Testosterone Deficiency Promotes Arterial Stiffening Independent of Sex Chromosome Complement

Background Testosterone plays a vital role in men’s health. Lower testosterone level is associated with cardiovascular and cardiometabolic diseases, including inflammation, atherosclerosis, and type 2 diabetes. Testosterone replacement is beneficial or neutral to men’s cardiovascular health. Testosterone deficiency is associated with cardiovascular events. Testosterone supplementation to hypogonadal men improves libido, increases muscle strength, and enhances mood. We hypothesized that sex chromosomes (XX and XY) interaction with testosterone plays a role in arterial stiffening. Methods We used four core genotype male mice to understand the inherent contribution of sex hormones and sex chromosome complement in arterial stiffening. Age-matched mice were either gonadal intact or castrated for eight weeks, followed by an assessment of blood pressure, pulse wave velocity, echocardiography, and ex vivo passive vascular mechanics. Results Arterial stiffening but not blood pressure was more significant in castrated than testes-intact mice independent of sex chromosome complement. Castrated mice showed a leftward shift in stress-strain curves and carotid wall thinning. Sex chromosome complement (XX) in the absence of testosterone increased collagen deposition in the aorta and Kdm6a gene expression. Conclusion Testosterone deprivation increases arterial stiffening and vascular wall remodeling. Castration increases Col1α1 in male mice with XX sex chromosome complement. Our study shows decreased aortic contractile genes in castrated mice with XX than XY sex chromosomes.


Background
Endogenous sex hormones, estradiol in women and testosterone in men, prevent CVD; however, it is debatable whether the bene ts are mediated by sex hormones, sex chromosomes, or their interaction [1][2][3].Sexual dimorphism in mammals plays a signi cant role in the development of CVD, including arterial stiffening [4,5].Pulse wave velocity (PWV), an indicator of arterial stiffening, is an independent predictor of cardiovascular mortality and events [6].The Baltimore Longitudinal Study of Aging showed an increase in PWV and blood pressure in men compared to women, widening the sex difference gap [7].
Studies have shown that PWV either precedes or comes after the development of hypertension [8,9].Even so, increasing arterial stiffening is associated with cognitive decline, impairment of kidney function, and CVD in a sex-dependent and independent manner.[10][11][12] Sex chromosomes (XX and XY) independent of sex hormones show sex differences in females and males [13].The male Y chromosome is inherited paternally from father to son [13].The Y chromosome encodes the Sry gene that instructs the formation of male gonads (testes) and spermatogenic pathways [14].The absence of testes makes cells to differentiate into ovaries, indicated by the oogenesis pathway [14].Association the Y chromosome with hypertension is shown in the spontaneously hypertensive rat with identi cation of a Y-linked locus, suggesting a Y chromosome linkage to blood pressure [15,16].Recently, it was shown that the hematopoietic loss of the Y chromosome contributes to cardiac dysfunction [17].Altogether, the contribution of sex chromosomes is equally signi cant to sex hormones in understanding and delineating sex differences in CVD.
The X chromosome comprises less than ve percent of the human genome and plays a signi cant role in sex differences across various tissues, including the heart, aorta, and adipocytes [18][19][20].While females have two X chromosomes, males have one copy [21].For dosage compensation of X-linked genes, one of the X chromosomes in female cells is randomly selected for inactivation [22].X chromosome inactivation (XCI) is initiated and maintained by the long non-coding RNA X inactive speci c transcript (Xist), along with various heterochromatic histone modi cations and DNA methylation [23].The inactive X chromosome is epigenetically distinct from the active X in female cells [24].Genes that escape XCI are expressed from the inactive and active X chromosomes [18].The XCI escapee genes play a crucial role in CVD, including atherosclerosis and brosis in valvular interstitial cells [4,18,25,26].
To understand the contribution of sex hormones, sex chromosomes, and their interaction in CVD in males and females, we used the four-core genotype (FCG) mouse model [27,28].In this mouse model, Sry gene has been deleted from the Y chromosome and inserted into an autosome, resulting in (XX and XY − ) female with ovaries and (XX and XY − Sry ) male with testes [27,28].Therefore, the mice can be used to test for the effect of sex hormones, sex chromosomes, and their interaction.In this study, we sought to clarify whether the interaction of sex hormones and sex chromosomes promotes arterial stiffening in male mice.

