Polyamine Acetylation Modulates Polyamine Metabolic Flux, a Prelude to Broader Metabolic Consequences

Polyamine Acetylation Modulates Polyamine Metabolic Flux,a Prelude to Broader Metabolic Consequences*

Received for publication, August 15, 2007, and in revised form, November 6, 2007 Published, JBC Papers in Press, December 18, 2007, DOI 10.1074/jbc.M706806200

Debora L. Kramer‡, Paula Diegelman‡, Jason Jell‡, Slavoljub Vujcic‡, Salim Merali§, and Carl W. Porter‡1

From the ‡Department of Pharmacology and Therapeutics, Roswell Park Cancer Institute, Buffalo, New York 14263 and §FelsInstitute and Biochemistry Department, Temple University School of Medicine, Philadelphia, Pennsylvania 19140

Recent studies suggest that overexpression of the polyamine-acetylating enzyme spermidine/spermine N1-acetyltransferase(SSAT) significantly increases metabolic flux through the poly-amine pathway. The concept derives from the observation thatSSAT-induced acetylation of polyamines gives rise to a compen-satory increase in biosynthesis and presumably to increasedflow through the pathway. Despite the strength of this deduc-tion, the existenceof heightenedpolyamine fluxhasnot yet beenexperimentally demonstrated. Here, we use the artificial poly-amine precursor 4-fluoro-ornithine to measure polyamineflux by tracking fluorine unit permeation of polyamine poolsin human prostate carcinoma LNCaP cells. Conditional over-expression of SSATwas accompanied by amassive increase inintracellular and extracellular acetylated spermidine and by a6–20-fold increase in biosynthetic enzyme activities. Inthe presence of 300 �M 4-fluoro-ornithine, SSAT overexpres-sion led to the sequential appearance of fluorinated putres-cine, spermidine, acetylated spermidine, and spermine. Asfluorinated polyamines increased, endogenous polyaminesdecreased, so that the total polyamine pool size remained rela-tively constant. At 24 h, 56%of the spermine pool in the inducedSSAT cells was fluorine-labeled compared with only 12% inuninduced cells. Thus, SSAT induction increasedmetabolic fluxby �5-fold. Flux could be interrupted by inhibition of poly-amine biosynthesis but not by inhibition of polyamine oxida-tion. Overall, the findings are consistent with a paradigmwhereby flux is initiated by SSAT acetylation of spermine andparticularly spermidine followed by a marked increase in keybiosynthetic enzymes. The latter sustains the flux cycle by pro-viding a constant supply of polyamines for subsequent acetyla-tion by SSAT. The broader metabolic implications of this futilemetabolic cycling are discussed in detail.

Intracellular levels of the polyamines, putrescine, spermi-dine, and spermine are thought to be homeostatically main-tained within a relatively constant range by three main effectorsystems: biosynthesis, acetylation/export, and transport (1). Inmammalian cells, biosynthesis is regulated by ornithine andS-adenosylmethionine decarboxylases (ODC2 and SAMDC);acetylation and export are regulated by spermidine/spermineN1-acetyltransferase (SSAT); and transport is regulated by agroup of unidentifiedmembrane protein(s). Each of these effec-tors is sensitively responsive to intracellular polyamine pools,with uptake and biosynthesis being negatively regulated bypolyamines and acetylation being positively regulated (2, 3). Inantiproliferative strategies utilizing biosynthetic enzyme inhib-itors (4–8), these various homeostatic effectors counter poly-amine pool depletion and thus compromise growth inhibition(9–11). In an alternative approach, we exploited homeostaticresponses by using polyamine analogs to create a circumstanceof apparent polyamine excess (1, 12). The resulting down-reg-ulation of polyamine biosynthesis export was accompanied byrapid depletion of endogenous polyamine pools and inhibitionof cell growth (13). Subsequent mechanistic studies showedthat rapid polyamine pool depletion was at least partially due topotent induction of the acetylating enzyme SSAT (14, 15),which facilitated polyamine oxidation and their export out ofthe cell (16, 17). Thus, induction of SSATbecame recognized asa major contributor to analog-mediated growth inhibitionand/or apoptosis (18, 19).The cytosolic enzyme SSAT acetylates spermine (Spm) and

spermidine (Spd) to formN1-acetylspermine (AcSpm),N1,N12-diacetylspermine (DAS), and N1-acetylspermidine (AcSpd),which are then rendered susceptible to either export out of thecell or oxidation to lower polyamines by polyamine oxidase(PAO) (20, 21). Attempts to define the cellular consequences ofselective SSAT induction on cell growth have relied on trans-fection strategies (22, 23). Conditional overexpression of SSATin LNCaP human prostate carcinoma cells led to growth inhi-

* This work was supported by National Institutes of Health (NIH) GrantsCA-22153, CA 109619, and CA-76428 and NIH Predoctoral Training GrantCA-09072. The costs of publication of this article were defrayed in part bythe payment of page charges. This article must therefore be herebymarked “advertisement” in accordance with 18 U.S.C. Section 1734 solely toindicate this fact.

This paper is dedicated to the memory of Professor Nikolaus Seiler, an out-standing scientist who, before others, understood the biological impor-tance of polyamine acetylation and catabolism and whose pioneeringmetabolic studies in this area greatly facilitated the studies and interpre-tations presented here.

1 To whom correspondence should be addressed: Dept. of Pharmacology andTherapeutics, Roswell Park Cancer Institute, Buffalo, NY 14263. Tel.: 716-845-3002; Fax: 716-845-2353; E-mail: [email protected].

2 The abbreviations and trivial names used are: ODC, ornithine decarb-oxylase; DAS, N1,N12-diacetylspermine; Fl-Orn, 4-fluoro-L-ornithine;Fl-Put, 2-fluoroputrescine; Fl-Spd, 6-fluorospermidine; Fl-Spm, 6-fluorospermine; Fl-AcSpd, 6-fluoro-N1-acetylspermidine; Fl-AcSpm,6-fluoro-N1-acetylspermine; HPLC, high pressure liquid chromatogra-phy; Orn, ornithine; Put, putrescine; SAMDC, S-adenosylmethioninedecarboxylase; Spd, spermidine; Spm, spermine; AcSpd, N1-acetylsper-midine; AcSpm, N1-acetylspermine; SSAT, spermidine/spermineN1-acetyltransferase; PAO, polyamine oxidase; Tet, tetracycline; nor-NOHA, N�-hydroxy-nor-L-arginine; MDL-72527, N1-methyl-N2-(2,3-butadienyl)butane-1,4-diamine; MDL-73811, 5�-[(Z)-4-amino-2-bute-nyl] methyl-amino-5�-deoxyadenosine.

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 283, NO. 7, pp. 4241–4251, February 15, 2008© 2008 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.

