ndash289 2010
[56] A Struthers H Krum and G H Williams ldquoA comparisonof the aldosterone-blocking agents eplerenone and spironolac-tonerdquo Clinical Cardiology vol 31 no 4 pp 153ndash158 2008
[57] K Yamasaki M Sawaki S Noda et al ldquoComparison of theHershberger assay and androgen receptor binding assay oftwelve chemicalsrdquo Toxicology vol 195 no 2-3 pp 177ndash1862004
[58] C Eil and S K Edelson ldquoThe use of human skin fibroblasts toobtain potency estimates of drug binding to androgen recept-orsrdquo The Journal of Clinical Endocrinology amp Metabolism vol59 no 1 pp 51ndash55 1984
[59] P Ye T Yamashita D M Pollock H Sasano and W ERainey ldquoContrasting effects of eplerenone and spironolac-tone on adrenal cell steroidogenesisrdquo Hormone and MetabolicResearch vol 41 no 1 pp 35ndash39 2009
[60] B D Albertson K L Frederick N C Maronian et al ldquoTheeffect of ketoconazole on steroidogenesis I Leydig cell enzymeactivity in vitrordquo Research Communications in Chemical Patho-logy and Pharmacology vol 61 no 1 pp 17ndash26 1988
[61] B D Albertson N C Maronian K L Frederick et al ldquoTheeffect of ketoconazole on steroidogenesis II Adrenocorticalenzyme activity in vitrordquoResearch Communications in ChemicalPathology and Pharmacology vol 61 no 1 pp 27ndash34 1988
[62] DM Stocco andB J Clark ldquoRegulation of the acute productionof steroids in steroidogenic cellsrdquo Endocrine Reviews vol 17 no3 pp 221ndash244 1996
[63] L K Christenson and J F Strauss III ldquoSteroidogenic acute regu-latory protein (StAR) and the intramitochondrial translocationof cholesterolrdquo Biochimica et Biophysica Acta (BBA)mdashMolecularand Cell Biology of Lipids vol 1529 no 1ndash3 pp 175ndash187 2000
[64] E R Simpson ldquoCholesterol side-chain cleavage cytochromeP450 and the control of steroidogenesisrdquo Molecular and Cel-lular Endocrinology vol 13 no 3 pp 213ndash227 1979
[65] H A LaVoie and S R King ldquoTranscriptional regulation of ster-oidogenic genes STARD1 CYP11A1 and HSD3Brdquo ExperimentalBiology and Medicine vol 234 no 8 pp 880ndash907 2009
[66] F Raucci A DrsquoAniello and M M Di Fiore ldquoStimulation ofandrogen production by d-aspartate through the enhancementof StAR P450scc and 3120573-HSD mRNA levels in vivo rat testisand in culture of immature rat Leydig cellsrdquo Steroids vol 84pp 103ndash110 2014
[67] M Doi Y Takahashi R Komatsu et al ldquoSalt-sensitive hyper-tension in circadian clock-deficient Cry-null mice involves dys-regulated adrenal Hsd3b6rdquo Nature Medicine vol 16 no 1 pp67ndash74 2010
[68] H Okamura M Doi Y Yamaguchi and J-M Fustin ldquoHyper-tension due to loss of clock novel insight from the molecu-lar analysis of Cry1Cry2-deleted micerdquo Current HypertensionReports vol 13 no 2 pp 103ndash108 2011
[69] M Doi F Satoh T Maekawa et al ldquoIsoform-specific mono-clonal antibodies against 3120573-hydroxysteroid dehydrogenaseisomerase family provide markers for subclassification ofhuman primary aldosteronismrdquo The Journal of Clinical Endo-crinology amp Metabolism vol 99 no 2 pp E257ndashE262 2014
[70] S Konosu-Fukaya Y Nakamura F Satoh et al ldquo3120573-hydroxy-steroid dehydrogenase isoforms in human aldosterone-produc-ing adenomardquo Molecular and Cellular Endocrinology vol 408pp 205ndash212 2015
[71] G C Verwoert J Hofland N Amin et al ldquoExpression and genevariation studies deny association of humanHSD3B1 gene withaldosterone production or blood pressurerdquoAmerican Journal ofHypertension vol 28 no 1 pp 113ndash120 2015
[72] D Hough K Storbeck SW P Cloete A C Swart and P SwartldquoRelative contribution of P450c17 towards the acute cortisolresponse lessons from sheep and goatsrdquoMolecular and CellularEndocrinology vol 408 pp 107ndash113 2015
[73] I Sakuma S Suematsu Y Matsuzawa et al ldquoCharacterizationof steroidogenic enzyme expression in aldosterone-producingadenoma a comparison with various human adrenal tumorsrdquoEndocrine Journal vol 60 no 3 pp 329ndash336 2013
[74] A Tong A Jia S Yan Y Zhang Y Xie and G Liu ldquoEctopiccortisol-producing adrenocortical adenoma in the renal hilumhistopathological features and steroidogenic enzyme profilerdquoInternational Journal of Clinical and Experimental Pathologyvol 7 no 7 pp 4415ndash4421 2014
[75] A Hirsch D Hahn P Kempna et al ldquoMetformin inhibitshuman androgen production by regulating steroidogenicenzymes HSD3B2 and CYP17A1 and complex I activity of therespiratory chainrdquo Endocrinology vol 153 no 9 pp 4354ndash43662012
[76] B Xu L Gao Y Cui et al ldquoSET protein up-regulated test-osterone production in the cultured preantral folliclesrdquo Repro-ductive Biology and Endocrinology vol 11 article 9 2013
[77] S Kikuchi D Tominaga M Arita M Takahashi and MTomita ldquoDynamic modeling of genetic networks using geneticalgorithm and S-systemrdquo Bioinformatics vol 19 no 5 pp 643ndash650 2003
[78] N Shikata YMaki M Nakatsui et al ldquoDetermining importantregulatory relations of amino acids from dynamic networkanalysis of plasma amino acidsrdquo Amino Acids vol 38 no 1 pp179ndash187 2010
[79] M Nakatsui T Ueda Y Maki I Ono and M OkamotoldquoMethod for inferring and extracting reliable genetic interac-tions from time-series profile of gene expressionrdquoMathematicalBiosciences vol 215 no 1 pp 105ndash114 2008
[80] A Komori Y Maki M Nakatsui I Ono and M OkamotoldquoEfficient numerical optimization algorithm based on new real-coded genetic algorithm AREX + JGG and application to theinverse problem in systems biologyrdquo Applied Mathematics vol3 no 10 pp 1463ndash1470 2012
[81] J A Roubos G van Straten and A J B van Boxtel ldquoAn evolu-tionary strategy for fed-batch bioreactor optimization conceptsand performancerdquo Journal of Biotechnology vol 67 no 2-3 pp173ndash187 1999
[82] D Sarkar and J M Modak ldquoGenetic algorithms with filters foroptimal control problems in fed-batch bioreactorsrdquo Bioprocessand Biosystems Engineering vol 26 no 5 pp 295ndash306 2004
[83] K Harada K Ikeda J Sakuma I Ono and S KobayashildquoHybridization of genetic algorithm with local search in mul-tiobjective function optimization recommendation of GA thenLSrdquoTransactions of the Japanese Society for Artificial Intelligencevol 21 no 6 pp 482ndash492 2006
Submit your manuscripts athttpwwwhindawicom
PainResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
