Transient Exposure of Neonatal Female Mice to Testosterone Abrogates the Sexual Dimorphism of Abdominal Aortic Aneurysms

Transient Exposure of Neonatal Female Mice to TestosteroneAbrogates the Sexual Dimorphism of Abdominal AorticAneurysms

Xuan Zhang1, Sean E. Thatcher2, Debra L. Rateri3, Dennis Bruemmer2,3, RichardCharnigo4, Alan Daugherty1,2,3, and Lisa A. Cassis1,2

1Graduate Center for Toxicology, University of Kentucky, Lexington, KY, USA2Graduate Center for Nutritional Sciences, University of Kentucky, Lexington, KY, USA3Saha Cardiovascular Research Center, University of Kentucky, Lexington, KY, USA4Department of Statistics, University of Kentucky, Lexington, KY, USA

AbstractRationale—Abdominal aortic aneurysms (AAAs) exhibit marked sexual dimorphism with higherprevalences in men. Similarly, AAAs induced by angiotensin II (AngII) infusion into mice exhibita higher prevalence in males. Testosterone promotes AAA pathology in adult male mice throughregulation of angiotensin type 1A receptors (AT1aR) in abdominal aortas. However, mechanismsfor sexual dimorphism of regional aortic angiotensin receptor expression and AAA formation areunknown.

Objective—To define the role of developmental testosterone exposures in sexual dimorphism ofAAAs, we determined if exposure of neonatal female mice to testosterone confers adultsusceptibility to AngII-induced AAAs.

Methods and Results—One day old female hypercholesterolemic mice were administered asingle dose of either vehicle or testosterone. Neonatal testosterone administration increasedabdominal aortic AT1aR mRNA abundance and promoted a striking increase in AngII-inducedAAAs in adult females exhibiting low serum testosterone concentrations. AngII-inducedatherosclerosis and ascending aortic aneurysms were also increased by testosterone administrationto neonatal females. In contrast, neonatal testosterone administration in males had no effect onAngII-induced vascular pathologies. Deficiency of AT1aR in smooth muscle cells (SMCs)reduced effects of neonatal testosterone to promote AAAs in adult females, but did not alteratherosclerosis or ascending aortic aneurysms. Testosterone increased AT1aR mRNA abundanceand hydrogen peroxide generation in cultured abdominal aortic SMCs. Increased AT1aR mRNAabundance was maintained during progressive passaging of female SMCs.

Conclusions—These data reveal an unrecognized role of transient sex hormone exposuresduring neonatal development as long-lasting mediators of regional aortic AT1aR expression andsexual dimorphism of AAAs.

Corresponding author: Lisa Cassis, Ph.D., Professor and Chair, Graduate Center for Nutritional Sciences, University of Kentucky,Room 521b Wethington Building, 900 S. Limestone, Lexington, KY 40536-0200, Tel: (859)-323-4933 ext 81400, Fax:(859)-257-3646, [email protected].

DisclosuresNone.

NIH Public AccessAuthor ManuscriptCirc Res. Author manuscript; available in PMC 2013 May 25.

Published in final edited form as:Circ Res. 2012 May 25; 110(11): e73–e85. doi:10.1161/CIRCRESAHA.111.253880.

$waterm

ark-text$w

atermark-text

$waterm

ark-text

Keywordsaneurysm; sexual dimorphism; testosterone; vascular smooth muscle; angiotensin

IntroductionAbdominal aortic aneurysms (AAA) are permanent dilations of the abdominal aorta with an85% chance of death after rupture.1-3 Recent estimates of AAA prevalence are 1.1 millionpeople in the US aged 50 to 84.4 As the population lives longer, it is anticipated that AAAprevalence will increase, which is of concern since there is no proven medical therapy tochange aneurysm growth or rupture. A variety of risk factors influence AAA formation andprogression including age,5, 6 smoking,7-9 obesity,10, 11 and gender.12, 13 Several studieshave demonstrated that male sex is a prominent risk factor for human AAAs,6, 14, 15 withrecent results from the Tromsǿ study estimating that males are 2.66 times as likely asfemales to develop AAAs.15 Despite a strong influence of male sex on AAA formation,mechanisms contributing to increased AAA formation in males are unclear.

Angiotensin II (AngII) infusion into hypercholesterolemic mice increased atherosclerosisand induced formation of aneurysms in both the ascending and suprarenal aortas of malemice.16-18 Notably, the renin-angiotensin system has been clinically implicated in theetiology of atherosclerosis, ascending aortic aneurysms, and AAAs.19, 20 In mice, theseAngII-induced vascular pathologies are mediated through angiotensin type 1a receptors(AT1aR).21-23 While each of these vascular pathologies are induced by infusion of AngII,only AAAs exhibit marked sexual dimorphism with a higher prevalence (4-fold) in malecompared to female mice.16, 17, 24, 25 Testosterone was determined to be a primary mediatorof higher AAA prevalence in adult AngII-infused male mice.25, 26 As a mechanistic target oftestosterone to promote AngII-induced AAAs, AT1aR mRNA abundance was greater inabdominal compared to thoracic aortas of male, but not female mice. This regionaldifference was abolished by orchiectomy and restored in castrated males by androgenadministration.26 However, the mechanistic contributions of the renin-angiotensin system tothe sexual dimorphism of AAA formation have not been defined.

Sexual dimorphism has been studied extensively in several organ systems. For example,sexual dimorphism of the mammalian brain has been examined by exposing females totestosterone during the neonatal period.27 This androgenization imposes upon femalestransient exposures to testosterone shortly after birth, and mimics developmental effects ofandrogen in males during the neonatal period.28, 29 Sex hormone differences between adultmales and females have been implicated as mediators of sexual dimorphism of severalcardiovascular diseases. However, the role of sex hormone exposures during neonataldevelopment on vascular diseases that display pronounced sexual dimorphism, such asAAAs, has not been defined. Moreover, mechanisms for region-specific regulation ofAT1aR expression by testosterone in different aortic regions are unknown, but may relate toembryonic origins of vascular wall cells and epigenetic influences of sex hormone exposuresduring critical periods of development. In this study, we sought to quantify effects oftransient exposures of neonatal females to testosterone on regional expression of aorticAT1aR and adult susceptibility to AngII-induced vascular diseases. Remarkably, we foundthat exposure of 1 day old female mice to a single dose of testosterone increased AT1aRmRNA abundance that was restricted to abdominal aortas of adult females. In addition,females exposed to testosterone as neonates exhibited a striking increase in adultsusceptibility to AngII-induced AAAs, atherosclerosis and ascending aortic aneurysms.Deficiency of AT1aR in smooth muscle cells (SMCs) reduced effects of neonataltestosterone to promote AngII-induced AAAs in adult females, but had no significant impact

Zhang et al. Page 2

Circ Res. Author manuscript; available in PMC 2013 May 25.

$waterm

ark-text$w

atermark-text

$waterm

ark-text

on the effects of neonatal testosterone in promoting atherosclerosis or ascending aorticaneurysms. Testosterone promoted AT1aR mRNA abundance in SMCs from abdominal, butnot thoracic aortas of male and female mice. Notably, effects of testosterone to promoteAT1aR mRNA abundance were heritable in passaged abdominal aortic SMCs from female,but not male mice. Our findings indicate that transient exposure of neonatal female mice totestosterone abrogates the sexual dimorphism of AngII-induced AAAs.

MethodsMice

Female ApoE-/- and LDLR-/- mice (both N10 C57BL/6 background) were bred to males inhouse. Within 24 hours of birth, female mice were injected once with either vehicle (cornoil) or testosterone propionate (100 or 400 μg/mouse, s.c., Sigma Aldrich, St. Louis, MO).Female ApoE-/- and LDLR-/- mice administered vehicle or testosterone propionate (400 μg/mouse) as neonates were maintained on standard diet and terminated at 5 months of age foraortic gene analysis and aortic vascular morphology measurements without AngII infusion.In separate studies, 3 month old female mice (ApoE-/-, LDLR-/-, AT1aRfl/fl, AT1aRSM22

KO on an LDLR-/-background) administered vehicle or testosterone (400 μg/mouse) asneonates were infused with either saline or AngII (500, 750 or 1,000 ng/kg/min, Bachem,Torrance, CA) using osmotic pumps (Alzet, model 1004, Durect Co., Cupertino, CA) for 4weeks. For studies in male mice, 1 day old male ApoE-/- or LDLR-/- mice were injectedwith either vehicle or testosterone propionate (400 μg/mouse) and then infused with AngII(500 or 750 ng/kg/min for 4 weeks) at 3 months of age. In separate studies, female LDLR-/-mice were injected with a lower dose of testosterone (100 μg/mouse) within 24 hours ofbirth, and then infused with AngII (1,000 ng/kg/min for 28 days) at 2 months of age.ApoE-/- mice were maintained on standard diet; LDLR-/- mice were fed a high fat dietcontaining 21% milk fat and 0.2% cholesterol (TD88137, Harlan Teklad, Indianapolis, IN) 1week prior to pump implantation and throughout the study. All experiments were performedin accordance with the University of Kentucky Institutional Animal Care and UseCommittee.