Animals
Four core genotype (FCG) mice on C57BL/6J background were used.Studies were conducted in mice between 15-16 weeks old.Male (M) mice XY − Sry were obtained from Dr. Franck Mauvais-Jarvis and were mated with female (F) breeders (XX) on C57bl/6J (RRID: IMSR_JAX:000664) background purchased from Jax Labs (ME, USA).Mating XY − Sry and XX mice produced the four-core genotypes, including XXF, XXM, XY − F, and XY − Sry M. Mice were genotyped with the following primers: Transgene Forward Sry: 5' AGC CCT ACA GCC ACA TGA TA 3', Transgene Reverse Sry: 5' GTC TTG CCT GTA TGT GAT GG 3', Forward: 5' CTG GAG CTC TAC AGT GAT GA 3', Reverse: 5' CAG TTA CCA ATC AAC ACA TCA C 3', Internal positive control: 5' CAA ATG TTG CTT GTC TGG TG 3' and Internal positive control: 5' GTC AGT CGA GTG CAC AGT TT 3'.All mice were maintained at Tulane University in a temperature-controlled vivarium under a 12-hour dark and light cycle with free access to standard chow and drinking water.Animal experiments followed the ARRIVE guidelines [29] approved by Augusta and Tulane University Institutional Animal Care and Use Committee.

Castration
Mice were at eight weeks of age followed by an additional eight weeks to clear endogenous testosterone predominantly from testes.All procedures followed aseptic techniques.Brie y, mice were weighed and placed under a heating pad with 3-4% iso urane-oxygen mixture, eye cream was applied, and hair was shaved around the incision site.Alcohol (70%) and betadine were used to clean incision sites.This was followed by administration of Buprenorphine 0.1 mg/kg.Castration was performed by locating the scrotum and making a small midline cut above the bladder followed by excision of testicles.Muscle tissue was sutured with absorbable suture and the skin was stapled.Mice were left on heating pads to recover and monitored for pain and stress.Awake and alert mice were placed back in their cages with daily monitoring post-surgical monitoring for wound healing and distress.

Tail Cuff Plethysmography
Blood pressure (BP) was measured non-invasively using the CODA 4-Channel Blood Pressure System (Kent Scienti c, Torrington, CT) in conscious mice as previously described [30].Three days of acclimation was followed by one week of BP measurements in the morning.Tail temperature and platform was warmed to 30ºC before beginning the in ation protocol, which consisted of 10 cycles of cuff in ation to 250 mmHg followed by a 20 second de ation.Measurements without a de nitive in ection point indicating the return of blood ow or with a blood ow volume of less than 30 µl were excluded.The average BP was recorded for each day, and the nal BP was reported as the mean of the daily averages after excluding days that were ± 2 standard deviations [30,31].

Pulse Wave Velocity and Echocardiography
PWV was performed as previously described by our group [30,31].Brie y, Vevo® 1100 ultrasound (VisualSonics, Toronto, ON) was used for cardiovascular analysis.Anesthesia was induced using 3% iso urane/oxygen mixture and data obtained under 1.5% iso urane/oxygen mixture.Mice were maintained in supine position on a 37°C heated EKG platform.Shaving cream was applied to the chest, abdomen, and around the throat and wiped with wet gauze.Intracarotid and abdominal pulse wave velocity PWV were measured.Cardiac function was assessed in short axis view in M-mode for left ventricular function.

Passive Biaxial Mechanical Testing
The carotid artery was dissected and cannulated onto 500µm needle secured with a nylon suture on a pressure myograph system in Hank's Balanced Salt Solution as previously described [31,32].Pressure was applied to the lumen of the vessel and the outer diameter was optically tracked.Biaxial phenotyping was performed as previously described [31,32].Pressure-diameter preconditioning was performed from 10 to 150 mmHg.The stretch ratio was assessed by dividing the loaded to unloaded axial length and used to determine the adjusted wall thickness during pressurization.Distensibility was calculated as the percent of starting external diameter.Stress was calculated as (σ) = (P in dyns/cm 2 *D internal )/(2*WT), while strain (ε) = D internal -D 10mmHg /D 10mmHg .

Quantitative Polymerase Chain Reaction
Mice aorta samples were homogenized in lysis buffer (RLT buffer, Qiagen) using bead homogenizer.The total RNA was isolated using RNeasy plus mini kit (Qiagen, cat.No. 74136).Finally, RNA pellet was resuspended in 25µl of Rnase free water, purity and concentrations were measured by nano spectrometer (Implen Nano photometer, N50).cDNA synthesis was performed with superscript IV Mater mix (cat.No. 11766050).Then real time PCR was carried using PowerTrack SYBR master mix (cat.No A46109) on QuantStudio 3 Real-Time PCR System (Applied Biosystems).Comparative cycle method (2- (ΔΔCt) ) was applied for gene expression analysis and β-actin gene was used as an internal control.All the target gene primers were procured from IDT technologies, the IDT predesigned primer IDs were mentioned in the supplementary (Table 1).