FEBRUARY 15, 2008 • VOLUME 283 • NUMBER 7 JOURNAL OF BIOLOGICAL CHEMISTRY 4241

by guest on August 14, 2016

http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

bition, but in contrast to expectations, polyamine pools werenot depleted (23). Despite intracellular accumulation of hugeamounts of acetylated polyamines, intracellular polyaminepools were only marginally reduced due, apparently, to a com-pensatory increase in ODC and SAMDC (23). We now knowfrom other studies that growth inhibition in the absence ofpolyamine pool depletion, as seen in LNCaP cells, is a typicalcellular response to SSAT overexpression (24, 25).Similar biochemical findings to those seen in LNCaP cells

were made in transgenic overexpression of SSAT, which led toa hairless phenotype (26) and altered tumor growth (27, 28).Wenow believe that the various effects seen in cells and mice maybe due to altered metabolic flux through the polyamine path-way (23). In other words, SSAT acetylation of polyamines leadsto increased biosynthesis and heightened metabolic flux (Fig.1). The adverse downstream consequences of sustained meta-bolic flux that account for growth inhibition and other effectsare potentially 2-fold: pathway precursor depletion or accumu-lation of toxic pathway by-products (23, 27–30). Examples ofthe former could include depletion of the SAMDC substrateand methylation precursor, S-adenosylmethionine, or deple-tion of the SSAT substrate and fatty acid precursor, acetyl-CoA(23, 27, 30). Possible by-product examples could include accu-mulation of the SAMDC by-product 5�-methylthioadenosineor accumulation of the PAO by-product, hydrogen peroxide(20). Detailed studies in LNCaP cells strongly relate growthinhibition to the fatty acid precursor acetyl-CoA, a finding sup-ported by the lean phenotype seen in SSAT transgenicmice (26,27, 29, 30) as well as the fatty phenotype seen in SSAT knock-out mice (30). More particularly, acetyl-CoA was depleted inSSAT-overexpressing LNCaP cells and in the adipose tissue ofSSAT transgenic mice. Thus far, the existence of SSAT-in-duced metabolic flux has only been indirectly inferred on thebasis of increased SSAT activity, increased ODC and SAMDCactivities, accumulation of acetylated polyamines, and persist-ence of Spd and Spm pools. Although the heightened flux par-adigm is compelling and seemingly without an alternativeinterpretation, it has not yet been experimentally demonstratedby direct measurement, which is the intention of this presentstudy.Previously, we reported a method for quantifying polyamine

flux thatmakes use of (a) the ability of fluoro-ornithine (Fl-Orn)to substitute for ornithine as a precursor to putrescine (Put)biosynthesis, (b) the metabolic processing of fluorinated Put tohigher polyamines in the absence of overt cytotoxicity, and (c)our ability to distinguish fluorinated polyamines from endoge-nous polyamines by high performance liquid chromatography(HPLC) (31). The method was used to characterize metabolicflux under various growth states and pharmacological condi-tions butnotduring selective induction of SSAT. In similarity toflux measurements based on radioactive tracers, this approachallows for unambiguous separation of newly synthesized poly-amines from preexisting polyamines. However, fluorotaggingallows these pools to be visualized and quantified on the samechromatogram and without the need for additional processingsteps or equipment. Here, this methodology was used to dem-onstrate that sustained induction of SSAT increases metabolicflux through the polyamine pathway. Given the many factors

that induce SSAT (32–34), these findings could have importantimplications for better understanding the role of SSAT in drugaction and for realizing their potential to affect other metabolicprocesses.

EXPERIMENTAL PROCEDURES

Materials—4-Fluoro-L-ornithine (Fl-Orn) was synthesizedaccording to a published procedure (31) and provided as a giftby Dr. Janos Kollonitsch (Merck). The PAO inhibitorN1-meth-yl-N2-(2,3-butadienyl)butane-1,4-diamine (MDL-72527) andthe SAMDC inhibitor 5�-[(Z)-4-amino-2-butenyl]methyl-ami-no-5�-deoxyadenosine (MDL-73811) were generously pro-vided by Aventis Pharmaceuticals, Inc. (Bridgewater, NJ). TheODC inhibitor �-difluoromethylornithine was obtained fromILEX, Inc. (San Antonio, TX). The arginase inhibitor N�-hy-droxy-nor-L-arginine (nor-NOHA) was purchased from EMDChemicals, Inc. (La Jolla, CA). Tetracycline (Tet), aminoguani-dine, polyamines, and the acetylated polyamines AcSpd andAcSpm were purchased from Sigma. The 2-fluoroputrescine(Fl-Put) andDASwas obtained from the lateDr.Nikolaus Seiler(University of Rennes, France). Radioactive [3H]Spd (27.6Ci/mmol) was purchased from PerkinElmer Life Sciences.Cell Culture Conditions—LNCaP prostate carcinoma cells

constitutively expressing the Tet-repressible transactivator(35) were used to transfect human SSAT cDNA and select forconditional overexpression of SSAT (23). Designated SSAT/LNCaP, these cells were routinely grown in RPMI 1640medium supplemented with 2 mM glutamine (Invitrogen), 10%Tet-approved fetal bovine serum (Clontech), 100 units/ml pen-icillin, 100 units/ml streptomycin (Mediatech Inc., Herndon,VA), 150 �g/ml hygromycin B, 150 �g/ml G418, and 1 �g/mlTet at 37 °C in the presence of humidified 5% CO2. Of signifi-cance, RPMI 1640 medium is ornithine-free. For optimal cellattachment enzyme induction and cell growth, experimentswere conducted with poly-D-lysine-coated dishes (BD Bio-sciences). This differs from our previous studies in whichLNCaP cells were grown on standard uncoated dishes (23).Aminoguanidine (1 mM) was routinely added to inhibit serumamino oxidases and prevent conversion of extracellular poly-amines to toxic products. Cells were harvestedwith 0.25% tryp-sin and counted electronically (Coulter model ZM, CoulterElectronics, Hialeah, FL). Just prior to plating, exponentiallygrowing cells were placed inmedium containing 10%NuSerumIV (Collaborative Research, Boston,MA), no addition of hygro-mycin and G418, and 100 ng/ml Tet. Cells were plated onto100-mm dishes in the absence (induced SSAT) and presence(basal SSAT) of 1 �g/ml Tet. After 24 h, fresh medium wasapplied immediately prior to the addition of 300 �M Fl-Orn.Control cells were treated with 300�M ornithine (Orn) in placeof Fl-Orn. Some experiments included co-treatments at fixedtime points with a 50 �M concentration of the polyamine oxi-dase inhibitor MDL-72527, 30 �M SAMDC inhibitorMDL-73811, or 100 �M arginase inhibitor nor-NOHA. From 0to 48 h following the addition of Fl-Orn, bothmedium and cellswere collected and frozen at �20 °C until HPLC analysis wasperformed.HPLC Detection of Polyamines and Fluoropolyamines—Cell

samples were extracted with 0.6 M perchloric acid and centri-

Polyamine Acetylation Modulates Metabolic Flux

4242 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 283 • NUMBER 7 • FEBRUARY 15, 2008


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

fuged. The supernatant extract was assayed for polyamines byreverse phase HPLC.We obtained sharp polyamine peaks withgood separation (Fig. 3) by introducing major modifications toa previously described procedure (36) as detailed below. Poly-amines in 50 �l of each acid extract were injected onto anautosampler (model 717 Plus,Waters Associates,Milford,MA)using a C-18 column (250 � 3 mm; 5 �m) (Alltech Associates,Deerfield, IL) with an acetonitrile/sodium acetate gradientbuffer system and detected with an o-phthalaldehyde postcol-umn derivatization system. The column was maintained at35 °C in a column heater (Waters Associates, Milford, MA).The mobile phase consisted of Buffer A (0.1 M sodium acetateand 15 mM octane sulfonic acid, pH 4.5) and Buffer B (0.25 Msodium acetate and 15 mM octane sulfonic acid, pH 4.5, with30% acetonitrile) at a flow rate of 0.75 ml/min. A low pressuregradient generator (model 6005 controller; Waters Associates)and a programmable pump (model 616; Waters Associates)were used to mix buffers A and B as a linear gradient: at 0 min,100%A; at 25min, 100% B; and at 30min, re-equilibration with100% A. The column eluate was derivatized according to thePickering Laboratories (Mountain View, CA) reagent bulletin,mixed with column eluates in a flow cell at 0.5 ml/min, andpassed through a fluorescent detector (model 2475; WatersAssociates) with fixed excitation and emission wavelengths of330 and 465 nm, respectively. The data were collected and ana-lyzed using Millennium 32 chromatography software version3.05 (Waters Associates). Authentic standards of the naturalpolyamines AcSpd and AcSpm were analyzed separately toidentify and quantitate each peak. Fl-Put was verified by anauthentic standard, and the other fluorinated compounds (Fl-Spd, Fl-Spm, Fl-AcSpd, and Fl-AcSpm) were presumed fromtheir faster retention time (�1min ahead) relative to the parentnatural polyamine. Elution times were as follows: Fl-Put, 14.2min; Put, 14.9 min; Fl-AcSpd, 17.6 min; AcSpd, 18.4 min; 6-Fl-Spd, 20.4min; Spd, 20.9min; Fl-AcSpm, 22.2min; AcSpm, 22.8min; 6-Fl-Spm, 23.2 min; and Spm, 23.6 min. The ratio of Fl-Put/Put was used for calculating the relative values of all otherfluorinated compounds. With sensitivity in the range of 20pmol, polyamine pools were expressed as nmol/106 cells. TheOPA reagent used for postcolumn detection will only bind toprimary amines; thus, monoacetylated polyamines bind thedye, but diacetylated polyamines like DAS (22) do not. DASwas analyzed by rerunning the samples using a precolumndansylation methodology (37) with additional modifications(38).Extracellular Polyamines—Due to high levels of Put found in