Generation of AT1aRSM22 KO miceAs described previously, AT1aRfl/fl were generated by InGenious Targeting Laboratory(Stony Brook, NY) directly in ES cells of C57BL/6 mice and subsequently crossed toLDLR-/- mice.1 SM22-Cre mice (The Jackson Laboratory, Tg(Tagln-cre)1Her/J, #004746,Bar Harbor, ME) were bred to LDLR-/- mice and were identified to be the equivalent ofN10. Details regarding breeding of experimental mice are provided in the OnlineSupplement.

Quantification of atherosclerosis, AAA and ascending aortic aneurysmsAtherosclerosis was quantified in the aortic arch and sinus, details are provided in the OnlineSupplement. AAAs were defined as > 50% dilation of lumen diameters from saline-infusedmice. The extent of ascending aortic dilation as an index of ascending aortic aneurysms wasquantified by measuring aortic arch intimal areas (to 3 mm distal to the subclavian branch)and ascending aortic diameters.18 The incidences of ulceration in ascending aortas wereassessed by observers who were blinded to the experimental design. Details regardingquantification of vascular pathologies are provided in the Online Supplement.

All other procedures used in this study were described previously, although an expandedMethods section is available in the Online Data Supplement at http://circres.ahajournals.org.

Zhang et al. Page 3


$waterm

ark-text$w

atermark-text

$waterm

ark-text

http://circres.ahajournals.org

ResultsAbdominal aortic AT1aR mRNA abundance was increased in adult females administeredtestosterone as neonates

The embryonic origins of cells can influence site-specific gene expression patterns indifferent aortic regions.30, 31 To define potential gene targets in aortic regions that areinfluenced by transient exposures to testosterone during development, we administered asingle dose of either testosterone or vehicle to neonatal LDLR-/- mice within the first 24hours after birth. This dosing regimen was chosen to mimic testosterone surges in neonatalmales.28, 29 Targeted PCR arrays in aortas from adult females examined 40 genes that wereimplicated previously in both AngII-induced AAAs in mice 32 and human AAAs (OnlineTable I). Of the 40 genes examined, 16 (40%) were differentially expressed betweenthoracic and abdominal aortas of adult female LDLR-/- mice administered vehicle asneonates (Online Table II). The majority of these genes continued to be expressed atdifferent levels in thoracic compared to abdominal aortas of adult female LDLR-/- miceadministered testosterone as neonates (Online Table III). In aortas from adult femalesadministered testosterone as neonates, 8 genes (20%) exhibited differences compared tovehicle in abdominal (5 genes) or thoracic (3 genes) aortas (Online Table IV). Notably,AT1aR mRNA abundance was increased in abdominal, but not in thoracic aortas from adultfemales administered testosterone as neonates. Since this receptor mediates AngII-inducedatherosclerosis, ascending aortic aneurysms and AAA formation in adult male mice,21-23 weused RT-PCR to confirm region-specific increases in AT1aR mRNA abundance inabdominal aortas from adult ApoE-/- and LDLR-/- female mice that were administeredvehicle or testosterone within 24 hours after birth. As reported previously,26 AT1aR mRNAabundance was not significantly different between thoracic and abdominal aortas of adultfemale ApoE-/- or LDLR-/- mice administered vehicle as neonates (Figure 1A and B,respectively). In contrast, administration of testosterone to neonatal females significantlyincreased AT1aR mRNA abundance in abdominal, but not in thoracic aortas of adult femalemice of each strain (Figure 1A and B). Thus, transient exposures of neonatal female mice totestosterone imparted a long-lasting increase in AT1aR mRNA abundance specifically inabdominal aortas of adult females. Unfortunately, we were unable to quantify AT1aRprotein expression in aortas due to a lack of antibodies exhibiting specificity to AT1aR.23

To determine whether neonatal exposures to testosterone changed aortic structure and/orfunction, we quantified morphology (medial thickness, Figure 1C and D) of selected regionsof aortas from adult LDLR-/- females that were administered either testosterone or vehicleas neonates. Medial thickness was greatest in ascending aortas from adult femalesadministered either testosterone or vehicle compared to other aortic regions (Figure 1D).However, medial thickness was not significantly influenced by neonatal administration oftestosterone. We also quantified contractile responses of different aortic regions to 5-hydroxytryptamine (5-HT) or AngII (Figure 1E and F). Contractile responses to 5-HT weregreatest in suprarenal aortic segments from mice in each group (Figure 1E), while responsesto AngII were greatest in infrarenal aortas (Figure 1F). However, there were no significantdifferences in contractile responses to either agonist in aortic ring segments extracted fromadult females administered testosterone within 24 hours after birth compared to vehiclecontrols. While it may appear surprising that the contractile response to AngII was notinfluenced by neonatal exposures to testosterone given the ability of the hormone to increaseAT1aR mRNA abundance, previous investigators inferred that AT1bR, rather than AT1aRmediate aortic smooth muscle contractions.33 In addition, mRNA abundance of AT1bR wasgreater in abdominal compared to thoracic aortas of adult female mice administered eithervehicle or testosterone as neonates, with no differences in expression of this angiotensinreceptor subtype between females administered vehicle or testosterone as neonates (OnlineTables I-III). Elevated expression of AT1bR in abdominal aortas is consistent with increased

Zhang et al. Page 4


$waterm

ark-text$w

atermark-text

$waterm

ark-text

contractile responses to AngII in this aortic region (Figure 1F). These results demonstratethat administration of testosterone to neonatal females had no overt effects on aorticmorphology or contractile function.

AngII-induced AAAs were strikingly increased in adult females administered testosteroneas neonates

We next investigated if elevated AT1aR mRNA abundance in abdominal aortas of adultfemales administered testosterone as neonates translated into increased susceptibility toAngII-induced AAAs. Adult ApoE-/- females administered vehicle or testosterone asneonates were infused with either saline or AngII (1,000 ng/kg/min) for 28 days. Inagreement with previous reports,34, 35 neonatal administration of testosterone resulted in amodest but significant increase in body weights of adult females (Online Table V for AngII-infused groups). However, there was no significant effect of testosterone on systolic bloodpressure, blood monocyte counts, serum cholesterol concentrations (Online Table V), orlipoprotein cholesterol distribution (Online Figure I) in adult females infused with AngII atdifferent rates. Notably, serum testosterone concentrations were not significantly elevated inadult female mice administered testosterone as neonates (Online Table V), and were 7-foldlower than those observed previously in adult male ApoE-/- mice.25

In saline-infused adult female ApoE-/- mice, neonatal administration of testosterone had nosignificant effect on maximal aortic diameters of suprarenal abdominal aortas (Figure 2B).AAA incidence, which included mice that died from aneurysmal rupture, was increaseddose-dependently by AngII in female mice administered vehicle as neonates (Figure 2A).However, it did not rise above 21% at the highest infusion rate (1,000 ng/kg/min) of AngII.Infusion of AngII (1,000 ng/kg/min) to adult females administered vehicle as neonatesresulted in modest but significant increases in external diameters of abdominal aortas(Figure 2B), indicative of aneurysm formation in a small percentage of mice (Figure 2A and2C). In contrast, AAA incidence was strikingly elevated with increasing infusion rates ofAngII in adult females administered testosterone as neonates compared to vehicle-administered controls (Figure 2A), with a 3-fold increase in AAA incidence at infusions of1,000 ng/kg/min of AngII (21% vs. 64%; P<0.05). In addition, adult female mice exposed totestosterone as neonates exhibited striking increases in external diameters of abdominalaortas with pronounced aneurysmal pathology similar to that observed previously in adultmales (Figure 2A-C).24, 25 An increase in mortality from aneurysmal rupture contributed tohigher AAA incidences in females administered testosterone as neonates (rupture: vehicle,7%; testosterone, 20%). These results demonstrate that transient exposure of neonatalfemales to testosterone abrogates the sexual dimorphism of AngII-induced AAAs in adultfemales that exhibited low concentrations of serum testosterone.