Transmission Electron Microscopy
Testicles from XYM and XXM mice were xed in paraformaldehyde and glutaraldehyde in 0.1 M sodium cacodylate (NaCac) buffer, pH 7.4.Post xing was performed in osmium tetroxide in NaCac, stained in a block with uranyl acetate, and dehydration was performed with ethanol series, and embedded in Epon-Araldite resin.The block was trimmed to permit proper orientation of the testicle during imaging.Diamond knife was used to cut sections on a Leica EM UC7 ultramicrotome (Leica Microsystems, Inc, Bannockburn, IL), collected on copper grids, and uranyl acetate and lead citrate was used for staining.JEM 1400 ash transmission electron microscope (JEOL USA Inc., Peabody, MA) tted with a CMOS CCD camera was used for imaging.

Statistics
Data was analyzed using GraphPad Prism version 9.1 (GraphPad Software, San Diego, CA).Outliers were identi ed by the ROUT method (Q = 1%).2-way ANOVA was used to compute the main effect, while Sidak's and uncorrected Fisher's Least Signi cant Difference multiple comparisons test were used to determine the difference between groups.The unpaired t-test was used to compare the area under the curve difference between the two groups.All data are presented as means ± S.E.M, and P < 0.05 was considered signi cant.

Results
Castration increases pulse wave velocity independent of blood pressure.
Castration reduces distensibility in XXM mice at higher pressures.
Gonadectomy decreases compliance and shifts stress-strain curves to the left.
Castration-induced carotid wall thinning and aortic collagen deposition.
To determine which collagen speci cally is increasingly deposited, we performed immuno uorescence (Fig. 4H), showing Col1a1 stained green was increased by castration (Fig. 4F and 4H; P = 0.004) and shown to be higher in gonadectomized than testes intact XXM mice (Fig. 4F and 4H; P = 0.002) but no signi cant difference in XYM aortic cross-sections.Smooth muscle α-actin stained red (Fig. 4F and 4H; P = 0.5) used as a counterstain was decreased in castrated XXM mice, but no statistical signi cance was detected.Representative immuno uorescence staining for αSMA and Col1a1 is shown in (Fig. 4H).
Gonadectomy downregulates contractile genes in the aorta of XX mice.
The decrease in contractile gene expression in GDX XXM but not GDX XYM compared to testes intact mice prompted us to assess X-linked escapee genes.First, we assessed Xist gene expression, distinguishing XX from XY samples.The data shows a signi cant (Fig. 6A; P < 0.0001) increase of Xist in GDX and testes of intact XXM but not XYM mice.Lysine demethylase 5c (Fig. 6B; Kdm5c) was not signi cantly different between groups.However, an increase in Kdm6a was observed in gonadectomized versus testes intact XX mice (Fig. 6C; P = 0.01).Additional escapee genes assessed, including DEAD-Box Helicase 3 X-linked (Ddx3x) and Ubiquitin Speci c Peptidase 9 X-linked (Usp9x), were not signi cantly different between groups (Fig. 6D and Fig. 6E).Eukaryotic translation initiation factor 2 subunit 3, Xlinked (Eif2s3x) indicated decreased expression in castrated XXM versus testes intact mice (Fig. 6F; P = 0.002).
Further evaluation of testicular morphology using a transmission electron microscope revealed structural disorganization in XXM than XYM mice.Leydig cells (Supplementary Fig. 2A; White arrow) were abundant and spread out in XYM than XXM sections.Lipid droplets (Supplementary Fig. 2B; Red arrow) were dense in XXM and scarcely distributed in the XYM section.Endoplasmic reticulum (ER) appeared to be organized and interconnected in XYM; however, XXM mice indicated swollen ER with cisternae disorganization (Supplementary Fig. 2C and D; Red asterisk).Mitochondrial morphology in XYM was elongated and numerous in XYM, but XXM showed swollen circular-like mitochondria (Supplementary Fig. 2C and D; Dollar sign).