ToxinsJournal of
VaccinesJournal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AntibioticsInternational Journal of
ToxicologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
StrokeResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Drug DeliveryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Pharmacological Sciences
Tropical MedicineJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AddictionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Emergency Medicine InternationalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Autoimmune Diseases
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anesthesiology Research and Practice
ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Pharmaceutics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
MEDIATORSINFLAMMATION
of
Journal of Toxicology 17
Authorsrsquo Contribution
Ryuta Saito Natsuko Terasaki and Naohisa Tsutsui designedthe study Naohisa Tsutsui and Naoya Masutomi supervisedthe experiments Natsuko Terasaki performed all in vitroexperiments and collected dataMakotoYamazaki performedall measurements of steroids Ryuta Saito conceived the insilico strategies analyzed data developed all programs andthe mathematical model and performed all simulation anal-yses with Masahiro Okamoto Ryuta Saito Natsuko TerasakiMakoto Yamazaki NaoyaMasutomi andMasahiroOkamotowrote the paper
Acknowledgments
The authors would like to thank Dr Michael S Breenand Dr Miyuki Breen for helpful discussions concerningprevious mathematical models of adrenal steroidogenesisThis work was partially supported by the Grant-in-Aid forScientific Research on Innovative Areas ldquoSynthetic Biologyrdquo(No 23119001 (MO)) from Ministry of Education CultureSports Science and Technology in Japan
References
[1] P W Harvey D J Everett and C J Springall ldquoAdrenal toxi-cology a strategy for assessment of functional toxicity to theadrenal cortex and steroidogenesisrdquo Journal of Applied Toxi-cology vol 27 no 2 pp 103ndash115 2007
[2] W E Ribelin ldquoThe effects of drugs and chemicals upon thestructure of the adrenal glandrdquo Fundamental and Applied Toxi-cology vol 4 no 1 pp 105ndash119 1984
[3] H D Colby and P A Longhurst ldquoToxicology of the adrenalglandrdquo in Endocrine Toxicology C K Atterwill and J D FlackEds pp 243ndash281 CambridgeUniversity Press Cambridge UK1992
[4] H D Colby ldquoIn vitro assessment of adrenocortical toxicityrdquoJournal of Pharmacological and Toxicological Methods vol 32no 1 pp 1ndash6 1994
[5] T J Rosol J T Yarrington J Latendresse and C C CapenldquoAdrenal gland structure function and mechanisms of toxic-ityrdquo Toxicologic Pathology vol 29 no 1 pp 41ndash48 2001
[6] P W Harvey and D J Everett ldquoThe adrenal cortex and ster-oidogenesis as cellular and molecular targets for toxicity crit-ical omissions from regulatory endocrine disrupter screeningstrategies for human healthrdquo Journal of Applied Toxicology vol23 no 2 pp 81ndash87 2003
[7] A F Gazdar H K Oie C H Shackleton et al ldquoEstablishmentand characterization of a human adrenocortical carcinoma cellline that expresses multiple pathways of steroid biosynthesisrdquoCancer Research vol 50 no 17 pp 5488ndash5496 1990
[8] W E Rainey I M Bird C Sawetawan et al ldquoRegulation ofhuman adrenal carcinoma cell (NCI-H295) production of C19steroidsrdquo The Journal of Clinical Endocrinology amp Metabolismvol 77 no 3 pp 731ndash737 1993
[9] B Staels D W Hum andW L Miller ldquoRegulation of steroido-genesis in NCI-H295 cells a cellular model of the human fetaladrenalrdquo Molecular Endocrinology vol 7 no 3 pp 423ndash4331993
[10] M Hecker J L Newsted M B Murphy et al ldquoHuman adreno-carcinoma (H295R) cells for rapid in vitro determination of
effects on steroidogenesis hormone productionrdquo Toxicologyand Applied Pharmacology vol 217 no 1 pp 114ndash124 2006
[11] A Oskarsson E Ulleras K E Plant J P Hinson and P SGoldfarb ldquoSteroidogenic gene expression in H295R cells andthe human adrenal gland adrenotoxic effects of lindane invitrordquo Journal of Applied Toxicology vol 26 no 6 pp 484ndash4922006
[12] UMuller-Vieira M Angotti and RW Hartman ldquoThe adreno-cortical tumor cell line NCI-H295R as an in vitro screeningsystem for the evaluation of CYP11B2 (aldosterone synthase)and CYP11B1 (steroid-11120573-hydroxylase) inhibitorsrdquoThe Journalof Steroid Biochemistry and Molecular Biology vol 96 no 3-4pp 259ndash270 2005
[13] K Imagawa S Okayama M Takaoka et al ldquoInhibitory effectof efonidipine on aldosterone synthesis and secretion in humanadrenocarcinoma (H295R) cellsrdquo Journal of CardiovascularPharmacology vol 47 no 1 pp 133ndash138 2006
[14] M Voets UMuller-Vieira S Marchais-Oberwinkler and RWHartmann ldquoSynthesis of amidinohydrazones and evaluation oftheir inhibitory effect towards aldosterone synthase (CYP11B2)and the formation of selected steroidsrdquo Archiv der Pharmazievol 337 no 7 pp 411ndash416 2004
[15] F K Nielsen C H Hansen J A Fey et al ldquoH295R cells as amodel for steroidogenic disruption a broader perspective usingsimultaneous chemical analysis of 7 key steroid hormonesrdquoToxicology in Vitro vol 26 no 2 pp 343ndash350 2012
[16] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cortisolproduction and on activities of steroidogenic enzymes inhuman adrenocortical H295R cellsrdquo Journal of Steroid Biochem-istry and Molecular Biology vol 80 no 3 pp 355ndash363 2002
[17] R F Canton J T Sanderson S Nijmeijer A Bergman R JLetcher and M van den Berg ldquoIn vitro effects of brominatedflame retardants and metabolites on CYP17 catalytic activity anovel mechanism of actionrdquo Toxicology and Applied Pharma-cology vol 216 no 2 pp 274ndash281 2006
[18] K Hilscherova P D Jones T Gracia et al ldquoAssessment ofthe effects of chemicals on the expression of ten steroidogenicgenes in the H295R cell line using real-time PCRrdquo ToxicologicalSciences vol 81 no 1 pp 78ndash89 2004
[19] T Gracia K Hilscherova P D Jones et al ldquoThe H295R systemfor evaluation of endocrine-disrupting effectsrdquo Ecotoxicologyand Environmental Safety vol 65 no 3 pp 293ndash305 2006
[20] X Zhang J L NewstedM Hecker E B Higley P D Jones andJ P Giesy ldquoClassification of chemicals based on concentration-dependent toxicological data using ToxClustrdquo EnvironmentalScience and Technology vol 43 no 10 pp 3926ndash3932 2009