We also examined effects of the same dose of testosterone administered to 1 day old maleApoE-/- mice on susceptibility to AngII-induced vascular diseases as adults. In contrast tofindings in females, AAA incidence and maximal external diameters were not altered inadult male mice administered testosterone as neonates (Online Figure II, A and B). Theseresults suggest that in male mice, with the natural occurrence of neonatal androgen surges,the dose of testosterone administered in this study did not impose supraphysiological effectson AngII-induced AAAs.

SMC deficiency of AT1aR blunted the effects of neonatal testosterone exposure topromote AngII-induced AAAs

AT1aRs are expressed in several cell types within the vascular wall. Previous studiesexamining temporal changes in aneurysmal pathology with AngII infusion demonstratedelastin degradation within the medial layer of abdominal aortas early in AngII-induced

Zhang et al. Page 5


$waterm

ark-text$w

atermark-text

$waterm

ark-text

AAAs.36 These results suggested that testosterone exposures in neonatal females mayregulate AT1aR mRNA abundance within SMCs of the abdominal aorta to promote AngII-induced AAA formation. Therefore, we investigated if SMC-specific AT1aR deficiencyreduced the effect of neonatal testosterone in promoting AngII-induced AAAs. As describedpreviously,23 we used Cre-LoxP technology to generate mice with deficiency of AT1aR inSMCs. Mice were engineered with loxP sites flanking exon 3 of AT1aR which includes theentire coding region (Online Figure III, A). AT1aRfl/fl females on an LDLR-/- backgroundwere mated to male mice expressing Cre recombinase under the control of an SM22promoter to generate smooth muscle cell-specific AT1aR deficient mice (AT1aRSM22 KO)and wild type littermates (AT1aRfl/fl).23, 37, 38 Depletion of AT1aR from SMCs was verifiedby PCR of genomic DNA from aortas (Online Figure III, B). Female AT1aRfl/fl control andAT1aRSM22 KO mice were administered a single dose of vehicle or testosterone within 24hours of birth, and then infused with AngII at 2 months of age. Notably, neonataladministration of testosterone, as well as SMC-specific AT1aR deficiency, had nosignificant effect on baseline systolic blood pressures, AngII-induced hypertension,23, 39

plasma aldosterone concentrations, serum cholesterol concentrations (Table 1) or serumlipoprotein cholesterol distributions (Online Figure IV). Moreover, serum testosteroneconcentrations did not significantly differ across study groups (Table 1).

Similar to findings from adult ApoE-/- females exposed to testosterone within 24 hours ofbirth, adult AT1aRfl/fl LDLR-/- females administered testosterone as neonates exhibited astriking increase in the incidence (Figure 3A) and size (Figure 3B and 3C) of AngII-inducedAAAs. Interestingly, SMC specific AT1aR deficiency significantly reduced, but did notablate effects of neonatal testosterone to promote AngII-induced AAAs (Figure 3A-C).Notably, even though AAA incidence was significantly decreased in AT1aRSM22 KOfemales administered testosterone as neonates, females of each genotype that exhibited anAAA developed aneurysms of similar sizes (Figure 3D). Moreover, we examined tissuecharacteristics of AAAs of equivalent sizes that formed in adult female mice of eachgenotype exposed to testosterone as neonates. In both genotypes, AAAs exhibited typicalcharacteristics of pronounced adventitial thrombus, elastin degradation within the aorticmedia and pronounced lumen dilation (Figure 3E). These data demonstrate that exposure ofneonatal females to testosterone results in a long-lasting increase in adult susceptibility offemales to AngII-induced AAAs partially through smooth muscle cell AT1aR.

AngII-induced atherosclerosis and ascending aortic aneurysms were increased in adultfemales administered testosterone as neonates, but SMC deficiency of AT1aR had noeffect

We also quantified ascending aortic aneurysms18 and atherosclerosis17 as additionalvascular pathologies induced by infusion of AngII that do not display sexual dimorphismbetween adult males and females.25, 26 In females administered vehicle as neonates, therewas no significant effect of SMC AT1aR deficiency on AngII-induced atherosclerosis inaortic sinuses (Figure 4A) or arches (Online Figure V, A). Surprisingly, testosteroneadministration to neonatal female mice resulted in pronounced increases in AngII-inducedatherosclerosis in AT1aRfl/fl and AT1aRSM22 KO adult female mice (Figure 4A and OnlineFigure V, A). However, deficiency of AT1aR in SMCs did not significantly influence theability of neonatal testosterone administration to promote AngII-induced atherosclerosis(Figure 4A-C). Lesional phagocytes, as defined by quantification of CD68 immunostainingin aortic sinus sections from adult females, were increased significantly by neonataladministration of testosterone in lesions from AT1aRfl/fl and AT1aRSM22 female mice(Figure 4B). Moreover, cellular compositions (phagocytes, SMCs) of atherosclerotic lesionsin aortic sinus sections were similar between AT1aRfl/fl and AT1aRSM22 female miceadministered testosterone as neonates (Figure 4C).

Zhang et al. Page 6


$waterm

ark-text$w

atermark-text

$waterm

ark-text

In addition to promoting atherosclerosis, neonatal testosterone administration to femaleLDLR-/- mice augmented AngII-induced increases in ascending aortic diameters (Figure4D) and ascending aortic intimal areas (Online Figure V, B), indicative of ascending aorticaneurysms.18 Cross sections from ascending aortas of female mice exposed to testosteroneas neonates and then infused with AngII as adults exhibited significant medial thickening(Figure 4E and F) and aortic ulceration (Online Figure V, C and D). Macrophageimmunostaining was prominent in media and adventitia of aortic arch sections from AngII-infused adult female mice administered testosterone as neonates (Figure 4E), similar toresults previously observed in ascending aortas of adult male mice.18 SMC deficiency ofAT1aR did not significantly influence the ability of neonatal testosterone administration topromote AngII-induced aneurysms in the ascending aorta (Figure 4D-F). Thus, SMC AT1aRdeficiency resulted in a selective reduction in AngII-induced AAAs, but not atherosclerosisor ascending aortic aneurysms, in female mice administered testosterone as neonates.

Neonatal female mice were more sensitive compared to neonatal males to testosteroneexposures to promote AngII-induced vascular pathologies

Our approach has been to attempt to mimic neonatal surges of testosterone in male miceshortly after birth28, 29 and superimpose similar systemic concentrations of testosterone onneonatal females. It is technically challenging to measure serum testosterone concentrationsin 1 day old male and female mice to confirm whether we had indeed achieved similarcirculating testosterone concentrations between males and females. Since administration oftestosterone to neonatal ApoE-/- males had no significant effect on adult male susceptibilityto AngII-induced AAAs, it is likely that the dose of testosterone was not supra-physiologicin addition to endogenous neonatal surges of testosterone. However, as an alternativeapproach, we determined if administration of a 4-fold lower dose of testosterone (100 μg) to1 day old female LDLR-/- mice was sufficient to confer increased adult susceptibility toAngII-induced AAAs. We also administered a higher dose (400 μg) of testosterone toneonatal LDLR-/- male mice and then defined adult sensitivity to AngII-induced AAAs.AAA incidence (from 30 to 64%, Figure 5A) and size (Figure 5B and 5C) increased in adultfemale LDLR administered the lower dose of testosterone as neonates. Atherosclerosis wasincreased significantly in aortic sinus sections from adult AngII-infused femalesadministered the lower dose of testosterone as neonates (Figure 5D). In addition, aortic archareas (vehicle: 17.6 ± 0.8; testosterone: 20.4 ± 0.8 mm2; P<0.05) and ulceration of the aorticarch (vehicle: 20%; testosterone: 57%) were significantly increased in females administeredthe lower dose of testosterone as neonates. In contrast, male LDLR-/- mice exhibited noresponse to a higher dose (400 μg) of neonatal testosterone exposure in the development ofAngII-induced AAAs, atherosclerosis and ascending aortic aneurysms (Online Figure VI, A-C). These results demonstrate pronounced sensitivity of neonatal females to testosterone topromote a long-lasting increased susceptibility to AngII-induced vascular diseases.