fetal bovine serum, duplicate experiments were performedusing 10% NuSerum containing 5-fold less Put. Experimentalmedium was also prepared without the addition of antibioticshygromycin and G418 due to their interference with peakdetection by HPLC analysis. At the indicated times, 10 ml ofmedium collected from each dish was passed through a Milli-pore Centriprep filter chamber (Millipore Corp., Bedford, MA)per the manufacturer’s instructions to remove serum proteins�50 kDa. To 6 ml of each filtered sample, 12 ml of 0.01 Mammoniumphosphate buffer was added and brought to pH 8.0.Samples were then loaded onto preconditioned Bond Elut LRCCBA columns (Varian Associates, Harbor City, CA) and

washed with 2 ml of 0.01 M ammonium phosphate buffer. Toelute each sample, 1.5 ml of 0.1 N HCl in methanol was applied.Each eluate was dried on a Savant SpeedVac (Thermo-Savant,Holbrook, NY) for 1.5 h, reconstituted with 400 �l of double-distilledH2O, and injected (50�l) onto theHPLC for analysis asdescribed above. The data expressed as pmol/50 �l were con-verted to nmol/ml equivalents for 106 cells.Enzyme Activities—SSAT, ODC, and SAMDC enzyme activ-

ities were measured with soluble protein extracts as previouslydescribed (39, 40). SSATwas expressed as pmol ofN1-[14C]ace-tylspermidine generated/min/mg of protein. ODC andSAMDC were expressed as nmol of [14C]O2 released/h/mg ofprotein.Spd Transport—This assay was performed as previously

described (41) with the following modifications. Cells weregrown in 6-well coated dishes (BD Biosciences) in the absenceand presence of Tet 48 h prior to assay, washed with serum-free RPMI 1640 medium, and placed on ice. Triplicate wellswere exposed to 2 ml of [3H]Spd (220 cpm/pmol) mixed withcold Spd to final concentrations of 2 and 10 �M in RMPI 1640medium and incubated in a 37 °C water bath for 30 min (a timewithin the linear range of uptake for these concentrations).After three washes with PBS containing excess cold Spd, cellswere dissolved in 0.1 N NaOH, neutralized with an equal vol-ume of 0.1 N HCl, and counted using a scintillation detector(Beckman Coulter, Fullerton, CA). Data were expressed aspmol/min/mg protein.Measurement of Acetyl-CoA andMalonyl-CoA—Coenzymes

A were monitored using a high performance capillary electro-phoresis as previously described (30), and data obtained fromtriplicate samples were expressed as nmol/108 cells.Statistics—Analysis of data obtained from three ormore sep-

arate experiments included averages � S.D. values. Significantdifferences determined by Student’s t test are indicated as pvalues �0.05 to compare basal SSAT with induced SSATconditions.

RESULTS

Effects of SSAT Overexpression on Polyamine MetabolicParameters and Cell Growth—In the presence of Tet, theSSAT/LNCaP cells grew logarithmically with a doubling timeof �25–30 h. Following Tet removal, SSAT activity was maxi-mally induced by �50-fold at 24 h, after which growth inhibi-tion became apparent (Fig. 1). Although the �50-fold induc-tion in activity achieved with the Tet-regulatable SSATplasmid-borne gene (Table 1) is probably higher than mostphysiological modulation of the endogenous gene, it is lessthan the increases seen with various pharmacological per-turbations (32–34) and particularly those involving poly-amine analogs, where inductions of �500-fold are notuncommon (14, 15).In the first set of experiments, 300 �M Fl-Orn or Orn was

added in order tomonitor flux in cells over a 48-h period underconditions of basal and induced SSAT activity. As previouslyreported (31), 300 �M Fl-Orn had no adverse effects on cellgrowth over this 48-h period (data not shown). In order tointerpret flux findings, polyamine pools and biosyntheticenzyme activities were also measured at the end of this same




http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

48-h period (Table 1 and Fig. 2A). There was a 50-fold increasein SSAT activity, which was accompanied by compensatoryincreases in ODC and SAMDC activities of �7- and 20-fold,respectively, relative to basal SSAT cells. The 50-fold inductionof SSAT activity is much higher than we previously reported(i.e. 20-fold) (23) due to the fact that in the present experiments,the SSAT/LNCaP cells were grown on poly-D-lysine-coateddishes, which increased both enzyme induction and cell toler-ance to it. Because regulation of the polyamine transporter typ-ically parallels ODC in response to changes in polyamine pools,we also examined [3H]Spd uptake (Fig. 2B) and found that itincreased by 3-fold with SSAT induction.Taken together, these data suggest that acetylation of Spd

and Spm by SSAT initiates changes in polyamine homeostasisthat affect biosynthesis and uptake (Fig. 2). Analysis of poly-amine pools (Table 1) shows that induction of SSAT leads to a

modest decrease in Spm pools and a huge increase in Put pools.More significantly, intracellular and extracellular AcSpdincrease remarkably. Since ODC and SAMDC are known to berepressed by intracellular polyamine pools (1, 12, 42), it appearsthat acetylation of Spm and particularly Spd relieves feedbackrepression of the enzymes.Identification of Fluorine-labeled Polyamines byHPLC—Pre-

vious HPLC studies indicate that Fl-Orn is decarboxylated byODC to produce Fl-Put, which is further processed to thededuced products Fl-Spd and Fl-Spm (31, 43). As shown in Fig.3, these various fluorinated and nonfluorinated polyamineswere clearly separable and quantifiable in the sameHPLC chro-matograph. Typically, each fluorinated polyamine analog runsjust ahead of its respective endogenous polyamine. The identityand quantitation of the fluorinated polyamines was based onthe relative association of the authentic Fl-Put standard withPut and previous reports by our laboratory and others showingthat 2-Fl-Put is metabolically converted to 6-Fl-Spd and 6-Fl-Spm (31, 43). Here, those findings were confirmed andextended to show that Fl-Spd and Fl-Spm are apparently acety-lated by SSAT and form products that similarly elute ahead ofthe endogenous acetylated polyamines. Interestingly, the intra-cellular fluorotagged polyamines were perceived as being sim-ilar to natural polyamine pools, since the individual polyaminepools remained relatively constant although the proportion offluorinated polyamines were freely exchanging and increasedsteadily with time. Because DAS lacks primary amines for bind-ingOPA reagent, it was assayed using aHPLCmethod based onprecolumn derivatization (22). Unfortunately, Fl-DAS andnative DAS co-eluted with this method and therefore could notbe separately quantified.Effect of Endogenous Orn on Fl-Orn Permeation—Intracellu-

lar Orn, a product of the citric acid cycle, is synthesized byarginase from the essential amino acid arginine (20). To assessthe extent to which endogenous Orn might compete with Fl-Orn as a substrate for ODC and hence mask flux, we made useof the inhibitor of arginase, nor-NOHA, to reduce endogenousOrn. In the presence of 100 �M nor-NOHA, each of the fluoro-tagged polyamine species examined under basal and inducedSSAT conditions was uniformly 2-fold greater than in theabsence of the inhibitor, whereas total pools sized remainedrelatively constant (data not shown). Thus, nor-NOHA effec-tively reduced Orn production and increased the ratio of

FIGURE 1. Effects of SSAT induction on cell growth. SSAT/LNCaP cells weregrown in the presence and absence of Tet for a total of 72 h. Cell growth (lefty axis) and SSAT activity (right y axis) were determined at 24, 48, and 72 h. Notethat the 70-fold SSAT induction is accompanied by growth inhibition. Theshaded bar drawn from 24 to 72 h indicates the period of time in subsequentexperiments when cells were treated with 300 �M Fl-Orn to monitor flux. Datarepresent average values � S.D. (n � 6).