[21] M S Breen M Breen N Terasaki M Yamazaki and R BConolly ldquoComputational model of steroidogenesis in humanH295R cells to predict biochemical response to endocrine-active chemicalsmodel development formetyraponerdquoEnviron-mental Health Perspectives vol 118 no 2 pp 265ndash272 2010
[22] M Breen M S Breen N Terasaki M Yamazaki A LLloyd and R B Conolly ldquoMechanistic computational model ofsteroidogenesis in H295R cells role of oxysterols and cell pro-liferation to improve predictability of biochemical response toendocrine active chemical-metyraponerdquo Toxicological Sciencesvol 123 no 1 pp 80ndash93 2011
[23] F B Kraemer ldquoAdrenal cholesterol utilizationrdquo Molecular andCellular Endocrinology vol 265-266 pp 42ndash45 2007
[24] W L Miller ldquoSteroidogenic acute regulatory protein (StAR)a novel mitochondrial cholesterol transporterrdquo Biochimica et
18 Journal of Toxicology
Biophysica Acta (BBA)mdashMolecular and Cell Biology of Lipidsvol 1771 no 6 pp 663ndash676 2007
[25] S Szabo and I T Lippe ldquoAdrenal gland chemically inducedstructural and functional changes in the cortexrdquo ToxicologicPathology vol 17 no 2 pp 317ndash329 1989
[26] C Chen-Pan I-J Pan Y Yamamoto T Sakogawa J Yamadaand YHayashi ldquoPrompt recovery of damaged adrenalmedullaeinduced by salinomycinrdquoToxicologic Pathology vol 27 no 5 pp563ndash572 1999
[27] R J Santen and R I Misbin ldquoAminoglutethimide review ofpharmacology and clinical userdquo Pharmacotherapy vol 1 no 2pp 95ndash120 1981
[28] M J Carella N V Dimitrov V V Gossain L Srivastava andD R Rovner ldquoAdrenal effects of low-dose aminoglutethimidewhen used alone in postmenopausal women with advancedbreast cancerrdquoMetabolism vol 43 no 6 pp 723ndash727 1994
[29] M K Johansson J T Sanderson and B-O Lund ldquoEffects of3-MeSO2-DDE and some CYP inhibitors on glucocorticoidsteroidogenesis in the H295R human adrenocortical carcinomacell linerdquo Toxicology in Vitro vol 16 no 2 pp 113ndash121 2002
[30] Y Touitou A Bogdan J C Legrand and P Desgrez ldquoAmino-glutethimide and glutethimide effects on 18-hydroxycorti-costerone biosynthesis by human and sheep adrenals in vitrordquoActa Endocrinologica vol 80 no 3 pp 517ndash526 1975
[31] M M Hart and J A Straw ldquoStudies on the site of action ofop1015840-DDD in the dog adrenal cortex 1 Inhibition of ACTH-mediated pregnenolone synthesisrdquo Steroids vol 17 no 1ndash5 pp559ndash574 1971
[32] M Ojima M Saitoh N Itoh Y Kusano S Fukuchi and HNaganuma ldquoThe effects of op1015840-DDD on adrenal steroidogen-esis and hepatic steroid metabolismrdquo Nippon Naibunpi Gakkaizasshi vol 61 no 3 pp 168ndash178 1985
[33] L F Canosa and N R Ceballos ldquoEffects of different steroid-biosynthesis inhibitors on the testicular steroidogenesis of thetoad Bufo arenarumrdquo Journal of Comparative Physiology B vol171 no 6 pp 519ndash526 2001
[34] D C Kossor S Kominami S Takemori and H D ColbyldquoRole of the steroid 17 alpha-hydroxylase in spironolactone-mediated destruction of adrenal cytochrome P-450rdquoMolecularPharmacology vol 40 no 2 pp 321ndash325 1991
[35] S C Cheng K Suzuki W Sadee and B W Harding ldquoEffectsof spironolactone canrenone and canrenoate K on cytochromeP450 and 11120573 and 18 hydroxylation in bovine and humanadrenal corticalmitochondriardquo Endocrinology vol 99 no 4 pp1097ndash1106 1976
[36] K Yanagibashi M Haniu J E Shively W H Shen and P HallldquoThe synthesis of aldosterone by the adrenal cortex Two zones(fasciculata and glomerulosa) possess one enzyme for 11120573- 18-hydroxylation and aldehyde synthesisrdquoThe Journal of BiologicalChemistry vol 261 no 8 pp 3556ndash3562 1986
[37] K Morishita H Okumura N Ito and N Takahashi ldquoPrimaryculture system of adrenocortical cells from dogs to evaluatedirect effects of chemicals on steroidogenesisrdquo Toxicology vol165 no 2-3 pp 171ndash178 2001
[38] S W J Lamberts E G Bons H A Bruining and F H de JongldquoDifferential effects of the imidazole derivatives etomidateketoconazole and miconazole and of metyrapone on the secre-tion of cortisol and its precursors by human adrenocorticalcellsrdquo Journal of Pharmacology and Experimental Therapeuticsvol 240 no 1 pp 259ndash264 1987
[39] L Yin S Lucas F Maurer U Kazmaier Q Hu and RW Hartmann ldquoNovel imidazol-1-ylmethyl substituted 1256-tetrahydropyrrolo[321-ij]quinolin-4-ones as potent and selec-tive CYP11B1 inhibitors for the treatment of Cushingrsquos syn-dromerdquo Journal ofMedicinal Chemistry vol 55 no 14 pp 6629ndash6633 2012
[40] M DiMattina N Maronian H Ashby D L Loriaux andB D Albertson ldquoKetoconazole inhibits multiple steroidogenicenzymes involved in androgen biosynthesis in the humanovaryrdquo Fertility and Sterility vol 49 no 1 pp 62ndash65 1988
[41] M Ayub and M J Levell ldquoInhibition of human adrenal steroi-dogenic enzymes in vitro by imidazole drugs including keto-conazolerdquo Journal of Steroid Biochemistry vol 32 no 4 pp 515ndash524 1989
[42] D Engelhardt M M Weber T Miksch F Abedinpour andC Jaspers ldquoThe influence of ketoconazole on human adrenalsteroidogenesis incubation studies with tissue slicesrdquo ClinicalEndocrinology vol 35 no 2 pp 163ndash168 1991
[43] S Bhasin S Sikka T Fielder et al ldquoHormonal effects of keto-conazole in vivo in the male rat mechanism of actionrdquo Endo-crinology vol 118 no 3 pp 1229ndash1232 1986
[44] D M Zisterer and D C Williams ldquoCalmidazolium and otherimidazole compounds affect steroidogenesis in Y1 cells lack ofinvolvement of the peripheral-type benzodiazepine receptorrdquoJournal of Steroid Biochemistry and Molecular Biology vol 60no 3-4 pp 189ndash195 1997
[45] L P Walsh C N Kuratko and D M Stocco ldquoEconazoleand miconazole inhibit steroidogenesis and disrupt steroido-genic acute regulatory (StAR) protein expression post-trans-criptionallyrdquo Journal of Steroid Biochemistry and Molecular Bio-logy vol 75 no 4-5 pp 229ndash236 2000
[46] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cor-tisol production and on activities of steroidogenic enzymesin human adrenocortical H295R cellsrdquo The Journal of SteroidBiochemistry and Molecular Biology vol 80 no 3 pp 355ndash3632002
[47] D P Dempsher D S Gann and R D Phair ldquoA mechanisticmodel of ACTH-stimulated cortisol secretionrdquo American Jour-nal of Physiology vol 246 no 4 part 2 pp R587ndashR596 1984