Testosterone-induced increases in AT1aR mRNA abundance in SMCs were region-specificand sexually dimorphic

Neonatal testosterone exposures exerted a region-specific effect to promote a long-lastingincrease in AT1aR mRNA abundance in abdominal aortic SMCs to increase susceptibility toAngII-induced AAAs in adult females. To define mechanisms for these effects, wequantified androgen receptor mRNA abundance in thoracic and abdominal aortas of adultmale and female ApoE-/- mice. Interestingly, similar to regional differences in AT1aRexpression (Figure 1A and 1B), androgen receptor mRNA abundance was expressed at asignificantly higher abundance in abdominal compared to thoracic aortas of male and femalemice (Online Figure VII). To determine if regional differences in androgen receptorexpression abundance arose from SMC cells within different aortic regions, we culturedSMCs from thoracic and abdominal aortic regions of adult male and female mice. Notably,

Zhang et al. Page 7


$waterm

ark-text$w

atermark-text

$waterm

ark-text

androgen receptor mRNA and protein expression were significantly increased in SMCscultured from abdominal compared to thoracic aortas from male and female mice (Figure 6Aand 6B). Furthermore, testosterone significantly increased AT1aR mRNA abundance inprimary cultures of SMCs from abdominal, but not thoracic aortas of male and female mice(Figure 6C and 6D). These results suggest that regional differences in androgen receptorexpression contribute to site-specific increases in abdominal aortic SMC AT1aR expressionby testosterone.

Since heritability is a characteristic of epigenetic gene regulation, we examined heritabilityof testosterone-induced increases in AT1aR mRNA abundance in passaged SMCs fromabdominal aortas of male and female mice. After incubation of abdominal aortic SMCs withvehicle or testosterone for 24 hours, cells from both male and female mice were passaged toproliferate for one additional generation. Elevated AT1aR mRNA abundance persisted inpassaged abdominal SMCs from female, but not male mice (Figure 6E). These resultsdemonstrate sexual dimorphism of testosterone-induced regulation of AT1aR expression inabdominal SMCs.

To determine if testosterone-induced increases in AT1aR mRNA abundance in abdominalaortic SMCs result in increased functional responses to AngII, we examined AngII-inducedhydrogen peroxide (H2O2) generation in testosterone-treated SMCs cultured from thoracicand abdominal aortas of female mice. Incubation of thoracic and abdominal aortic SMCswith AngII significantly increased H2O2 generation, which was abolished by losartan(Figure 6F and 6G). In abdominal, but not thoracic aortic SMCs previously exposed totestosterone, AngII-induced H2O2 generation was increased compared to vehicle (DMSO).These results demonstrate that testosterone-induced increases in AT1aR mRNA abundancespecifically in abdominal aortic SMCs result in heightened regional-specific functionalresponses to AngII.

DiscussionOur findings reveal that exposure of neonatal female mice to a single dose of testosterone issufficient to increase the adult susceptibility to AngII-induced AAAs. Notably, exposure ofneonatal females to testosterone promoted a regional increase in abdominal aortic AT1aRmRNA abundance in adult females, a mechanism that was suggested to contribute to sexualdimorphism of AngII-induced AAAs between adult male and female mice.24, 25 To gaininsights into the cell targets of testosterone during development to increase aortic AT1aRmRNA abundance and enhance AngII-induced AAAs, we performed studies in female micelacking AT1aR in SMCs. SMC-specific deficiency of AT1aR reduced neonatal effects oftestosterone to promote AngII-induced AAAs. Notably, neonatal exposures of female miceto testosterone also increased AngII-induced atherosclerosis and ascending aorticaneurysms, even though these vascular pathologies do not exhibit sexual dimorphismbetween adult males and females. However, SMC deficiency of AT1aR had no effect onaugmented atherosclerosis or ascending aortic aneurysms in adult females exposed totestosterone as neonates. Mechanisms for site-specific effects of testosterone to regulateAT1aR mRNA expression and AngII-induced reactive oxygen species production includedifferential expression of androgen receptors in SMCs from abdominal compared to thoracicaortas. Moreover, heritable effects of testosterone to promote AT1aR expression inabdominal aortic SMCs from female, but not male mice demonstrate sexual dimorphism oftestosterone regulation of AT1aR expression. These results demonstrate that distinctmechanisms mediate effects of neonatal testosterone exposures to promote 3 differentAngII-induced vascular pathologies in adult females. Moreover, these studies are the first todemonstrate that transient exposure of neonatal females to testosterone abrogates the sexualdimorphism phenotype of AAAs.

Zhang et al. Page 8


$waterm

ark-text$w

atermark-text

$waterm

ark-text

Exposure of neonatal females to testosterone, designed to mimic a surge of testosterone inmales shortly after birth,28, 29 has been used extensively by many investigators to studysexual dimorphism of brain structure and function. Permanent changes in brain neurons offemales androgenized during development are considered to be “organizational”, allowingfor male-like behaviors that persist to adulthood.27 In contrast to studies focused on brain,little is known about developmental effects of testosterone on other traits. We focused oneffects of testosterone during neonatal development on regional variations in AT1aRabundance in aortas since previous studies demonstrated a critical requirement of AT1aR forAngII-induced AAAs, atherosclerosis and ascending aortic aneurysms in male mice.18, 22

Moreover, our previous results demonstrated higher expression of AT1aR in abdominalcompared to thoracic aortas of male, but not female mice.26 In this study we demonstratedthat administration of testosterone to neonatal female mice initiated greater abdominal aorticAT1aR mRNA abundance and markedly enhanced AAA formation in adult females.Interestingly, unlike males,25, 26 exposure of neonatal females to testosterone resulted inlong-lasting effects that persisted into adulthood and did not require the continued presenceof high concentrations of testosterone in serum. In SMCs cultured from abdominal aortas offemale, but not male mice, testosterone promotion of AT1aR expression was heritable,supporting epigenetic mechanisms contribute to sexual dimorphism of testosterone’s effects.Androgen receptors, which exhibited a region-specific increase in abdominal compared tothoracic aortic SMCs in the present study, regulate transcription of target genes by accessingDNA in complexes with co-activators and co-repressors, many of which have enzymaticactivities for histone modification.40, 41 Previous studies demonstrated that administration ofa histone deacetylase (HDAC) inhibitor to neonatal male mice blocked masculinization ofthe adult male brain.42 Moreover, administration of an HDAC inhibitor to neonatal femalemice that were androgenized by testosterone blocked masculinization of the adult brain.42

These findings suggest that epigenetic effects of androgen receptor stimulation duringneonatal development may have contributed to long-lasting increases in vascular diseasesusceptibility of adult females. Moreover, regional differences in androgen receptorexpression in aortic SMCs most likely contributed to site-specific increases in AT1aRexpression in abdominal aortas of adult females administered testosterone as neonates.

Our findings demonstrate that male mice experiencing normal surges of testosterone duringthe neonatal period administered the single dose of testosterone did not respond the same asfemales exhibiting striking increases in AngII-induced vascular diseases. Thus, it is unlikelythat the dose of testosterone administered in these studies was supra-physiologic. Rather,these results suggest that neonatal male and female mice respond differently to testosterone.This hypothesis is supported by augmented AngII-induced atherosclerosis and ascendingaortic aneurysms in adult females administered testosterone as neonates, even though adultmale and female mice do not display sex differences in these AngII-initiated vasculardiseases. Moreover, even though male mice exhibited no response to the higher dose oftestosterone during neonatal development, a 4-fold lower dose of testosterone administeredto neonatal females promoted their adult susceptibility to AngII-induced vascular diseases.Finally, abdominal aortic SMCs from female, but not male mice exhibited heritability oftestosterone regulation of AT1aR. Taken together, these results suggest that responses totestosterone differed between male and female mice. Recent studies support sexchromosomes influence sexual dimorphism of AngII-induced hypertension.43 Similarly,sexual dimorphism of AngII-bradycardiac baroreflex responses were ascribed to differencesin sex chromosomes in a 4 core genotype model allowing for dissection of effects of gonadalsex from sex chromosome complement.44 Differences in the complement of genes in XXversus XY cells, or X-linked gene differences may have contributed to differential responsesto testosterone between male and female mice.