TABLE 1Effects of SSAT induction on polyamine metabolism

SSAT/LNCaPEnzyme activitiesa Polyamine poolsb

SSAT ODC SAMDC Put Spd Spm AcSpdpmol/min/mg protein nmol/h/mg protein nmol/h/mg protein

BasalCells 23 � 5 3 � 1 3 � 2 0.06 � 0.09 1.38 � 0.31 4.26 � 0.63 0.13 � 0.23Medium 6.43 � 0.06 6.05 � 0.23 2.37 � 0.12 1.77 � 0.15

InducedCells 1165 � 270c 20 � 4c 60 � 1c 1.39 � 0.53d 1.46 � 0.40 2.99 � 0.51e 10.80 � 2.9dMedium 6.20 � 1.84 2.83 � 0.35d 0.06 � 0.08f 45.17 � 11.62e

a Data represent mean � S.D., where n � 6.b Intracellular pools, nmol/106 cells; extracellular (medium) pools, nmol/106 cell equivalents. Data represent mean � S.D., where n � 4.c p � 0.0001, Student’s t test, differences between basal and induced enzyme levels.d p � 0.001, Student’s t test, differences between basal and induced pools.e p � 0.01, Student’s t test, differences between basal and induced pools.f p � 0.0001, Student’s t test, differences between basal and induced pools.




http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

endogenous Fl-Orn/Orn to favor decarboxylation of Fl-Orn,thereby enhancing the rate of Fl permeation. Since Fl perme-ation of the SSAT/LNCaP cell polyamine pools was rapid and

substantial, the inhibitor was notroutinely used in the presentexperiments.Time-dependent Fluorine Unit

Permeation of Polyamine Pools—SSAT/LNCAP cells grown in thepresence (basal SSAT) or absence(induced SSAT) of Tet for 24 h andthen exposed to 300 �M Fl-Orn for48 h showed a time-dependentincrease in the newly synthesizedfluorinated polyamines that re-placed endogenous polyamines,whereas total pool sizes remainedrelatively constant. Thus, HPLCanalysis of Fl-Orn permeation ofpools allows us to distinguishnewly synthesized from existingpolyamine pools, thereby provid-ing an indication of the metabolicflux rate. Under basal SSAT condi-tions, Fl-Spd was apparent by 4 h,and Fl-Spm was apparent by 12 h,after which both increased

steadily. Under conditions of induced SSAT, far greateramounts of Fl-Spd were present at 4 h, and Fl-Spm appearedat 4 h instead of 12 h (Fig. 4). In fact, Fl-analogs were signif-icantly greater at every time point in SSAT-induced cellsrelative to cells with basal SSAT. By 48 h, fluorinated poly-amines comprised �25% of the total Spd and Spm poolsunder basal SSAT, whereas under induced SSAT, Fl-Spd andFl-Spm represented �50% of the Spd and Spm pools, respec-tively. Using fluorine tagging of Spm at 48 h as an indicator,the rate of metabolic flux for SSAT-induced cells was �4.5times faster than for the uninduced cells. We presume thatthis is due to increased acetylation of polyamines in conjunc-tion with increased biosynthesis of polyamines.Acetylated Polyamine Accumulation—A fixed time point

comparison of acetylated and nonacetylated polyamines wasconducted in cells under basal and induced conditions. Cellswere grown in the presence and absence of Tet for 24 h andthen exposed to 300 �M Fl-Orn or Orn for an additional 24 h.As shown in Fig. 5, individual polyamine levels were similarwhether cells were exposed to Orn or Fl-Orn, indicating thatthe latter was metabolically processed as efficiently as thenatural substrate Orn. Although acetylated products wererarely seen under basal conditions, they comprised themajority of polyamines under SSAT-induced conditions,with AcSpd being the main species. In fact, AcSpd accumu-lated to levels that were �10 times higher than those of Spd.This accumulation indicates that SSAT-induced cells areacetylating polyamines faster than they can be oxidized byPAO and/or exported out of the cell. The fact that �50% ofthe AcSpd is fluorine-labeled suggests that Fl-Spd is acety-lated as efficiently by SSAT as the endogenous Spd. Giventhe massive amounts of AcSpd that were generated in theinduced cells, it would seem that ODC and SAMDC feedback

FIGURE 2. Compensatory increases in polyamine biosynthetic enzymes and uptake following SSATinduction. SSAT/LNCaP cells were grown in the presence (; basal SSAT) and absence (�; induced SSAT) of Tetfor 48 h. A, ODC (left y axis), SAMDC (left y axis), and SSAT (right y axis) activities were determined, and -foldincreases shown above each set of bars represent the activity ratio of induced versus basal SSAT. B, Spd uptakewas determined by exposing SSAT/LNCaP cells with or without Tet to 2 and 10 �M [3H]Spd and quantifying theamount of intracellular radioactivity after a 30-min incubation at 37 °C. Data represent the average � S.D. ofthree separate assays performed in triplicate. -Fold increases are the average ratios of induced to basal activities.

FIGURE 3. Representative chromatogram of fluorotagged polyamines.A, polyamine pool analysis of SSAT/LNCaP cells incubated in the absence ofTet for 24 h and then exposed to 300 �M Fl-Orn for an additional 24 h. Post-column fluorescent derivatization of primary amines in acid-soluble cellextracts were detected by the HPLC methodology outlined under “Experi-mental Procedures.” B, polyamine standard mix containing authentic Fl-Putfor verification. The identities of fluorinated compounds (Fl-Spd, Fl-Spm, Fl-AcSpd, Fl-AcSpm) were presumed from peaks with retention times �1 minahead of the parent natural polyamine. Production of DAS, not detectable bythis method, was determined using precolumn dansylation methodologyoutlined under “Experimental Procedures.”




http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

regulation is insensitive to acetylated polyamines, since bothenzymes remain at high levels (Table 1).Time-dependent Polyamine Acetylation—Having shown that

induced SSAT cells generate acetylated polyamines whereasbasal SSAT cells do not, we next examined the time-dependentfluorine unit permeation of acetylated polyamine pools. SSAT/LNCaP cells were grown in the absence of Tet for 24 h and thenexposed to 300 �M Fl-Orn for 0–48 h. Due to SSAT inductionduring the Tet preincubation, cells contained high levels ofunlabeled AcSpd pools at the initiation of the Fl-Orn incuba-tion. Thereafter, there is a steady fluorine unit permeation ofboth the intracellular Spd andAcSpd pools that continuedwithtime (Fig. 6A). Although not present in comparable amounts,Spm, AcSpm, and DAS were also fluorine-labeled in a time-de-pendent manner, but the acetylated Spm products failed toaccumulate to levels as high as those of AcSpd (Fig. 6A). Atevery time point, the ratio of AcSpd/Spd is �5, whereas theratio of AcSpm/Spm is�0.25. Since intracellular Spmpools are2–3 times higher than Spd and since SSAT has a lower Km forSpm than Spd (32, 39), these ratios are opposite to expectations.One explanation is that the endogenous Spmpool is bound (32)or sequestered in the nucleus (or elsewhere) and hence lessavailable than Spd for SSAT acetylation in the cytosol.