[48] N Sarai S Matsuoka and A Noma ldquosimBio a Java package forthe development of detailed cell modelsrdquo Progress in Biophysicsand Molecular Biology vol 90 no 1ndash3 pp 360ndash377 2006
[49] K Levenberg ldquoA method for the solution of certain non-linearproblems in least squaresrdquo Quarterly of Applied Mathematicsvol 2 pp 164ndash168 1944
[50] D W Marquardt ldquoAn algorithm for least-squares estimation ofnonlinear parametersrdquo SIAM Journal on Applied Mathematicsvol 11 no 2 pp 431ndash441 1963
[51] P E Gill and W Murray ldquoAlgorithms for the solution of thenonlinear least-squares problemrdquo SIAM Journal on NumericalAnalysis vol 15 no 5 pp 977ndash992 1978
[52] Y Akimoto R Hasada J Sakuma I Ono and S KobayashildquoGeneration alternation model for real-coded GA using multi-parent proposal and evaluation of just generation gap (JGG)rdquoin Proceedings of the 19th SICE Symposium on DecentralizedAutonomous Systems pp 341ndash346 Tokyo Japan January 2007
[53] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoProposal and evaluation of adaptive real-coded crossoverAREXrdquo Transactions of the Japanese Society for Artificial Intel-ligence vol 24 no 6 pp 446ndash458 2009
Journal of Toxicology 19
[54] S Kobayashi ldquoThe frontiers of real-coded genetic algorithmsrdquoJournal of Japanese Society for Artificial Intelligence vol 24 no1 pp 147ndash162 2009
[55] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoAnalysis of the behavior of MGG and JGG as a selection modelfor real-coded genetic algorithmsrdquo Transactions of the JapaneseSociety for Artificial Intelligence vol 25 no 2 pp 281ndash289 2010
[56] A Struthers H Krum and G H Williams ldquoA comparisonof the aldosterone-blocking agents eplerenone and spironolac-tonerdquo Clinical Cardiology vol 31 no 4 pp 153ndash158 2008
[57] K Yamasaki M Sawaki S Noda et al ldquoComparison of theHershberger assay and androgen receptor binding assay oftwelve chemicalsrdquo Toxicology vol 195 no 2-3 pp 177ndash1862004
[58] C Eil and S K Edelson ldquoThe use of human skin fibroblasts toobtain potency estimates of drug binding to androgen recept-orsrdquo The Journal of Clinical Endocrinology amp Metabolism vol59 no 1 pp 51ndash55 1984
[59] P Ye T Yamashita D M Pollock H Sasano and W ERainey ldquoContrasting effects of eplerenone and spironolac-tone on adrenal cell steroidogenesisrdquo Hormone and MetabolicResearch vol 41 no 1 pp 35ndash39 2009
[60] B D Albertson K L Frederick N C Maronian et al ldquoTheeffect of ketoconazole on steroidogenesis I Leydig cell enzymeactivity in vitrordquo Research Communications in Chemical Patho-logy and Pharmacology vol 61 no 1 pp 17ndash26 1988
[61] B D Albertson N C Maronian K L Frederick et al ldquoTheeffect of ketoconazole on steroidogenesis II Adrenocorticalenzyme activity in vitrordquoResearch Communications in ChemicalPathology and Pharmacology vol 61 no 1 pp 27ndash34 1988
[62] DM Stocco andB J Clark ldquoRegulation of the acute productionof steroids in steroidogenic cellsrdquo Endocrine Reviews vol 17 no3 pp 221ndash244 1996
[63] L K Christenson and J F Strauss III ldquoSteroidogenic acute regu-latory protein (StAR) and the intramitochondrial translocationof cholesterolrdquo Biochimica et Biophysica Acta (BBA)mdashMolecularand Cell Biology of Lipids vol 1529 no 1ndash3 pp 175ndash187 2000
[64] E R Simpson ldquoCholesterol side-chain cleavage cytochromeP450 and the control of steroidogenesisrdquo Molecular and Cel-lular Endocrinology vol 13 no 3 pp 213ndash227 1979
[65] H A LaVoie and S R King ldquoTranscriptional regulation of ster-oidogenic genes STARD1 CYP11A1 and HSD3Brdquo ExperimentalBiology and Medicine vol 234 no 8 pp 880ndash907 2009
[66] F Raucci A DrsquoAniello and M M Di Fiore ldquoStimulation ofandrogen production by d-aspartate through the enhancementof StAR P450scc and 3120573-HSD mRNA levels in vivo rat testisand in culture of immature rat Leydig cellsrdquo Steroids vol 84pp 103ndash110 2014
[67] M Doi Y Takahashi R Komatsu et al ldquoSalt-sensitive hyper-tension in circadian clock-deficient Cry-null mice involves dys-regulated adrenal Hsd3b6rdquo Nature Medicine vol 16 no 1 pp67ndash74 2010
[68] H Okamura M Doi Y Yamaguchi and J-M Fustin ldquoHyper-tension due to loss of clock novel insight from the molecu-lar analysis of Cry1Cry2-deleted micerdquo Current HypertensionReports vol 13 no 2 pp 103ndash108 2011
[69] M Doi F Satoh T Maekawa et al ldquoIsoform-specific mono-clonal antibodies against 3120573-hydroxysteroid dehydrogenaseisomerase family provide markers for subclassification ofhuman primary aldosteronismrdquo The Journal of Clinical Endo-crinology amp Metabolism vol 99 no 2 pp E257ndashE262 2014
[70] S Konosu-Fukaya Y Nakamura F Satoh et al ldquo3120573-hydroxy-steroid dehydrogenase isoforms in human aldosterone-produc-ing adenomardquo Molecular and Cellular Endocrinology vol 408pp 205ndash212 2015
[71] G C Verwoert J Hofland N Amin et al ldquoExpression and genevariation studies deny association of humanHSD3B1 gene withaldosterone production or blood pressurerdquoAmerican Journal ofHypertension vol 28 no 1 pp 113ndash120 2015
[72] D Hough K Storbeck SW P Cloete A C Swart and P SwartldquoRelative contribution of P450c17 towards the acute cortisolresponse lessons from sheep and goatsrdquoMolecular and CellularEndocrinology vol 408 pp 107ndash113 2015
[73] I Sakuma S Suematsu Y Matsuzawa et al ldquoCharacterizationof steroidogenic enzyme expression in aldosterone-producingadenoma a comparison with various human adrenal tumorsrdquoEndocrine Journal vol 60 no 3 pp 329ndash336 2013
[74] A Tong A Jia S Yan Y Zhang Y Xie and G Liu ldquoEctopiccortisol-producing adrenocortical adenoma in the renal hilumhistopathological features and steroidogenic enzyme profilerdquoInternational Journal of Clinical and Experimental Pathologyvol 7 no 7 pp 4415ndash4421 2014
[75] A Hirsch D Hahn P Kempna et al ldquoMetformin inhibitshuman androgen production by regulating steroidogenicenzymes HSD3B2 and CYP17A1 and complex I activity of therespiratory chainrdquo Endocrinology vol 153 no 9 pp 4354ndash43662012
[76] B Xu L Gao Y Cui et al ldquoSET protein up-regulated test-osterone production in the cultured preantral folliclesrdquo Repro-ductive Biology and Endocrinology vol 11 article 9 2013