Zhang et al. Page 9


$waterm

ark-text$w

atermark-text

$waterm

ark-text

Our previous results demonstrated medial elastin degradation was an early event in AAAformation in adult male ApoE-/- mice.36 Results from this study demonstrate thattestosterone increases AT1aR mRNA expression in primary cultured mouse abdominalaortic SMCs. Moreover, this is the first demonstration that deficiency of AT1aR in SMCsdecreased neonatal effects of testosterone to promote AngII-induced AAAs, but had noeffect on 2 other vascular pathologies induced by infusion of AngII. Interestingly, in thisstudy, AAAs that did form in female mice with smooth muscle cell-specific AT1aRdeficiency had similar size and pathologic characteristics to those of wild type controls.Thus, as opposed to downstream signaling pathways, testosterone may influence initiatingevents in the formation of AngII-induced AAAs by promoting AT1aR expression in pivotalcell types, including vascular SMCs. Apoptosis of SMCs contribute to AngII-inducedAAAs, since administration of a caspase inhibitor reduced medial apoptosis andsignificantly decreased AAA formation.45 Moreover, genetic deficiency of cyclophilin A, achaperone protein abundantly expressed in SMCs, abolished AngII-induced AAAs.46 Sinceeffects of testosterone to increase AT1aR mRNA abundance were heritable in SMCsspecifically from abdominal aortas of female mice, these results suggest that testosteroneexerts epigenetic effects to regulate AT1aR expression in a site-specific and sexuallydimorphic manner. Mechanisms for site-specific effects of testosterone to increaseabdominal aortic AT1aR expression include greater androgen receptor expression inabdominal aortic SMCs. These results suggest that promotion of AT1aR expression inabdominal aortic SMCs by testosterone influenced the initiating event in AngII-inducedAAAs. Of note, increased AT1aR mRNA abundance by testosterone exposures in SMCsisolated from abdominal aortas resulted in enhanced functional responses to AngII.

The diversity of SMC embryonic origins has been suggested to contribute to region-specificaortic pathologies, including those induced by infusion of AngII.47, 48 For example, thoracicand abdominal aortic SMCs respond distinctively to transforming growth factor-β (TGF-β).30 Inhibition of TGF-β has a beneficial effect in a mouse model of ascending aorticaneurysms while inhibition of TGF-β promotes aortic dissection of AngII-inducedAAAs. 49, 50 Our results demonstrate that thoracic and abdominal aortas differ in theregulation of expression of several genes, most importantly AT1aR, and that effects oftestosterone to increase aortic AT1aR expression were specific to abdominal aortas. Recentresults demonstrated that infusion of AngII to C57BL/6 male mice resulted in hyperplasia ofSMCs in the ascending aorta, but hypertrophy in other aortic regions.51 Interestingly, despitedifferences in AT1aR-mediated regulation of SMCs growth along the aortic length, allgrowth-related responses of SMCs to AngII were abolished in AT1aR deficient mice.51 Instudies focused on AT1aR located to endothelial or SMCs as targets of AngII to induceaneurysms in the ascending aorta, results demonstrated that deficiency of AT1aR inendothelial cells, but not SMCs, reduced AngII-induced ascending aortic aneurysm.23 Ourresults demonstrate that even though testosterone promoted a region-specific increase inAT1aR mRNA abundance to abdominal aortas, all 3 AngII-induced pathologies wereincreased in neonatal females exposed to testosterone. Moreover, while SMC-specificdeficiency of AT1aR reduced effects of testosterone to promote AngII-induced AAAs,ascending aortic aneurysms and atherosclerosis were not influenced by deficiency of AT1aRon this cell type. A lack of effect of SMC AT1aR deficiency to reduce ascending aorticaneurysms in adult females exposed to testosterone as neonates is consistent with recentresults obtained from adult male mice lacking SMC AT1aR.23

In conclusion, these data demonstrate that a single dose of testosterone administered toneonatal female mice confers permanently increased susceptibility to AngII-induced AAAs,atherosclerosis and ascending aortic aneurysms in adulthood. An ability of a single dose oftestosterone administered to neonatal females to promote three distinct vascular pathologiesindicates that short term hormonal exposures at pivotal periods of development can

Zhang et al. Page 10


$waterm

ark-text$w

atermark-text

$waterm

ark-text

markedly impact vascular disease susceptibility of adults. Moreover, mechanisms for sexualdimorphism of these vascular pathologies are distinct, since smooth muscle AT1aRdeficiency reduced the ability of neonatal administration to increase AAAs, but not AngII-induced atherosclerosis or ascending aortic aneurysms in adult females. Heritability oftestosterone’s effect to increase SMC AT1aR in abdominal aortic SMCs from female, butnot male mice demonstrates sexual dimorphism of SMC responses to testosterone. Theseresults demonstrate that adult AAA susceptibility can be regulated by neonatal exposures totestosterone, indicating a pivotal role for testosterone during critical periods of developmentas an initiator of the sexual dimorphism of AAAs.

Supplementary MaterialRefer to Web version on PubMed Central for supplementary material.

AcknowledgmentsWe thank Jessica Moorleghen for technical assistance with aortic contractility measurement, breeding andadministration of testosterone to neonatal female mice, Anju Balakrishnan for assistance with mouse breeding,Deborah Howatt for tissue pathology and FPLC, and Victoria English for technical assistance with biochemicalassays.

Sources of Funding

This study was funded by research grants (HL08100, HL107326, LAC, AD) from the National Institutes of Health.Morphologic assessment of aortic segments was performed through a pathology core supported by an NIH NCRRCenter of Biomedical Research Excellence (P20RR021954; LAC).

References1. Brown LC, Powell JT. Risk factors for aneurysm rupture in patients kept under ultrasound

surveillance. UK Small Aneurysm Trial Participants. Ann Surg. 1999; 230:289–296. [PubMed:10493476]

2. Heikkinen M, Salenius JP, Auvinen O. Ruptured abdominal aortic aneurysm in a well-definedgeographic area. J Vasc Surg. 2002; 36:291–296. [PubMed: 12170209]

3. Qureshi NA, Rehman A, Slater N, Moss E, Shiralkar S, Patel RT, Grimley RP, Jayatunga AP.Abdominal aortic aneurysm surgery in a district general hospital: a 15-years experience. Ann VascSurg. 2007; 21:749–753. [PubMed: 17512166]

4. Kent KC, Zwolak RM, Egorova NN, Riles TS, Manganaro A, Moskowitz AJ, Gelijns AC, Greco G.Analysis of risk factors for abdominal aortic aneurysm in a cohort of more than 3 millionindividuals. J Vasc Surg. 2010; 52:539–548. [PubMed: 20630687]

5. Pleumeekers HJ, Hoes AW, van der Does E, van Urk H, Hofman A, de Jong PT, Grobbee DE.Aneurysms of the abdominal aorta in older adults. The Rotterdam Study. Am J Epidemiol. 1995;142:1291–1299. [PubMed: 7503049]

6. Singh K, Bonaa KH, Jacobsen BK, Bjork L, Solberg S. Prevalence of and risk factors for abdominalaortic aneurysms in a population-based study : The Tromso Study. Am J Epidemiol. 2001; 154:236–244. [PubMed: 11479188]

7. Lederle FA, Johnson GR, Wilson SE, Chute EP, Littooy FN, Bandyk D, Krupski WC, Barone GW,Acher CW, Ballard DJ. Prevalence and associations of abdominal aortic aneurysm detected throughscreening. Aneurysm Detection and Management (ADAM) Veterans Affairs Cooperative StudyGroup. Ann Intern Med. 1997; 126:441–449. [PubMed: 9072929]

8. Lee AJ, Fowkes FG, Carson MN, Leng GC, Allan PL. Smoking, atherosclerosis and risk ofabdominal aortic aneurysm. Eur Heart J. 1997; 18:671–676. [PubMed: 9129900]

9. Lederle FA, Nelson DB, Joseph AM. Smokers’ relative risk for aortic aneurysm compared withother smoking-related diseases: a systematic review. J Vasc Surg. 2003; 38:329–334. [PubMed:12891116]



$waterm

ark-text$w

atermark-text

$waterm

ark-text

10. Golledge J, Clancy P, Jamrozik K, Norman PE. Obesity, adipokines, and abdominal aorticaneurysm: Health in Men study. Circulation. 2007; 116:2275–2279. [PubMed: 17967974]

11. Police SB, Thatcher SE, Charnigo R, Daugherty A, Cassis LA. Obesity promotes inflammation inperiaortic adipose tissue and angiotensin II-induced abdominal aortic aneurysm formation.Arterioscler Thromb Vasc Biol. 2009; 29:1458–1464. [PubMed: 19608970]