Medium collected from the aforementioned experimentswas analyzed by HPLC (Fig. 6B). By far, the most abundantextracellular polyamine was AcSpd. Accumulation of extracel-lular AcSpd began immediately following the medium changebetween the pre- and postincubations and rose to �50 nmol/106 cell equivalents over the 48-h postincubation. By compari-son, only minor amounts of acetylated Spm products appearedin the medium. Nonacetylated Fl-Spd and Fl-Spm alsoappeared in the medium, whereas Spd and Spm did not. Thissuggests that newly synthesized Fl-Spd and Fl-Spm ismore sus-ceptible to export than preexisting Spd and Spm.Role of PAO Inhibition in Flux—The above data are con-

sistent with the idea that acetylation of Spd and its excretionout of the cell may be responsible for initiating the height-ened metabolic flux that was experimentally shown to takeplace. It is also possible that a significant portion of AcSpdand AcSpm may be oxidized to Put, and this could help todrive flux. In order to examine the possible contribution ofpolyamine oxidation in perpetuating flux, SSAT-inducedcells were exposed to a 50 �M concentration of the PAOinhibitorMDL-72527 during the 24-h postincubation periodwith the expectation that if acetylated polyamines werebeing oxidized to a significant degree, the level of intracellu-lar and extracellular acetylated polyamines would increase(44). As shown in Fig. 7, intracellular AcSpd levels were sim-

FIGURE 4. Effects of SSAT induction on metabolic flux, as indicated byfluorine permeation of polyamine pools. SSAT/LNCaP were grown in thepresence and absence of Tet for 24 h and then exposed to 300 �M Fl-Orn for 0,4, 8, 12, 24, and 48 h, after which the relative levels of Put, Fl-Put, Spd, Fl-Spd,Spm, and Fl-Spm were determined. The stacked bars represent total poly-amine pool, depicted as fluorinated polyamine (hatched upper) and nativepolyamine (solid lower) pool. It is relevant that the basal SSAT cells continue todivide, whereas the induced SSAT cells do not (see Fig. 1). Data represent theaverage values (n � 3) with S.D. � 15%.

FIGURE 5. Fixed time point comparison of polyamine pools in SSAT/LN-CaP cells exposed to Orn or Fl-Orn. SSAT/LNCaP cells were grown in thepresence and absence of 300 �M Tet for 24 h and then exposed to either 300�M Orn or 300 �M Fl-Orn for 24 h prior to analyzing polyamine pool content.The data depict the total level of each polyamine with Orn treatment (solidbars) graphed beside Fl-Orn treatment (endogenous polyamine (solid) andFl-analog (hatched), stacked bars) for comparison. Note that the individualpolyamine levels are relatively constant for Orn and Fl-Orn-exposed cells.Data represent the average of three separate determinations with S.D. �15%.




http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

ilar in the presence or absence of MDL-72527, and extracel-lular AcSpd actually decreased with MDL-72527. Minorincreases were observed for AcSpm and DAS, suggestingsome oxidation and back-conversion of these products toSpd and AcSpd, respectively. However, since AcSpm andDAS represent only a small portion of the total SSAT-acety-lated products, they would seem at best to be minor contrib-utors to the heightened flux. Instead, the data indicate thatacetylation functions independently of oxidation to initiatethe flux cycle. Nearly identical results were obtained if cellswere exposed to Fl-Orn during MDL-72527 treatment (datanot shown).Effects of Biosynthetic Enzyme Inhibition on Flux—As shown

above, SSAT induction induces flux, as indicated by theincreased rate of Fl permeation of polyamine pools. For fluxto be sustained, new polyamines must be synthesized, takenup, or both. To examine the contribution of biosynthesis,SSAT-induced cells were exposed for 48 h to 300 �M Fl-Ornin the absence and presence of the SAMDC inhibitor MDL-

73811. The more commonly usedODC inhibitor, �-difluoromethylo-rnithine, was not suitable for theseexperiments, since the high levels ofFl-Orn would be expected to com-pete with it for uptake and enzymebinding. In cells not treated withMDL-73811 (Table 2), the levels ofFl-Spd, Fl-Spm, and Fl-AcSpd areabout equal (�1:1 ratio) to nativepolyamine levels.MDL-73811 treat-ment blocked synthesis of thesethree fluorotagged polyamines,whereas the pools of native Spd andSpm decreased by half. This con-firms (a) that new synthesis isrequired for flux to be sustained and(b) that the fluorotagged poolsderive from newly synthesized Spdand Spm. In addition, we noted thatsignificantly less AcSpd was pro-duced and exported in the treatedcells, indicating that these processesare also dependent on increasedbiosynthetic flux.Treatment of cells with SAMDC

inhibitors typically increases Putsignificantly due to a compensatoryincrease in ODC (45, 46) and theinability of the enzyme product Putto be processed forward to Spd dueto SAMDC inhibition. Thus, MDL-78311 treatment gave rise to intra-and extracellular accumulation ofboth Put and Fl-Put. Although thenative Put seemed to be partitionedevenly between the cells andmedium, most of the Fl-Put wasrapidly exported into the medium,

reaching levels 2.5-fold those of extracellular Put.Effects of Flux on Coenzymes A—Wepreviously reported (23)

that SSAT induction in the SSAT/LNCaP cells depletes theSSAT cofactor, acetyl-CoA pools. Since the downstreammetabolite of acetyl-CoA, malonyl-CoA, is known to potentlyregulate fatty acid oxidation (47), wemeasured both acetyl- andmalonyl-CoApools at 48 and 72 h followingTet removal. Thesetime points were chosen to coordinate with fluxmeasurementsindicated in Fig. 1. Relative to basal SSAT cells, we found sig-nificant time-dependent decreases in both CoA molecules,such that 72 h after SSAT induction, acetyl-CoA pools werereduced from 17 � 2 to 8.7 � 1.2 nmol/106 cells (50%), andmalonyl-CoA pools were reduced from 35.3� 2.8 to 12.0� 1.6nmol/106 cells (65%).

DISCUSSION

Except for a 1995 publication from this laboratory (31), littleattention has been given in the literature to the issue of poly-amine metabolic flux. The hypothesis being tested here is that

FIGURE 6. Time course analysis of SSAT effects on metabolic flux, as indicated by fluorine permeationof intracellular (A) and extracellular (B) polyamines. Following 24 h in the absence of Tet (inducedSSAT), SSAT/LNCaP cells were placed in fresh medium and exposed to 300 �M Fl-Orn from 0 to 48 h.Medium and cells were collected for polyamine analysis. A, comparison of intracellular Spd and AcSpd(left) and intracellular Spm, AcSpm and DAS (right) showing time-dependent fluorine labeling. The stackedbars depict levels of natural polyamine (lower) and Fl-polyamine (upper) and their acetylated products.Data expressed as nmol/106 cells represent four separate determinations with S.D. � 15%. B, comparisonof extracellular Spd and AcSpd (left) and extracellular Spm, AcSpm, and DAS showing time-dependentlabeling with fluorine. Stacked bars depict the natural and the acetylated Spd (left) and Spm (right) prod-ucts and the Fl-polyamines (hatched upper) for both. Note that the major products exported into themedium with SSAT activation are AcSpd and Fl-AcSpd. Data expressed as nmol/106 cell equivalentsrepresent the average of four separate determinations with S.D. � 15%.