[77] S Kikuchi D Tominaga M Arita M Takahashi and MTomita ldquoDynamic modeling of genetic networks using geneticalgorithm and S-systemrdquo Bioinformatics vol 19 no 5 pp 643ndash650 2003
[78] N Shikata YMaki M Nakatsui et al ldquoDetermining importantregulatory relations of amino acids from dynamic networkanalysis of plasma amino acidsrdquo Amino Acids vol 38 no 1 pp179ndash187 2010
[79] M Nakatsui T Ueda Y Maki I Ono and M OkamotoldquoMethod for inferring and extracting reliable genetic interac-tions from time-series profile of gene expressionrdquoMathematicalBiosciences vol 215 no 1 pp 105ndash114 2008
[80] A Komori Y Maki M Nakatsui I Ono and M OkamotoldquoEfficient numerical optimization algorithm based on new real-coded genetic algorithm AREX + JGG and application to theinverse problem in systems biologyrdquo Applied Mathematics vol3 no 10 pp 1463ndash1470 2012
[81] J A Roubos G van Straten and A J B van Boxtel ldquoAn evolu-tionary strategy for fed-batch bioreactor optimization conceptsand performancerdquo Journal of Biotechnology vol 67 no 2-3 pp173ndash187 1999
[82] D Sarkar and J M Modak ldquoGenetic algorithms with filters foroptimal control problems in fed-batch bioreactorsrdquo Bioprocessand Biosystems Engineering vol 26 no 5 pp 295ndash306 2004
[83] K Harada K Ikeda J Sakuma I Ono and S KobayashildquoHybridization of genetic algorithm with local search in mul-tiobjective function optimization recommendation of GA thenLSrdquoTransactions of the Japanese Society for Artificial Intelligencevol 21 no 6 pp 482ndash492 2006
Submit your manuscripts athttpwwwhindawicom
PainResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
ToxinsJournal of
VaccinesJournal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AntibioticsInternational Journal of
ToxicologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
StrokeResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Drug DeliveryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Pharmacological Sciences
Tropical MedicineJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AddictionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Emergency Medicine InternationalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Autoimmune Diseases
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anesthesiology Research and Practice
ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Pharmaceutics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
MEDIATORSINFLAMMATION
of
18 Journal of Toxicology
Biophysica Acta (BBA)mdashMolecular and Cell Biology of Lipidsvol 1771 no 6 pp 663ndash676 2007
[25] S Szabo and I T Lippe ldquoAdrenal gland chemically inducedstructural and functional changes in the cortexrdquo ToxicologicPathology vol 17 no 2 pp 317ndash329 1989
[26] C Chen-Pan I-J Pan Y Yamamoto T Sakogawa J Yamadaand YHayashi ldquoPrompt recovery of damaged adrenalmedullaeinduced by salinomycinrdquoToxicologic Pathology vol 27 no 5 pp563ndash572 1999
[27] R J Santen and R I Misbin ldquoAminoglutethimide review ofpharmacology and clinical userdquo Pharmacotherapy vol 1 no 2pp 95ndash120 1981
[28] M J Carella N V Dimitrov V V Gossain L Srivastava andD R Rovner ldquoAdrenal effects of low-dose aminoglutethimidewhen used alone in postmenopausal women with advancedbreast cancerrdquoMetabolism vol 43 no 6 pp 723ndash727 1994
[29] M K Johansson J T Sanderson and B-O Lund ldquoEffects of3-MeSO2-DDE and some CYP inhibitors on glucocorticoidsteroidogenesis in the H295R human adrenocortical carcinomacell linerdquo Toxicology in Vitro vol 16 no 2 pp 113ndash121 2002
[30] Y Touitou A Bogdan J C Legrand and P Desgrez ldquoAmino-glutethimide and glutethimide effects on 18-hydroxycorti-costerone biosynthesis by human and sheep adrenals in vitrordquoActa Endocrinologica vol 80 no 3 pp 517ndash526 1975
[31] M M Hart and J A Straw ldquoStudies on the site of action ofop1015840-DDD in the dog adrenal cortex 1 Inhibition of ACTH-mediated pregnenolone synthesisrdquo Steroids vol 17 no 1ndash5 pp559ndash574 1971
[32] M Ojima M Saitoh N Itoh Y Kusano S Fukuchi and HNaganuma ldquoThe effects of op1015840-DDD on adrenal steroidogen-esis and hepatic steroid metabolismrdquo Nippon Naibunpi Gakkaizasshi vol 61 no 3 pp 168ndash178 1985
[33] L F Canosa and N R Ceballos ldquoEffects of different steroid-biosynthesis inhibitors on the testicular steroidogenesis of thetoad Bufo arenarumrdquo Journal of Comparative Physiology B vol171 no 6 pp 519ndash526 2001
[34] D C Kossor S Kominami S Takemori and H D ColbyldquoRole of the steroid 17 alpha-hydroxylase in spironolactone-mediated destruction of adrenal cytochrome P-450rdquoMolecularPharmacology vol 40 no 2 pp 321ndash325 1991
[35] S C Cheng K Suzuki W Sadee and B W Harding ldquoEffectsof spironolactone canrenone and canrenoate K on cytochromeP450 and 11120573 and 18 hydroxylation in bovine and humanadrenal corticalmitochondriardquo Endocrinology vol 99 no 4 pp1097ndash1106 1976
[36] K Yanagibashi M Haniu J E Shively W H Shen and P HallldquoThe synthesis of aldosterone by the adrenal cortex Two zones(fasciculata and glomerulosa) possess one enzyme for 11120573- 18-hydroxylation and aldehyde synthesisrdquoThe Journal of BiologicalChemistry vol 261 no 8 pp 3556ndash3562 1986
[37] K Morishita H Okumura N Ito and N Takahashi ldquoPrimaryculture system of adrenocortical cells from dogs to evaluatedirect effects of chemicals on steroidogenesisrdquo Toxicology vol165 no 2-3 pp 171ndash178 2001
[38] S W J Lamberts E G Bons H A Bruining and F H de JongldquoDifferential effects of the imidazole derivatives etomidateketoconazole and miconazole and of metyrapone on the secre-tion of cortisol and its precursors by human adrenocorticalcellsrdquo Journal of Pharmacology and Experimental Therapeuticsvol 240 no 1 pp 259ndash264 1987
[39] L Yin S Lucas F Maurer U Kazmaier Q Hu and RW Hartmann ldquoNovel imidazol-1-ylmethyl substituted 1256-tetrahydropyrrolo[321-ij]quinolin-4-ones as potent and selec-tive CYP11B1 inhibitors for the treatment of Cushingrsquos syn-dromerdquo Journal ofMedicinal Chemistry vol 55 no 14 pp 6629ndash6633 2012
[40] M DiMattina N Maronian H Ashby D L Loriaux andB D Albertson ldquoKetoconazole inhibits multiple steroidogenicenzymes involved in androgen biosynthesis in the humanovaryrdquo Fertility and Sterility vol 49 no 1 pp 62ndash65 1988