Discussion
We report for the rst time that castration induces arterial stiffening independent of sex chromosome complement without signi cantly changing the blood pressure of male four-core genotype (FCG) mice.We show carotid wall thinning and an overall leftward shift of stress-strain curves upon testosterone de ciency, indicating arterial stiffening.The present study shows arterial structural remodeling with increased collagen deposition, including Col1a1, without elastin strand breaks.Due to testosterone de ciency, a substantial decrease in smooth muscle contractile gene expression and an increase in the Xlinked escapee gene Kdm6a is indicated in XXM but not XYM mice.Additionally, we show a structural disorganization of testes in XXM than in XYM mice.The results demonstrate the role of sex chromosomes (XX and XY) and sex hormones (testosterone) in vascular biomechanics.
Aging results in a decline of sex hormones, parallel to an increase in arterial stiffening [7].The use of androgen deprivation therapy and castration in prostate cancer increases arterial stiffening [33].Furthermore, testosterone de ciency is associated with endothelial dysfunction, demonstrating the essential role of testosterone in vascular health [34].Study using FCG mice on MF1 background showed no signi cant difference in blood pressure in XX or XY mice before Angiotensin II infusion recapitulates our ndings, indicating lower but not signi cant systolic blood pressure after gonadectomy [35].Besides blood pressure, PWV was higher in testosterone-de cient men without CVD, suggesting a protective role for testosterone [36].It is shown that hypogonadal males have signi cantly bene ted from testosterone therapy to improve libido and mood [3].Studies have shown that gonadectomy unmasked effects of sex chromosome (XX) in hypertension, atherosclerosis, and ischemia-reperfusion injury in the heart [20,35,37].We showed increased Col1a1 and Kdm6a levels in castrated XX mice that is analogous to a study showing increased adiposity with Kdm5c overexpression, suggesting that the lysine demethylase family of proteins plays an essential role in autosomal gene regulation [19].However, we did not determine the mechanisms of transcriptional activation of Col1a1 by Kdm6a.
Genes that escape X-chromosome inactivation impact CVD, including atherosclerosis and brosis in valvular interstitial cells [4,18,25,26].We show that Kdm6a plays a role in arterial remodeling and is also implicated in autoimmunity [38].Additionally, Kdm5c is involved in adiposity, and recently discovered escapee genes Bms and Stx regulate aortic stenosis progression [19,25].Kdm6a is associated with adiposity of male and female mice with XX sex chromosome complement [19,39].Clinically, men born with an extra X chromosome (Klinefelter syndrome) have an increased incidence of CVD that is not reversed by testosterone replacement [39,40].Given that men with Klinefelter syndrome have XXY sex chromosomes, lysine demethylases may be a druggable target to mitigate CVD, including arterial stiffening and metabolic disorders [40][41][42].Arguably, Klinefelter individuals have a Y chromosome linked to the Sry gene unlike our mouse model (XXM) with an autosomal Sry gene [43].However, the Y chromosome, in comparison to the X, encodes very few genes; even so, hematopoietic mosaic loss of the Y chromosome increases CVD risk in men [17].
Extracellular matrix, including collagen deposition and ber arrangement on the arterial wall, has long been attributed to arterial stiffening and remodeling [30,31].In this study, gonadectomy increased collagen deposition in mice with XX sex chromosome suggesting a role for X-linked genes in driving arterial wall remodeling [20].Our data support a role for Col1a1 in arterial stiffening and remodeling, which aligns with a study in humans and mice showing polymorphism on COL1α1 Sp1 binding site impacts arterial stiffening [44,45].Arterial stiffening is coupled with changes in vascular cells; we show a decrease in Myocd, a master regulator of SMC contractile phenotype changes in a similar trend to Acta2, Calponin, and Myh10, and Myh11 [46].Compelling evidence implicate SMC phenotype switching promotes aortic diseases and our study shows how sex hormones perturbation in mice with XX sex chromosomes is impacted more [47,48].
The arterial geometry, including wall thickness, contributes to vascular distensibility, incremental elastic modulus, and PWV [45,49].Our data shows that carotid walls got thinner while the aorta wall had an increase in collagen deposition, suggesting higher mechanical wall stress associated with aging and hypertension [45,50].Stress-strain curves indicated a leftward shift in all groups after gonadectomy recapitulating PWV assessed in vivo [9,31].The leftward curve shift after gonadectomy primes the vessels to respond to arterial remodeling, predisposing the vessel to impaired ow-mediated dilation, hypertension, atherosclerosis, and abdominal aortic aneurysms [20,35,51].

Perspectives and signi cance
While the FCG mouse is a widely considered model for investigating the role of sex hormones, sex chromosomes, and their interaction, the present study has limitations [28].Unlike wild-type male mice, the FCG male (XYM) model has a Y chromosome independent of the Sry gene that resides in chromosome 3; therefore, we made a general assumption that the chromosomal mutation and Sry transgene location did not skew the cardiovascular phenotype [28].An alternative approach may involve comparing male mice with a Y-chromosome mutation and a Sry transgene (XY − Sry ) with a wild-type male mouse comprising a Y chromosome linked to Sry (XY Sry ).Our study did not assess active vascular properties, including vasoreactivity in smaller arteries shown to impact angiotensin-induced vasodilation involving the type 2 receptor [52].The Traverse study (NCT03518034) recently showed no adverse effect of testosterone replacement in hypogonadal men, which paves the way for novel testosterone therapies in CVD [53].Given that the presence of two X chromosomes and the Sry gene impact on CVD is rarely studied, [40] future studies can utilize the mouse XY* model that allows for the generation of mice with XXY sex chromosomes present in Klinefelter men to understand the role of sex chromosomes in vascular disease [43].

Figure 1 Increased
Figure 1