http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

induction of SSAT gives rise to heightened and sustained met-abolic flux through both the biosynthetic and acetylation armsof the polyamine pathway. Prior to this study, the notion thatinduced SSAT might lead to heightened metabolic flux wasdeduced on the basis of whatwe nowdesignate as four hallmarkobservations: (a) increased SSAT activity, (b) a rise in biosyn-thetic enzyme activities, and (c) persistence of polyamine poolsdespite (d) accumulation of intracellular and extracellularacetylated polyamines. In the face of these events, the mostcompelling conclusion is increased metabolic flux through thepathway. Despite the logic of this deduction, this concept hasnot yet been experimentally confirmed. Thus, wemeasured flu-orine permeation of polyamine pools in basal SSAT versusinduced SSAT cells incubated for various times in the presenceof Fl-Orn. As shown in Fig. 4, the rate of Fl-Spm accumulationin cells growing under basal conditions was increased �5-foldby induction of SSAT. In actuality, this value markedly under-estimates the rate of flux. For example, it does not account forthe huge amounts of intracellular and extracellular AcSpd thatare essentially lost from the forward-processing pathway toSpm. The combined amount of intracellular and extracellularAcSpd (�45 nmol) generated during the first 24 h was 22 timesthat of the intracellular Spd levels (�2 nmol). Since Spd pools

remained relatively constant over this period and since theamount of AcSpd generated by uninduced cells was insignifi-cant, by this analysis flux could have increased by �20-fold inSSAT-induced cells.The basis for this increase in flux in induced SSAT cells is

initiated by increased acetylation of polyamines. This is thenperpetuated by an increase in the biosynthetic enzymes ODCand SAMDC. These enzymes are known to be negativelyrepressed at the level of protein translation and degradation bybasal intracellular levels of polyamines (12, 48–50). We believethat acetylation of Spm and particularly Spd following SSATinduction relieves enzyme repression in some manner, givingrise to an overall increase in biosynthetic capabilities. Thus,polyamines lost to acetylation (and eventually export) are rap-idly replaced by biosynthesis and made available for continuedrecycling via acetylation. Under conditions of heightened met-abolic flux, sustained acetylation of these polyamines gives riseto a new steady state in which ODC and SAMDC activities arehomeostatically set at higher levels so that intracellular poly-amine pools are maintained despite massive acetylation andexport. The critical role of biosynthesis in sustaining flux wasclearly demonstrated by the ability of SAMDC inhibition tointerrupt flux (see Table 2). Thus, although flux is initiated byinduction of SSAT, it is critically sustained by increases in bio-synthetic enzyme activity for as long as SSAT remains overex-pressed. This sequence of events is diagrammed in Fig. 8.The basis for derepression of ODC and SAMDC following

SSAT induction is unclear. Although there is amodest decreasein Spm, we have previously reported that Spd and Spm poolsactually increase during the same timewhenODCand SAMDCare increasing (23). It seems likely that repression is controlledby a relatively small unbound portion of the total intracellularpolyamine pool, and acetylation could affect polyamine bindingby reducing the net molecular charge (32). More certainly,AcSpd is incapable of substituting for Spd in repressing thebiosynthetic enzymes, since huge amounts accumulate withinthe cell (Table 1 and Figs. 5 and 6).Our findings also show that in addition to derepressingODC

and SAMDC, acetylation of intracellular polyamines increasespolyamine transport. Although the latter is known to increasein response to polyamine pool depletion by biosyntheticenzyme inhibitors (41, 51, 52), it has not been shown to occur inresponse to SSAT induction. In both cases, the response is con-sistent with an attempt by the system to homeostatically nor-malize polyamine imbalances by importing extracellular poly-amines. Up-regulation of transport by SSAT could have

FIGURE 7. Effects of PAO inhibition on accumulation and export of acety-lated polyamines. SSAT/LNCaP cells were grown in the absence of Tet for24 h and then exposed for an additional 24 h to the presence (72527) orabsence (�72527) of the PAO inhibitor MDL-72527 at 50 �M. The cells andmedium were processed to determine intracellular (left) and extracellular(right) levels of AcSpd, AcSpm, and DAS. Note that PAO inhibition failed toincrease intracellular accumulation or export of acetylated Spd or Spm. Datarepresent average values (n � 3) with S.D. � 10%.

TABLE 2Interruption of polyamine flux by the SAMDC inhibitor, MDL-73811

Induced SSAT/LNCaP cellsaPolyamine poolsb

Put Fl-Put Spd Fl-Spd Spm Fl-Spm AcSpd Fl-AcSpdControlCells �0.05 �0.05 0.7 0.8 2.0 2.8 3.5 3.8Medium 2.2 3.1 0.1 1.2 �0.05 �0.05 17.2 13.1

MDL-RxCells 18.6 2.4 0.6 �0.05 1.7 �0.05 0.2 �0.05Medium 26.5 67.4 0.6 �0.05 �0.05 �0.05 6.1 �0.05

a Following 24-h Tet removal, cells were treated for 48 h with 300 �M Fl-Orn with or without 30 �M MDL-73811 (MDL-Rx). Data represent the average of two experimentsperformed in duplicate with S.D. � 15%.

b Intracellular pools, nmol/106 cells; extracellular (medium) pools, nmol/106 cell equivalents.




http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

significant implications in vivo, where, in the case of enzymeinhibitors, salvage of exogenous polyamines is known to com-promise antitumor effects (10, 11, 53).We also addressed the possible relationship of polyamine

oxidation to SSAT-induced flux. The SSAT products, AcSpdand AcSpm, are preferred substrates of PAO, which convertsthem to Put and Spd, respectively, via the back-conversionpathway (20, 32). By making these products readily available,SSAT could contribute to flux by increasing their oxidation.Using the PAO inhibitor, MDL-72527, we expected intracellu-lar acetylated polyamines to increase if they were being oxi-dized (44), but they did not. This confirms that flux is initiatedby polyamine acetylation and not by polyamine oxidation.It is informative to consider the potential of SSAT as a mod-

ulator of metabolic flux relative to other polyamine enzymes.Intuitively, it would seem that as key regulators of polyaminebiosynthesis, ODC and SAMDC might be better suited as reg-ulators of flux. However, they are not, because, as noted earlier,both are repressed at the levels of protein translation and deg-radation by intracellular Spd and Spm levels (12, 48). Thus, iftheir expression were up-regulated, enzyme activity (andhence, flux) would be rapidly repressed by rising levels of Spdand Spm. This does not preclude the possibility that forcedexpression of ODC, particularly if it is rendered stable to deg-radation (54), can induce flux.Polyamines have been traditionally thought to represent

functional entities and the processes of biosynthesis, catabo-lism, and transport as means to homeostatically control and

maintain their intracellular concentrations. Although the find-ings here do not dispute this view, several recent studies haveprovided evidence to indicate that SSAT-induced changes inmetabolic flux appear to have much broader metabolic conse-quences, particularly in vivo. For example, the futile metaboliccycling initiated by SSAT and sustained by increases in poly-amine biosynthesis can deplete pathway precursors, such asornithine,methionine, S-adenosylmethionine, and acetyl-CoA.Alternatively, it can lead to overproduction of toxic pathwayby-products, such as 5�-methylthioadenosine, hydrogen perox-ide, and aldehydes. These various possibilities have beenrecently identified as “enhanced flux” by Kee et al. (23, 27), as a“metabolic ratchet” by Tucker et al. (28), and as a “paddle-wheel” by Janne et al. (55).