[41] M Ayub and M J Levell ldquoInhibition of human adrenal steroi-dogenic enzymes in vitro by imidazole drugs including keto-conazolerdquo Journal of Steroid Biochemistry vol 32 no 4 pp 515ndash524 1989
[42] D Engelhardt M M Weber T Miksch F Abedinpour andC Jaspers ldquoThe influence of ketoconazole on human adrenalsteroidogenesis incubation studies with tissue slicesrdquo ClinicalEndocrinology vol 35 no 2 pp 163ndash168 1991
[43] S Bhasin S Sikka T Fielder et al ldquoHormonal effects of keto-conazole in vivo in the male rat mechanism of actionrdquo Endo-crinology vol 118 no 3 pp 1229ndash1232 1986
[44] D M Zisterer and D C Williams ldquoCalmidazolium and otherimidazole compounds affect steroidogenesis in Y1 cells lack ofinvolvement of the peripheral-type benzodiazepine receptorrdquoJournal of Steroid Biochemistry and Molecular Biology vol 60no 3-4 pp 189ndash195 1997
[45] L P Walsh C N Kuratko and D M Stocco ldquoEconazoleand miconazole inhibit steroidogenesis and disrupt steroido-genic acute regulatory (StAR) protein expression post-trans-criptionallyrdquo Journal of Steroid Biochemistry and Molecular Bio-logy vol 75 no 4-5 pp 229ndash236 2000
[46] S Ohno S Shinoda S Toyoshima H Nakazawa T Makinoand S Nakajin ldquoEffects of flavonoid phytochemicals on cor-tisol production and on activities of steroidogenic enzymesin human adrenocortical H295R cellsrdquo The Journal of SteroidBiochemistry and Molecular Biology vol 80 no 3 pp 355ndash3632002
[47] D P Dempsher D S Gann and R D Phair ldquoA mechanisticmodel of ACTH-stimulated cortisol secretionrdquo American Jour-nal of Physiology vol 246 no 4 part 2 pp R587ndashR596 1984
[48] N Sarai S Matsuoka and A Noma ldquosimBio a Java package forthe development of detailed cell modelsrdquo Progress in Biophysicsand Molecular Biology vol 90 no 1ndash3 pp 360ndash377 2006
[49] K Levenberg ldquoA method for the solution of certain non-linearproblems in least squaresrdquo Quarterly of Applied Mathematicsvol 2 pp 164ndash168 1944
[50] D W Marquardt ldquoAn algorithm for least-squares estimation ofnonlinear parametersrdquo SIAM Journal on Applied Mathematicsvol 11 no 2 pp 431ndash441 1963
[51] P E Gill and W Murray ldquoAlgorithms for the solution of thenonlinear least-squares problemrdquo SIAM Journal on NumericalAnalysis vol 15 no 5 pp 977ndash992 1978
[52] Y Akimoto R Hasada J Sakuma I Ono and S KobayashildquoGeneration alternation model for real-coded GA using multi-parent proposal and evaluation of just generation gap (JGG)rdquoin Proceedings of the 19th SICE Symposium on DecentralizedAutonomous Systems pp 341ndash346 Tokyo Japan January 2007
[53] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoProposal and evaluation of adaptive real-coded crossoverAREXrdquo Transactions of the Japanese Society for Artificial Intel-ligence vol 24 no 6 pp 446ndash458 2009
Journal of Toxicology 19
[54] S Kobayashi ldquoThe frontiers of real-coded genetic algorithmsrdquoJournal of Japanese Society for Artificial Intelligence vol 24 no1 pp 147ndash162 2009
[55] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoAnalysis of the behavior of MGG and JGG as a selection modelfor real-coded genetic algorithmsrdquo Transactions of the JapaneseSociety for Artificial Intelligence vol 25 no 2 pp 281ndash289 2010
[56] A Struthers H Krum and G H Williams ldquoA comparisonof the aldosterone-blocking agents eplerenone and spironolac-tonerdquo Clinical Cardiology vol 31 no 4 pp 153ndash158 2008
[57] K Yamasaki M Sawaki S Noda et al ldquoComparison of theHershberger assay and androgen receptor binding assay oftwelve chemicalsrdquo Toxicology vol 195 no 2-3 pp 177ndash1862004
[58] C Eil and S K Edelson ldquoThe use of human skin fibroblasts toobtain potency estimates of drug binding to androgen recept-orsrdquo The Journal of Clinical Endocrinology amp Metabolism vol59 no 1 pp 51ndash55 1984
[59] P Ye T Yamashita D M Pollock H Sasano and W ERainey ldquoContrasting effects of eplerenone and spironolac-tone on adrenal cell steroidogenesisrdquo Hormone and MetabolicResearch vol 41 no 1 pp 35ndash39 2009
[60] B D Albertson K L Frederick N C Maronian et al ldquoTheeffect of ketoconazole on steroidogenesis I Leydig cell enzymeactivity in vitrordquo Research Communications in Chemical Patho-logy and Pharmacology vol 61 no 1 pp 17ndash26 1988
[61] B D Albertson N C Maronian K L Frederick et al ldquoTheeffect of ketoconazole on steroidogenesis II Adrenocorticalenzyme activity in vitrordquoResearch Communications in ChemicalPathology and Pharmacology vol 61 no 1 pp 27ndash34 1988
[62] DM Stocco andB J Clark ldquoRegulation of the acute productionof steroids in steroidogenic cellsrdquo Endocrine Reviews vol 17 no3 pp 221ndash244 1996
[63] L K Christenson and J F Strauss III ldquoSteroidogenic acute regu-latory protein (StAR) and the intramitochondrial translocationof cholesterolrdquo Biochimica et Biophysica Acta (BBA)mdashMolecularand Cell Biology of Lipids vol 1529 no 1ndash3 pp 175ndash187 2000
[64] E R Simpson ldquoCholesterol side-chain cleavage cytochromeP450 and the control of steroidogenesisrdquo Molecular and Cel-lular Endocrinology vol 13 no 3 pp 213ndash227 1979
[65] H A LaVoie and S R King ldquoTranscriptional regulation of ster-oidogenic genes STARD1 CYP11A1 and HSD3Brdquo ExperimentalBiology and Medicine vol 234 no 8 pp 880ndash907 2009
[66] F Raucci A DrsquoAniello and M M Di Fiore ldquoStimulation ofandrogen production by d-aspartate through the enhancementof StAR P450scc and 3120573-HSD mRNA levels in vivo rat testisand in culture of immature rat Leydig cellsrdquo Steroids vol 84pp 103ndash110 2014
[67] M Doi Y Takahashi R Komatsu et al ldquoSalt-sensitive hyper-tension in circadian clock-deficient Cry-null mice involves dys-regulated adrenal Hsd3b6rdquo Nature Medicine vol 16 no 1 pp67ndash74 2010
[68] H Okamura M Doi Y Yamaguchi and J-M Fustin ldquoHyper-tension due to loss of clock novel insight from the molecu-lar analysis of Cry1Cry2-deleted micerdquo Current HypertensionReports vol 13 no 2 pp 103ndash108 2011
[69] M Doi F Satoh T Maekawa et al ldquoIsoform-specific mono-clonal antibodies against 3120573-hydroxysteroid dehydrogenaseisomerase family provide markers for subclassification ofhuman primary aldosteronismrdquo The Journal of Clinical Endo-crinology amp Metabolism vol 99 no 2 pp E257ndashE262 2014