12. Simoni G, Pastorino C, Perrone R, Ardia A, Gianrossi R, Decian F, Cittadini G Jr, Baiardi A,Bachi V. Screening for abdominal aortic aneurysms and associated risk factors in a generalpopulation. Eur J Vasc Endovasc Surg. 1995; 10:207–210. [PubMed: 7655973]

13. Cornuz J, Sidoti Pinto C, Tevaearai H, Egger M. Risk factors for asymptomatic abdominal aorticaneurysm: systematic review and meta-analysis of population-based screening studies. Eur JPublic Health. 2004; 14:343–349. [PubMed: 15542867]

14. Golledge J, Muller J, Daugherty A, Norman P. Abdominal aortic aneurysm: pathogenesis andimplications for management. Arterioscler Thromb Vasc Biol. 2006; 26:2605–2613. [PubMed:16973970]

15. Forsdahl SH, Singh K, Solberg S, Jacobsen BK. Risk factors for abdominal aortic aneurysms: a 7-year prospective study: the Tromso Study, 1994-2001. Circulation. 2009; 119:2202–2208.[PubMed: 19364978]

16. Daugherty A, Cassis L. Chronic angiotensin II infusion promotes atherogenesis in low densitylipoprotein receptor -/- mice. Ann N Y Acad Sci. 1999; 892:108–118. [PubMed: 10842656]

17. Daugherty A, Manning MW, Cassis LA. Angiotensin II promotes atherosclerotic lesions andaneurysms in apolipoprotein E-deficient mice. J Clin Invest. 2000; 105:1605–1612. [PubMed:10841519]

18. Daugherty A, Rateri DL, Charo IF, Owens AP, Howatt DA, Cassis LA. Angiotensin II infusionpromotes ascending aortic aneurysms: attenuation by CCR2 deficiency in apoE-/- mice. Clin Sci(Lond). 2010; 118:681–689. [PubMed: 20088827]

19. Lu H, Rateri DL, Cassis LA, Daugherty A. The role of the renin-angiotensin system in aorticaneurysmal diseases. Curr Hypertens Rep. 2008; 10:99–106. [PubMed: 18474175]

20. Daugherty A, Poduri A, Chen X, Lu H, Cassis LA. Genetic variants of the Renin Angiotensinsystem: effects on atherosclerosis in experimental models and humans. Curr Atheroscler Rep.2010; 12:167–173. [PubMed: 20425255]

21. Daugherty A, Rateri DL, Lu H, Inagami T, Cassis LA. Hypercholesterolemia stimulatesangiotensin peptide synthesis and contributes to atherosclerosis through the AT1A receptor.Circulation. 2004; 110:3849–3857. [PubMed: 15596561]

22. Cassis LA, Rateri DL, Lu H, Daugherty A. Bone marrow transplantation reveals that recipientAT1a receptors are required to initiate angiotensin II-induced atherosclerosis and aneurysms.Arterioscler Thromb Vasc Biol. 2007; 27:380–386. [PubMed: 17158350]

23. Rateri DL, Moorleghen JJ, Balakrishnan A, Owens AP 3rd, Howatt DA, Subramanian V, PoduriA, Charnigo R, Cassis LA, Daugherty A. Endothelial cell-specific deficiency of Ang II type 1areceptors attenuates Ang II-induced ascending aortic aneurysms in LDL receptor-/- mice. CircRes. 2011; 108:574–581. [PubMed: 21252156]

24. Manning MW, Cassis LA, Huang J, Szilvassy SJ, Daugherty A. Abdominal aortic aneurysms:fresh insights from a novel animal model of the disease. Vasc Med. 2002; 7:45–54. [PubMed:12083734]

25. Henriques TA, Huang J, D’Souza SS, Daugherty A, Cassis LA. Orchidectomy, but notovariectomy, regulates angiotensin II-induced vascular diseases in apolipoprotein E-deficientmice. Endocrinology. 2004; 145:3866–3872. [PubMed: 15105380]

26. Henriques T, Zhang X, Yiannikouris FB, Daugherty A, Cassis LA. Androgen increases AT1areceptor expression in abdominal aortas to promote angiotensin II-induced AAAs inapolipoprotein E-deficient mice. Arterioscler Thromb Vasc Biol. 2008; 28:1251–1256. [PubMed:18451329]

27. Arnold AP. The organizational-activational hypothesis as the foundation for a unified theory ofsexual differentiation of all mammalian tissues. Horm Behav. 2009; 55:570–578. [PubMed:19446073]



$waterm

ark-text$w

atermark-text

$waterm

ark-text

28. Motelica-Heino I, Castanier M, Corbier P, Edwards DA, Roffi J. Testosterone levels in plasma andtestes of neonatal mice. J Steroid Biochem. 1988; 31:283–286. [PubMed: 3419158]

29. Corbier P, Edwards DA, Roffi J. The neonatal testosterone surge: a comparative study. Arch IntPhysiol Biochim Biophys. 1992; 100:127–131. [PubMed: 1379488]

30. Topouzis S, Majesky MW. Smooth muscle lineage diversity in the chick embryo. Two types ofaortic smooth muscle cell differ in growth and receptor-mediated transcriptional responses totransforming growth factor-beta. Dev Biol. 1996; 178:430–445.

31. Gadson PF Jr, Dalton ML, Patterson E, Svoboda DD, Hutchinson L, Schram D, Rosenquist TH.Differential response of mesoderm- and neural crest-derived smooth muscle to TGF-beta1:regulation of c-myb and alpha1 (I) procollagen genes. Exp Cell Res. 1997; 230:169–180.[PubMed: 9024776]

32. Rush C, Nyara M, Moxon JV, Trollope A, Cullen B, Golledge J. Whole genome expressionanalysis within the angiotensin II-apolipoprotein E deficient mouse model of abdominal aorticaneurysm. BMC Genomics. 2009; 10:298. [PubMed: 19580648]

33. Zhou Y, Chen Y, Dirksen WP, Morris M, Periasamy M. AT1b receptor predominantly mediatescontractions in major mouse blood vessels. Circ Res. 2003; 93:1089–1094. [PubMed: 14563714]

34. Nilsson C, Niklasson M, Eriksson E, Bjorntorp P, Holmang A. Imprinting of female offspring withtestosterone results in insulin resistance and changes in body fat distribution at adult age in rats. JClin Invest. 1998; 101:74–78. [PubMed: 9421468]

35. Ramirez MC, Luque GM, Ornstein AM, Becu-Villalobos D. Differential neonatal testosteroneimprinting of GH-dependent liver proteins and genes in female mice. J Endocrinol. 2010;207:301–308. [PubMed: 20943812]

36. Saraff K, Babamusta F, Cassis LA, Daugherty A. Aortic dissection precedes formation ofaneurysms and atherosclerosis in angiotensin II-infused, apolipoprotein E-deficient mice.Arterioscler Thromb Vasc Biol. 2003; 23:1621–1626. [PubMed: 12855482]

37. Kuhbandner S, Brummer S, Metzger D, Chambon P, Hofmann F, Feil R. Temporally controlledsomatic mutagenesis in smooth muscle. Genesis. 2000; 28:15–22. [PubMed: 11020712]

38. Holtwick R, Gotthardt M, Skryabin B, Steinmetz M, Potthast R, Zetsche B, Hammer RE, Herz J,Kuhn M. Smooth muscle-selective deletion of guanylyl cyclase-A prevents the acute but notchronic effects of ANP on blood pressure. Proc Natl Acad Sci U S A. 2002; 99:7142–7147.[PubMed: 11997476]

39. Sparks MA, Parsons KK, Stegbauer J, Gurley SB, Vivekanandan-Giri A, Fortner CN, Snouwaert J,Raasch EW, Griffiths RC, Haystead TA, Le TH, Pennathur S, Koller B, Coffman TM.Angiotensin II type 1A receptors in vascular smooth muscle cells do not influence aorticremodeling in hypertension. Hypertension. 2011; 57:577–585. [PubMed: 21242463]

40. Metzger E, Wissmann M, Yin N, Muller JM, Schneider R, Peters AH, Gunther T, Buettner R,Schule R. LSD1 demethylates repressive histone marks to promote androgen-receptor-dependenttranscription. Nature. 2005; 437:436–439. [PubMed: 16079795]

41. Wissmann M, Yin N, Muller JM, Greschik H, Fodor BD, Jenuwein T, Vogler C, Schneider R,Gunther T, Buettner R, Metzger E, Schule R. Cooperative demethylation by JMJD2C and LSD1promotes androgen receptor-dependent gene expression. Nat Cell Biol. 2007; 9:347–353.[PubMed: 17277772]