Several independent pieces of evidence support the ideathat heightened polyamine flux leads to deprivation of theSSAT co-enzyme and fatty acid precursor acetyl-CoA inboth cultured cells and mice. In summary, the evidence is asfollows: (a) conditional overexpression of SSAT in LNCaPcells leads to a significant decrease in acetyl- and malonyl-CoA pools (23); (b) transgenic overexpression of SSAT inTRAMP mice is accompanied by a significant decrease inacetyl-CoA in prostatic tumors (27); (c) transgenic overex-pression of SSAT in C57Bl/6 mice depletes acetyl- and mal-onyl-CoA pools in adipose tissue, increases fatty acid oxida-tion, and results in a distinctly lean phenotype (30); and last,(d) genetic deletion of SSAT in C57Bl/6 mice (SSAT knock-out mice) leads to an accumulation of acetyl- and malonyl-CoA, decreased fatty acid oxidation, and a fat-prone pheno-type, particularly on a high fat diet (30). The importance ofthe malonyl-CoA finding in both SSAT transgenic and SSATknock-out mice is that it is synthesized from acetyl-CoA byacetyl-CoA carboxylase, and it is known to regulate �-oxida-tion of fatty acids at the level of carnitine palmitoyltrans-ferase 1 (47). Thus, a decrease in malonyl-CoA in adiposetissue is consistent with increased fatty oxidation and a leanphenotype in SSAT transgenic mice. Coincidentally, theSSAT transgenic phenotype is nearly identical to that ofmice lacking the enzyme responsible for malonyl-CoA syn-thesis, acetyl-CoA carboxylase-2, and they too have reducedlevels of malonyl-CoA in the adipose tissue (56) andincreased fatty acid oxidation in isolated adipocytes (57).Finally, the probable relationship between malonyl-CoAdepletion and growth inhibition has been recently rein-forced by studies showing that small molecule inhibition ofacetyl-CoA carboxylase increases fatty acid oxidation andinhibits growth of LNCaP cells (58).Although the findings with SSAT overexpression in cells

and mice represent an exaggeration of normalcy that may ormay not be physiologically relevant, the finding that SSATknock-out mice are fat-prone has clearer physiologicalimplications. In particular, it implies that in normal cells andtissues of wild type mice and at physiological enzyme levels,SSAT prevents fat accumulation. Taken together, our pastand present findings strongly relate polyamine acetylation tofatty acid metabolism via acetyl- and malonyl-CoA. Weemphasize that the metabolic consequences of increasedpolyamine flux may vary according to cells and tissues and

FIGURE 8. SSAT-induced increases in polyamine metabolic flux. Poly-amine biosynthesis is regulated by ODC and SAMDC, which generate thepolyamine Put and decarboxylated S-adenosylmethionine (dcSAM), respec-tively. The latter is used as an aminopropyl donor by spermidine synthase(Spd Syn) to form Spd and by spermine synthase (Spm Syn) to form Spm. Thisforward synthesis can be reversed by a back-conversion pathway in whichSpm and Spd are acetylated by SSAT to form AcSpm and AcSpd, which arethen oxidized by PAO to form Spd and Put, respectively. A major portion ofAcSpd is exported out of the cell. Data presented here suggest that the fol-lowing sequence of events mediates SSAT-induced metabolic flux: SSAT isinduced (1); Spm and particularly Spd are acetylated (2); ODC and SAMDCactivities increase (3); new polyamines are synthesized and become availablefor continued acetylation (4), and heightened metabolic flux (large arrow) isestablished for as long as SSAT remains induced (5). This heightened meta-bolic flux is initiated by SSAT acetylation of polyamines and sustained by ODCand SAMDC regulation of new polyamine biosynthesis. The end result ofthese events is futile metabolic cycling, which has untoward consequencesfor the cell. More specifically, it can cause depletion of pathway precursors,such as S-adenosylmethionine (SAM) or acetyl-CoA, or accumulation ofpotentially toxic by-products, such as 5�-methylthioadenosine (MTA). This fig-ure was adapted from Jell et al. (30).




http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

that in some contexts, other metabolic and cellular conse-quences may obtain. As an example, we have reported thatglobal overexpression of SSAT suppresses growth of prostatetumors (23) and enhances growth of intestinal tumors (28),although both effects appear to be due to increased flux.Thus, although acetyl-CoA depletion appears to be involvedin the former, alternative metabolic disturbances are likelyto be responsible for the latter. Similarly, a recent studyfound that selective SSAT expression in mouse skinincreased sensitivity to chemical carcinogens due to eleva-tions in ODC activity and Put levels, both indicative ofincreased flux with increased liberation of toxic metabolicby-products (59).In summary, this study provides direct evidence for the pre-

mise that conditional overexpression of SSAT leads to height-ened metabolic flux through the polyamine pathway. It alsoconfirms the validity of (a) increased SSAT activity, (b) a rise inbiosynthetic enzyme activities, and (c) persistence of polyaminepools despite (d) accumulation of intracellular and extracellularacetylated polyamines as hallmark indicators of metabolic flux.For now, the broader metabolic consequences of heightenedpolyamine flux are best characterized by effects on the SSAT-coenzyme and fatty acid precursor, acetyl-CoA (30) in associa-tion with profound phenotypic effects involving fat accumula-tion in mice. We point out that these consequences areprobably context-dependent, and alternativemetabolic pertur-bations are likely to be revealed in other cells and tissues. Ineither case, it is significant that SSAT-induced changes in poly-amine flux can impact other metabolic pathways, such as fatmetabolism.REFERENCES1. Porter, C. W., and Bergeron, R. J. (1988) Adv. Enzyme Regul. 27, 57–792. Porter, C. W., and Bergeron, R. J. (1988) Adv. Exp. Med. Biol. 250,

677–6903. Shappell, N.W., Fogel-Petrovic,M. F., and Porter, C.W. (1993) FEBS Lett.

321, 179–1834. Pegg, A. E. (1988) Cancer Res. 48, 759–7745. Janne, J., Poso, H., and Raina, A. (1978) Biochim. Biophys. Acta 473,

241–2936. Seiler, N. (2003) Curr. Drug. Targets 4, 537–5647. Kramer, D. L., and Gerner, E. W. (2004) in Cancer Chemoprevention

(Kelloff, G. J., Hawk, E. T., and Sigman, C. C., eds) Vol. 1, pp. 339–357,Humana Press, Inc., Totowa, NJ

8. Casero, R. A., Jr., and Marton, L. J. (2007) Nat. Rev. Drug. Discov. 6,373–390

9. Persson, L., Holm, I., Ask, A., and Heby, O. (1988) Cancer Res. 48,4807–4811

10. Sarhan, S., Knodgen, B., and Seiler, N. (1989) Anticancer Res. 9, 215–22311. Moulinoux, J. P., Quemener, V., Cipolla, B., Guille, F., Havouis, R., Martin,

C., Lobel, B., and Seiler, N. (1991) J. Urol. 146, 1408–141212. Porter, C.W., Berger, F. G., Pegg, A. E., Ganis, B., and Bergeron, R. J. (1987)

Biochem. J. 242, 433–440.13. Porter, C.W.,McManis, J., Casero, R. A., and Bergeron, R. J. (1987)Cancer

Res. 47, 2821–2825.14. Porter, C.W., Ganis, B., Libby, P. R., and Bergeron, R. J. (1991)Cancer Res.

51, 3715–372015. Casero, R. A., Jr., Celano, P., Ervin, S. J., Porter, C. W., Bergeron, R. J., and

Libby, P. R. (1989) Cancer Res. 49, 3829–383316. Wallace, H. M., Macgowan, S. H., and Keir, H. M. (1985) Biochem. Soc.

Trans. 13, 329–33017. Seiler, N. (2003) Curr. Drug Targets 4, 565–58518. Chen, Y., Kramer, D. L., Jell, J., Vujcic, S., and Porter, C. W. (2003) Mol.

Pharmacol. 64, 1153–115919. Chen, Y., Kramer, D. L., Diegelman, P., Vujcic, S., and Porter, C.W. (2001)

Cancer Res. 61, 6437–644420. Seiler, N. (2004) Amino Acids 26, 217–23321. Wallace, H. M., Nuttall, M. E., and Coleman, C. S. (1988) Adv. Exp. Med.