[70] S Konosu-Fukaya Y Nakamura F Satoh et al ldquo3120573-hydroxy-steroid dehydrogenase isoforms in human aldosterone-produc-ing adenomardquo Molecular and Cellular Endocrinology vol 408pp 205ndash212 2015
[71] G C Verwoert J Hofland N Amin et al ldquoExpression and genevariation studies deny association of humanHSD3B1 gene withaldosterone production or blood pressurerdquoAmerican Journal ofHypertension vol 28 no 1 pp 113ndash120 2015
[72] D Hough K Storbeck SW P Cloete A C Swart and P SwartldquoRelative contribution of P450c17 towards the acute cortisolresponse lessons from sheep and goatsrdquoMolecular and CellularEndocrinology vol 408 pp 107ndash113 2015
[73] I Sakuma S Suematsu Y Matsuzawa et al ldquoCharacterizationof steroidogenic enzyme expression in aldosterone-producingadenoma a comparison with various human adrenal tumorsrdquoEndocrine Journal vol 60 no 3 pp 329ndash336 2013
[74] A Tong A Jia S Yan Y Zhang Y Xie and G Liu ldquoEctopiccortisol-producing adrenocortical adenoma in the renal hilumhistopathological features and steroidogenic enzyme profilerdquoInternational Journal of Clinical and Experimental Pathologyvol 7 no 7 pp 4415ndash4421 2014
[75] A Hirsch D Hahn P Kempna et al ldquoMetformin inhibitshuman androgen production by regulating steroidogenicenzymes HSD3B2 and CYP17A1 and complex I activity of therespiratory chainrdquo Endocrinology vol 153 no 9 pp 4354ndash43662012
[76] B Xu L Gao Y Cui et al ldquoSET protein up-regulated test-osterone production in the cultured preantral folliclesrdquo Repro-ductive Biology and Endocrinology vol 11 article 9 2013
[77] S Kikuchi D Tominaga M Arita M Takahashi and MTomita ldquoDynamic modeling of genetic networks using geneticalgorithm and S-systemrdquo Bioinformatics vol 19 no 5 pp 643ndash650 2003
[78] N Shikata YMaki M Nakatsui et al ldquoDetermining importantregulatory relations of amino acids from dynamic networkanalysis of plasma amino acidsrdquo Amino Acids vol 38 no 1 pp179ndash187 2010
[79] M Nakatsui T Ueda Y Maki I Ono and M OkamotoldquoMethod for inferring and extracting reliable genetic interac-tions from time-series profile of gene expressionrdquoMathematicalBiosciences vol 215 no 1 pp 105ndash114 2008
[80] A Komori Y Maki M Nakatsui I Ono and M OkamotoldquoEfficient numerical optimization algorithm based on new real-coded genetic algorithm AREX + JGG and application to theinverse problem in systems biologyrdquo Applied Mathematics vol3 no 10 pp 1463ndash1470 2012
[81] J A Roubos G van Straten and A J B van Boxtel ldquoAn evolu-tionary strategy for fed-batch bioreactor optimization conceptsand performancerdquo Journal of Biotechnology vol 67 no 2-3 pp173ndash187 1999
[82] D Sarkar and J M Modak ldquoGenetic algorithms with filters foroptimal control problems in fed-batch bioreactorsrdquo Bioprocessand Biosystems Engineering vol 26 no 5 pp 295ndash306 2004
[83] K Harada K Ikeda J Sakuma I Ono and S KobayashildquoHybridization of genetic algorithm with local search in mul-tiobjective function optimization recommendation of GA thenLSrdquoTransactions of the Japanese Society for Artificial Intelligencevol 21 no 6 pp 482ndash492 2006
Submit your manuscripts athttpwwwhindawicom
PainResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
ToxinsJournal of
VaccinesJournal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AntibioticsInternational Journal of
ToxicologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
StrokeResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Drug DeliveryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Pharmacological Sciences
Tropical MedicineJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AddictionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Emergency Medicine InternationalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Autoimmune Diseases
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anesthesiology Research and Practice
ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Pharmaceutics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
MEDIATORSINFLAMMATION
of
Journal of Toxicology 19
[54] S Kobayashi ldquoThe frontiers of real-coded genetic algorithmsrdquoJournal of Japanese Society for Artificial Intelligence vol 24 no1 pp 147ndash162 2009
[55] Y Akimoto Y Nagata J Sakuma I Ono and S KobayashildquoAnalysis of the behavior of MGG and JGG as a selection modelfor real-coded genetic algorithmsrdquo Transactions of the JapaneseSociety for Artificial Intelligence vol 25 no 2 pp 281ndash289 2010
[56] A Struthers H Krum and G H Williams ldquoA comparisonof the aldosterone-blocking agents eplerenone and spironolac-tonerdquo Clinical Cardiology vol 31 no 4 pp 153ndash158 2008
[57] K Yamasaki M Sawaki S Noda et al ldquoComparison of theHershberger assay and androgen receptor binding assay oftwelve chemicalsrdquo Toxicology vol 195 no 2-3 pp 177ndash1862004
[58] C Eil and S K Edelson ldquoThe use of human skin fibroblasts toobtain potency estimates of drug binding to androgen recept-orsrdquo The Journal of Clinical Endocrinology amp Metabolism vol59 no 1 pp 51ndash55 1984
[59] P Ye T Yamashita D M Pollock H Sasano and W ERainey ldquoContrasting effects of eplerenone and spironolac-tone on adrenal cell steroidogenesisrdquo Hormone and MetabolicResearch vol 41 no 1 pp 35ndash39 2009
[60] B D Albertson K L Frederick N C Maronian et al ldquoTheeffect of ketoconazole on steroidogenesis I Leydig cell enzymeactivity in vitrordquo Research Communications in Chemical Patho-logy and Pharmacology vol 61 no 1 pp 17ndash26 1988
[61] B D Albertson N C Maronian K L Frederick et al ldquoTheeffect of ketoconazole on steroidogenesis II Adrenocorticalenzyme activity in vitrordquoResearch Communications in ChemicalPathology and Pharmacology vol 61 no 1 pp 27ndash34 1988
[62] DM Stocco andB J Clark ldquoRegulation of the acute productionof steroids in steroidogenic cellsrdquo Endocrine Reviews vol 17 no3 pp 221ndash244 1996
[63] L K Christenson and J F Strauss III ldquoSteroidogenic acute regu-latory protein (StAR) and the intramitochondrial translocationof cholesterolrdquo Biochimica et Biophysica Acta (BBA)mdashMolecularand Cell Biology of Lipids vol 1529 no 1ndash3 pp 175ndash187 2000
[64] E R Simpson ldquoCholesterol side-chain cleavage cytochromeP450 and the control of steroidogenesisrdquo Molecular and Cel-lular Endocrinology vol 13 no 3 pp 213ndash227 1979