42. Murray EK, Hien A, de Vries GJ, Forger NG. Epigenetic control of sexual differentiation of thebed nucleus of the stria terminalis. Endocrinology. 2009; 150:4241–4247. [PubMed: 19497973]

43. Ji H, Zheng W, Wu X, Liu J, Ecelbarger CM, Watkins R, Arnold AP, Sandberg K. Sexchromosome effects unmasked in angiotensin II-induced hypertension. Hypertension. 2010;55:1275–1282. [PubMed: 20231528]

44. Caeiro XE, Mir FR, Vivas LM, Carrer HF, Cambiasso MJ. Sex chromosome complementcontributes to sex differences in bradycardic baroreflex response. Hypertension. 2011; 58:505–511. [PubMed: 21810650]

45. Yamanouchi D, Morgan S, Kato K, Lengfeld J, Zhang F, Liu B. Effects of caspase inhibitor onangiotensin II-induced abdominal aortic aneurysm in apolipoprotein E-deficient mice. ArteriosclerThromb Vasc Biol. 2010; 30:702–707. [PubMed: 20075419]



$waterm

ark-text$w

atermark-text

$waterm

ark-text

46. Satoh K, Nigro P, Matoba T, O’Dell MR, Cui Z, Shi X, Mohan A, Yan C, Abe J, Illig KA, BerkBC. Cyclophilin A enhances vascular oxidative stress and the development of angiotensin II-induced aortic aneurysms. Nat Med. 2009; 15:649–656. [PubMed: 19430489]

47. Majesky MW. Developmental basis of vascular smooth muscle diversity. Arterioscler ThrombVasc Biol. 2007; 27:1248–1258. [PubMed: 17379839]

48. Tromp G, Kuivaniemi H, Hinterseher I, Carey DJ. Novel genetic mechanisms for aorticaneurysms. Curr Atheroscler Rep. 2010; 12:259–266. [PubMed: 20446064]

49. Habashi JP, Judge DP, Holm TM, Cohn RD, Loeys BL, Cooper TK, Myers L, Klein EC, Liu G,Calvi C, Podowski M, Neptune ER, Halushka MK, Bedja D, Gabrielson K, Rifkin DB, Carta L,Ramirez F, Huso DL, Dietz HC. Losartan, an AT1 antagonist, prevents aortic aneurysm in amouse model of Marfan syndrome. Science. 2006; 312:117–121. [PubMed: 16601194]

50. Wang Y, Ait-Oufella H, Herbin O, Bonnin P, Ramkhelawon B, Taleb S, Huang J, Offenstadt G,Combadiere C, Renia L, Johnson JL, Tharaux PL, Tedgui A, Mallat Z. TGF-beta activity protectsagainst inflammatory aortic aneurysm progression and complications in angiotensin II-infusedmice. J Clin Invest. 2010; 120:422–432. [PubMed: 20101093]

51. Owens AP 3rd, Subramanian V, Moorleghen JJ, Guo Z, McNamara CA, Cassis LA, Daugherty A.Angiotensin II induces a region-specific hyperplasia of the ascending aorta through regulation ofinhibitor of differentiation 3. Circ Res. 2010; 106:611–619. [PubMed: 20019328]

Non-standard Abbreviations and Acronyms

AngII Angiotensin II

AAAs Abdominal aortic aneurysms

AT1aR Angiotensin type 1a receptor

ApoE Apolipoprotein E

LDLR Low density lipoprotein receptor



$waterm

ark-text$w

atermark-text

$waterm

ark-text

Novelty and Significance

What is Known?

• Abdominal aortic aneurysms (AAAs) in humans and AngII-induced AAA inmice exhibit pronounced sexual dimorphism with greater prevalence in malesthan in females.

• In adult mice, testosterone increases angiotensin type 1a receptor (AT1aR) inabdominal aortas where the AAAs form. It also increases the incidence ofAAAs in male mice.

What New Information Does this Article Contribute?

• Exposure of female mice to a single dose of testosterone within 24 h of birth, tomimic testosterone surges in neonatal males, has a long lasting effect inincreasing abdominal aortic AT1aR expression and AngII-induced AAAs inadult females. It also increases AngII-induced atherosclerosis and ascendingaortic aneurysms.

• Smooth muscle cell (SMC)-specific deficiency of AT1aR results in a selectivereduction in the neonatal effects of testosterone in promoting AngII-inducedAAAs, but it has no effect on atherosclerosis or ascending aortic aneurysms.

• Heritability of testosterone-induced increases in AT1aR expression inabdominal aortic SMCs from female, but not male, mice suggests thatepigenetic mechanisms contribute to sexual dimorphism of testosteroneresponses.

AAAs exhibit pronounced sexual dimorphism with much higher prevalence in males.Thus, it is important to identify mechanisms underlying sexual dimorphism of AAAs inorder to understand the pathogenesis and develop more effective strategies for treatingthis life-threatening disease. We report here that transient exposures of female mice totestosterone shortly after birth induces a long-lasting increase in AT1aR mRNAabundance in abdominal aortas and a striking increase in susceptibility to AAA duringadulthood, Transient testosterone exposures in neonatal females promoted atherosclerosisand ascending aortic aneurysms. Our results demonstrate that deletion of AT1aR allelesin SMCs resulted in a selective decrease in AAAs, but not atherosclerosis or ascendingaortic aneurysms induced by exposures of neonatal females to testosterone. Finally,testosterone exerted a region-specific increase in abdominal aortic SMC AT1aRexpression that was heritable in passaged cells from female, but not male mice. Regionaldifferences in androgen receptor expression with increased localization to abdominalaortic SMCs may have contributed to site-specific effects of testosterone.



$waterm

ark-text$w

atermark-text

$waterm

ark-text



$waterm

ark-text$w

atermark-text

$waterm

ark-text



$waterm

ark-text$w

atermark-text

$waterm

ark-text



$waterm

ark-text$w

atermark-text

$waterm

ark-text



$waterm

ark-text$w

atermark-text

$waterm

ark-text



$waterm

ark-text$w

atermark-text

$waterm

ark-text

Figure 1.Administration of testosterone to neonatal females increased AT1aR mRNA abundance inabdominal aortas from adult female ApoE-/- and LDLR-/- mice, but did not change aorticvascular morphology or contractile function in adult females. (A,B) AT1aR mRNAabundance in thoracic and abdominal aortas from ApoE-/- (A) or LDLR-/- (B) female miceadministered vehicle or testosterone (400 μg/mouse) as neonates. Data are mean ± SEMfrom N=8-10/group ApoE-/- mice; N = 5-6 mice/group LDLR-/- mice. *P < 0.001 comparedto thoracic within treatment; **P<0.001 compared to vehicle group within aortic region. (C)Cross sections of aortas (H&E stain) from ascending, thoracic or abdominal aortic regions infemale adult LDLR-/- mice administered vehicle or testosterone as neonates. Scale bars, 100μm. (D) Medial thickness of sections from different aortic regions in female LDLR-/- micefrom each group (N = 3 mice/group). *, P<0.01 compared to thoracic and abdominal regionwithin treatment. (E,F) Contractile responses (expressed as a % of the KCl response, 80mM/L) of aortic rings from different regions to 5-HT (1 μM; E) or AngII (1 μM; F) infemale LDLR-/- mice (N = 5 mice/group). Data are represented as mean ± SEM. * P<0.05suprarenal vs ascending within treatment; ** P<0.05 infrarenal vs ascending and suprarenalwithin treatment.



$waterm

ark-text$w

atermark-text

$waterm

ark-text



$waterm

ark-text$w

atermark-text

$waterm

ark-text

Figure 2.Administration of testosterone to neonatal females significantly increased AngII-inducedAAAs in adult female ApoE-/- mice. (A) AAA incidences in mice infused with 500, 750 or1000 ng/kg/min of AngII for 28 days (N = 10-25/group). *P<0.05 compared to 500 ng/kg/



$waterm

ark-text$w

atermark-text

$waterm

ark-text

min AngII-infused group within treatment; **P<0.01 compared to vehicle within 1000 ng/kg/min AngII-infused group. (B) Maximal external diameters of excised abdominal aortas infemale mice infused with saline (N = 7-10) or AngII (1000 ng/kg/min, N = 14-25) for 28days. Data are summarized as mean ± SEM. *P<0.05 compared to saline infused groupwithin treatment (vehicle, testosterone); **P<0.01 compared to vehicle within 1000 ng/kg/min AngII-infused group. (C) Representative aortas from mice administered vehicle (top) ortestosterone (bottom). Scale bars, 2 mm.