Biol. 250, 331–34422. Vujcic, S., Halmekyto, M., Diegelman, P., Gan, G., Kramer, D. L., Janne, J.,

and Porter, C. W. (2000) J. Biol. Chem. 275, 38319–3832823. Kee, K., Vujcic, S., Merali, S., Diegelman, P., Kisiel, N., Powell, C. T.,

Kramer, D. L., and Porter, C. W. (2004) J. Biol. Chem. 279, 27050–2705824. McCloskey, D. E., Coleman, C. S., and Pegg, A. E. (1999) J. Biol. Chem. 274,

6175–618225. Alhonen, L., Karppinen, A., Uusi-Oukari, M., Vujcic, S., Korhonen, V. P.,

Halmekyto,M., Kramer, D. L., Hines, R., Janne, J., and Porter, C.W. (1998)J. Biol. Chem. 273, 1964–1969

26. Pietila, M., Alhonen, L., Halmekyto, M., Kanter, P., Janne, J., and Porter,C. W. (1997) J. Biol. Chem. 272, 18746–18751

27. Kee, K., Foster, B. A., Merali, S., Kramer, D. L., Hensen, M. L., Diegelman,P., Kisiel, N., Vujcic, S., Mazurchuk, R. V., and Porter, C. W. (2004) J. Biol.Chem. 279, 40076–40083

28. Tucker, J. M., Murphy, J. T., Kisiel, N., Diegelman, P., Barbour, K. W.,Davis, C., Medda, M., Alhonen, L., Janne, J., Kramer, D. L., Porter, C. W.,and Berger, F. G. (2005) Cancer Res. 65, 5390–5398

29. Pirinen, E., Kuulasmaa, T., Pietila, M., Heikkinen, S., Tusa,M., Itkonen, P.,Boman, S., Skommer, J., Virkamaki, A., Hohtola, E., Kettunen, M., Fatrai,S., Kansanen, E., Koota, S., Niiranen, K., Parkkinen, J., Levonen, A. L.,Yla-Herttuala, S., Hiltunen, J. K., Alhonen, L., Smith, U., Janne, J., andLaakso, M. (2007)Mol. Cell. Biol. 27, 4953–4967

30. Jell, J., Merali, S., Hensen, M. L., Mazurchuk, R., Spernyak, J. A., Di-egelman, P., Kisiel, N. D., Barrero, C., Deeb, K. K., Alhonen, L., Patel,M. S.,and Porter, C. W. (2007) J. Biol. Chem. 282, 8404–8413

31. Kramer, D., Stanek, J., Diegelman, P., Regenass, U., Schneider, P., andPorter, C. W. (1995) Biochem. Pharmacol. 50, 1433–1443

32. Seiler, N. (1987) Can. J. Physiol. Pharmacol. 65, 2024–203533. Casero, R. A., Jr., and Pegg, A. E. (1993) FASEB J. 7, 653–66134. Hector, S., Porter, C. W., Kramer, D. L., Clark, K., Prey, J., Kisiel, N.,

Diegelman, P., Chen, Y., and Pendyala, L. (2004) Mol. Cancer Ther. 3,813–822

35. Gschwend, J. E., Fair,W. R., and Powell, C. T. (1997) Prostate 33, 166–17636. Seiler, N., and Knodgen, B. (1980) J. Chromatogr. 221, 227–23537. Kabra, P. M., Lee, H. K., Lubich, W. P., and Marton, L. J. (1986) J. Chro-

matogr. 380, 19–3238. Kramer, D., Mett, H., Evans, A., Regenass, U., Diegelman, P., and Porter,

C. W. (1995) J. Biol. Chem. 270, 2124–213239. Libby, P. R., Ganis, B., Bergeron, R. J., and Porter, C. W. (1991) Arch.

Biochem. Biophys. 284, 238–24440. Porter, C. W., Cavanaugh, P. F., Jr., Stolowich, N., Ganis, B., Kelly, E., and

Bergeron, R. J. (1985) Cancer Res. 45, 2050–205741. Kramer, D. L., Miller, J. T., Bergeron, R. J., Khomutov, R., Khomutov, A.,

and Porter, C. W. (1993) J. Cell. Physiol. 155, 399–40742. Clark, J. L., and Fuller, J. L. (1975) Biochemistry 14, 4403–440943. Dezeure, F., Sarhan, S., and Seiler, N. (1988) Int. J. Biochem. 20,

1299–131244. Seiler, N., Duranton, B., and Raul, F. (2002) Prog. Drug Res. 59, 1–4045. Madhubala, R., Secrist, J. A., III, and Pegg, A. E. (1988) Biochem. J. 254,

45–5046. Kramer,D. L., Khomutov, R.M., Bukin, Y. V., Khomutov,A. R., and Porter,

C. W. (1989) Biochem. J. 259, 325–33147. McGarry, J. D., and Brown, N. F. (1997) Eur. J. Biochem. 244, 1–1448. Kameji, T., and Pegg, A. E. (1987) J. Biol. Chem. 262, 2427–243049. Fogel-Petrovic, M., Vujcic, S., Miller, J., and Porter, C. W. (1996) FEBS

Lett. 391, 89–9450. Porter, C. W., Pegg, A. E., Ganis, B., Madhabala, R., and Bergeron, R. J.

(1990) Biochem. J. 268, 207–21251. Kakinuma, Y., Hoshino, K., and Igarashi, K. (1988) Eur. J. Biochem. 176,

409–41452. Byers, T. L., and Pegg, A. E. (1990) J. Cell. Physiol. 143, 460–46753. Ask, A., Persson, L., and Heby, O. (1992) Cancer Lett. 66, 29–34




http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

54. Ghoda, L., van Daalen Wetters, T., Macrae, M., Ascherman, D., andCoffino, P. (1989) Science 243, 1493–1495

55. Janne, J., Alhonen, L., Pietila, M., Keinanen, T. A., Uimari, A., Hyvonen,M. T., Pirinen, E., and Jarvinen, A. (2006) J Biochem. (Tokyo) 139,155–160

56. Abu-Elheiga, L., Matzuk, M. M., Abo-Hashema, K. A., and Wakil, S. J.(2001) Science 291, 2613–2616

57. Oh, W., Abu-Elheiga, L., Kordari, P., Gu, Z., Shaikenov, T., Chirala, S. S.,and Wakil, S. J. (2005) Proc. Natl. Acad. Sci. U. S. A. 102, 1384–1389

58. Beckers, A., Organe, S., Timmermans, L., Scheys, K., Peeters, A., Brussel-mans, K., Verhoeven, G., and Swinnen, J. V. (2007) Cancer Res. 67,8180–8187

59. Wang, X., Feith, D. J., Welsh, P., Coleman, C. S., Lopez, C., Woster, P. M.,O’Brien T, G., and Pegg, A. E. (2007) Carcinogenesis 28, 2404–2411




http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Carl W. PorterDebora L. Kramer, Paula Diegelman, Jason Jell, Slavoljub Vujcic, Salim Merali and

Metabolic ConsequencesPolyamine Acetylation Modulates Polyamine Metabolic Flux, a Prelude to Broader

doi: 10.1074/jbc.M706806200 originally published online December 18, 20072008, 283:4241-4251.J. Biol. Chem.

10.1074/jbc.M706806200Access the most updated version of this article at doi:

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here

http://www.jbc.org/content/283/7/4241.full.html#ref-list-1

This article cites 59 references, 32 of which can be accessed free at


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/lookup/doi/10.1074/jbc.M706806200

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;283/7/4241&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/283/7/4241

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=283/7/4241&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/283/7/4241

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/283/7/4241.full.html#ref-list-1

http://www.jbc.org/

Polyamine Acetylation Modulates Polyamine Metabolic Flux, a Prelude to Broader Metabolic Consequences

Documents