[65] H A LaVoie and S R King ldquoTranscriptional regulation of ster-oidogenic genes STARD1 CYP11A1 and HSD3Brdquo ExperimentalBiology and Medicine vol 234 no 8 pp 880ndash907 2009
[66] F Raucci A DrsquoAniello and M M Di Fiore ldquoStimulation ofandrogen production by d-aspartate through the enhancementof StAR P450scc and 3120573-HSD mRNA levels in vivo rat testisand in culture of immature rat Leydig cellsrdquo Steroids vol 84pp 103ndash110 2014
[67] M Doi Y Takahashi R Komatsu et al ldquoSalt-sensitive hyper-tension in circadian clock-deficient Cry-null mice involves dys-regulated adrenal Hsd3b6rdquo Nature Medicine vol 16 no 1 pp67ndash74 2010
[68] H Okamura M Doi Y Yamaguchi and J-M Fustin ldquoHyper-tension due to loss of clock novel insight from the molecu-lar analysis of Cry1Cry2-deleted micerdquo Current HypertensionReports vol 13 no 2 pp 103ndash108 2011
[69] M Doi F Satoh T Maekawa et al ldquoIsoform-specific mono-clonal antibodies against 3120573-hydroxysteroid dehydrogenaseisomerase family provide markers for subclassification ofhuman primary aldosteronismrdquo The Journal of Clinical Endo-crinology amp Metabolism vol 99 no 2 pp E257ndashE262 2014
[70] S Konosu-Fukaya Y Nakamura F Satoh et al ldquo3120573-hydroxy-steroid dehydrogenase isoforms in human aldosterone-produc-ing adenomardquo Molecular and Cellular Endocrinology vol 408pp 205ndash212 2015
[71] G C Verwoert J Hofland N Amin et al ldquoExpression and genevariation studies deny association of humanHSD3B1 gene withaldosterone production or blood pressurerdquoAmerican Journal ofHypertension vol 28 no 1 pp 113ndash120 2015
[72] D Hough K Storbeck SW P Cloete A C Swart and P SwartldquoRelative contribution of P450c17 towards the acute cortisolresponse lessons from sheep and goatsrdquoMolecular and CellularEndocrinology vol 408 pp 107ndash113 2015
[73] I Sakuma S Suematsu Y Matsuzawa et al ldquoCharacterizationof steroidogenic enzyme expression in aldosterone-producingadenoma a comparison with various human adrenal tumorsrdquoEndocrine Journal vol 60 no 3 pp 329ndash336 2013
[74] A Tong A Jia S Yan Y Zhang Y Xie and G Liu ldquoEctopiccortisol-producing adrenocortical adenoma in the renal hilumhistopathological features and steroidogenic enzyme profilerdquoInternational Journal of Clinical and Experimental Pathologyvol 7 no 7 pp 4415ndash4421 2014
[75] A Hirsch D Hahn P Kempna et al ldquoMetformin inhibitshuman androgen production by regulating steroidogenicenzymes HSD3B2 and CYP17A1 and complex I activity of therespiratory chainrdquo Endocrinology vol 153 no 9 pp 4354ndash43662012
[76] B Xu L Gao Y Cui et al ldquoSET protein up-regulated test-osterone production in the cultured preantral folliclesrdquo Repro-ductive Biology and Endocrinology vol 11 article 9 2013
[77] S Kikuchi D Tominaga M Arita M Takahashi and MTomita ldquoDynamic modeling of genetic networks using geneticalgorithm and S-systemrdquo Bioinformatics vol 19 no 5 pp 643ndash650 2003
[78] N Shikata YMaki M Nakatsui et al ldquoDetermining importantregulatory relations of amino acids from dynamic networkanalysis of plasma amino acidsrdquo Amino Acids vol 38 no 1 pp179ndash187 2010
[79] M Nakatsui T Ueda Y Maki I Ono and M OkamotoldquoMethod for inferring and extracting reliable genetic interac-tions from time-series profile of gene expressionrdquoMathematicalBiosciences vol 215 no 1 pp 105ndash114 2008
[80] A Komori Y Maki M Nakatsui I Ono and M OkamotoldquoEfficient numerical optimization algorithm based on new real-coded genetic algorithm AREX + JGG and application to theinverse problem in systems biologyrdquo Applied Mathematics vol3 no 10 pp 1463ndash1470 2012
[81] J A Roubos G van Straten and A J B van Boxtel ldquoAn evolu-tionary strategy for fed-batch bioreactor optimization conceptsand performancerdquo Journal of Biotechnology vol 67 no 2-3 pp173ndash187 1999
[82] D Sarkar and J M Modak ldquoGenetic algorithms with filters foroptimal control problems in fed-batch bioreactorsrdquo Bioprocessand Biosystems Engineering vol 26 no 5 pp 295ndash306 2004
[83] K Harada K Ikeda J Sakuma I Ono and S KobayashildquoHybridization of genetic algorithm with local search in mul-tiobjective function optimization recommendation of GA thenLSrdquoTransactions of the Japanese Society for Artificial Intelligencevol 21 no 6 pp 482ndash492 2006
Submit your manuscripts athttpwwwhindawicom
PainResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
ToxinsJournal of
VaccinesJournal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AntibioticsInternational Journal of
ToxicologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
StrokeResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Drug DeliveryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Pharmacological Sciences
Tropical MedicineJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AddictionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Emergency Medicine InternationalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Autoimmune Diseases
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anesthesiology Research and Practice
ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Pharmaceutics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
MEDIATORSINFLAMMATION
of
Submit your manuscripts athttpwwwhindawicom
PainResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
ToxinsJournal of
VaccinesJournal of
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AntibioticsInternational Journal of
ToxicologyJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
StrokeResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Drug DeliveryJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Pharmacological Sciences
Tropical MedicineJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Medicinal ChemistryInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AddictionJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
BioMed Research International
Emergency Medicine InternationalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Autoimmune Diseases
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Anesthesiology Research and Practice
ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Pharmaceutics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
MEDIATORSINFLAMMATION
of
Related Documents