$waterm

ark-text$w

atermark-text

$waterm

ark-text



$waterm

ark-text$w

atermark-text

$waterm

ark-text



$waterm

ark-text$w

atermark-text

$waterm

ark-text



$waterm

ark-text$w

atermark-text

$waterm

ark-text

Figure 3.SMC-specific AT1aR deficiency reduced AngII-induced AAAs in adult female miceadministered testosterone as neonates. AAA incidence (A) and maximal external diameters(B) of abdominal aortas in female mice infused with AngII (1000 ng/kg/min) for 28 days.SMC AT1aR deficiency reduced the incidence and diameter of AAAs. Numeric data aresummarized as mean ± SEM from N = 18-33 mice/group. *P<0.001 compared to vehiclegroup within genotype; **P<0.05 compared to AT1aRfl/fl within treatment. (C)Representative aortas from adult female AT1aRfl/fl and AT1aRSM22 KO mice administeredtestosterone as neonates. Scale bars, 2 mm. (D) AAAs that did form in adult female mice ofeach genotype administered testosterone as neonates exhibited similar maximal externaldiameters (this measurement excluded aneurysm non-responders to AngII infusions).AT1aRfl/fl, N=16 AAAs; AT1aRSM22 KO, N=5 AAAs. (E) Cross sections of AngII-inducedAAAs in the proximal vs distal region of aneurysms from AT1aRfl/fl and AT1aRSM22 KOmice administered testosterone as neonates. Movat’s pentachrome stain for elastin (black),collagen (yellow) and smooth muscle (red). Black arrows indicate severe elastin degradationand medial break. L, lumen. Scale bars, 0.5 mm.



$waterm

ark-text$w

atermark-text

$waterm

ark-text



$waterm

ark-text$w

atermark-text

$waterm

ark-text



$waterm

ark-text$w

atermark-text

$waterm

ark-text



$waterm

ark-text$w

atermark-text

$waterm

ark-text



$waterm

ark-text$w

atermark-text

$waterm

ark-text

Figure 4.Testosterone administration to neonatal females increased AngII-induced atherosclerosisand ascending aortic aneurysms in adult AT1aRfl/fl and AT1aRSM22 KO female LDLR-/-mice. (A) Quantification of atherosclerotic lesions stained with Oil Red O in aortic sinussections (N =5 mice/group). Data are represented throughout the aorta root, with transitionsbetween the sinus and ascending arch depicted as 0. *, P<0.05 compared to vehicle withingenotype. (B) Quantification of CD68 staining as measurement of phagocytes in aortic sinusatherosclerotic lesions (N =5 mice/group). *, P<0.05 compared to vehicle within genotype.(C) Representative sections of atherosclerotic lesions in aortic sinus sections stained withCD68 (phagocytes) or α-actin (SMC). Scale bars, 100 μm. (D) Ascending aortic diametersas an index of ascending aortic aneurysms (N= 18-33 mice/group). *P<0.001 compared tovehicle group within genotype. (E) Cross sections of ascending aortas were stained withH&E (top), or for macrophages (red staining, bottom). The orientation of aortic sections isdescribed by L, lumen; M, media; A, adventitia. Scale bars, 50 μm. F, Medial thickness ofaortic arch sections (N = 5-6 mice/group). *, P<0.001 compared to vehicle within genotype.Data are represented as mean ± SEM.



$waterm

ark-text$w

atermark-text

$waterm

ark-text



$waterm

ark-text$w

atermark-text

$waterm

ark-text



$waterm

ark-text$w

atermark-text

$waterm

ark-text

Figure 5.Administration of a 4-fold lower dose (100 μg/mouse) of testosterone to neonatal femalessignificantly increased AngII-induced AAAs in adult female LDLR-/- mice. (A) AAAincidences (N =10-14 mice/group). (B) Representative aortas from adult female miceadministered vehicle or testosterone as neonates. Scale bars, 2 mm. (C) Maximal externaldiameters of abdominal aortas (N = 10-14 mice/group). Data are summarized as mean ±SEM. (D) Quantification of atherosclerotic lesions stained with Oil Red O in aortic sinussections (N = 5 mice/group). Data are represented throughout the aorta root, with transitionsbetween the sinus and ascending arch depicted as 0. *, P<0.05 compared to vehicle.



$waterm

ark-text$w

atermark-text

$waterm

ark-text



$waterm

ark-text$w

atermark-text

$waterm

ark-text



$waterm

ark-text$w

atermark-text

$waterm

ark-text



$waterm

ark-text$w

atermark-text

$waterm

ark-text



$waterm

ark-text$w

atermark-text

$waterm

ark-text



$waterm

ark-text$w

atermark-text

$waterm

ark-text

Figure 6.Regional specificity and sexually dimorphic heritability of testosterone-induced increased inAT1aR expression in SMCs. (A) Androgen receptor mRNA abundance in primary culturedSMCs from thoracic and abdominal aortas of male and female mice. *, P<0.01 compared tothoracic SMCs; **, P<0.01 compared to male within abdominal SMCs. N = 9-11 replicatesfor male cells and N= 6-7 replicates for female cells. (B) Androgen receptor proteinabundance in primary cultured SMCs from thoracic and abdominal aortas of male andfemale mice. *, P<0.01 compared to thoracic SMCs. N = 3-6 replicates/group. (C, D)Testosterone increased AT1aR mRNA abundance in SMCs from abdominal, but notthoracic aortas from male (C) and female (D) mice. *, P< 0.05 compared to DMSO withinabdominal region; **, P<0.01 compared to thoracic within treatment. N= 5-8 replicates/treatment. (E) AT1aR mRNA abundance in abdominal aortic SMCs passaged from cellspreviously incubated with DMSO or testosterone (10 nM; 24 hours of incubation duringpassage one, cells harvested in passage two). *, P<0.05 compared to DMSO within femalecells. N = 3-6 replicates/treatment. (F, G) Testosterone increased AngII-induced H2O2production in SMCs from abdominal (G), but not thoracic (F) aortas of female mice. Afterincubation with DMSO or testosterone (10 nM), thoracic and abdominal SMCs werestimulated with saline, AngII (1μM) or AngII plus losartan (1μM). H2O2 production wasmeasured using Amplex Red. *, P< 0.001 compared to saline within treatment; **, P<0.01compared to DMSO within AngII treatment. N = 3-6 replicates/group.



$waterm

ark-text$w

atermark-text

$waterm

ark-text

$waterm

ark-text$w

atermark-text

$waterm

ark-text


Tabl

e 1

Cha

ract

eris

tics

of a

dult

fem

ale

AT

1aR

fl/f

l and

AT

1aR

SM22

KO

adm

inis

tere

d ei

ther

veh

icle

or

test

oste

rone

as

neon

ates

.

Gro

ups

NB

W(g

)Se

rum

tes

tost

eron

e (p

g/m

l)T

otal

ser

um c

hole

ster

ol (

mg/

dl)

Pla

sma

aldo

ster

one

(pg/

ml)

Syst

olic

blo

od p

ress

ure

(mm

Hg)

base

line

Ang

II

AT

1aR

fl/f

l + v

ehic

le26

24.0

± 0

.618

5±28

1412

± 1

3673

9 ±

131

121

± 4

154

± 8

†

AT

1aR

SM22

KO

+ v

ehic

le22

24.4

± 0

.414

2±16

1456

± 1

8160

1 ±

123

112

± 5

150

± 8

†

AT

1aR

fl/f

l + te

stos

tero

ne33

26.6

± 0

.6*

155±

1514

24 ±

119

548

± 1

2511

5 ±

315

6 ±

4†

AT

1aR

SM22

KO

+ te

stos

tero

ne18

27.0

± 0

.7*

130±

2014

14 ±

147

456

± 7

711

6 ±

615

3 ±

8†

Dat

a ar

e re

pres

ente

d as

mea

n +

SE

M.

* P<0.

001

com

pare

d to

veh

icle

gro

up w

ithin

gen

otyp

e.

† P<0.

001

com

pare

d to

bas

elin

e bl

ood

pres

sure

.


Transient Exposure of Neonatal Female Mice to Testosterone Abrogates the Sexual Dimorphism of Abdominal Aortic Aneurysms

Documents