Greater TIMP‐1 protein levels and neointimal formation represent sex‐dependent cellular events limiting aortic vessel expansion in female rats

Abstract

Fragmentation/loss of the structural protein elastin represents the precipitating event translating to aortic expansion and subsequent aneurysm formation. The present study tested the hypothesis that greater protein expression of tissue inhibitor of matrix metalloproteinase‐1 (TIMP‐1) and neointimal growth secondary to a reduction of medial elastin content represent sex‐dependent events limiting aortic vessel expansion in females. TIMP‐1 protein levels were higher in the ascending aorta of female versus male patients diagnosed with a bicuspid aortic valve (BAV). The latter paradigm was recapitulated in the aorta of adult male and female rats complemented by greater TIMP‐2 expression in females. CaCl₂ (0.5 M) treatment of the infrarenal aorta of adult male and female rats increased the in situ vessel diameter and expansion was significantly smaller in females despite a comparable reduction of medial elastin content. The preferential appearance of a neointimal region of the CaCl₂‐treated infrarenal aorta of female rats may explain in part the smaller in situ expansion and neointimal growth correlated positively with the % change of the in situ diameter. Neointimal formation was secondary to a significant increase in the density of medial/neointimal vascular smooth muscle cells (VSMCs) that re‐entered the G₂‐M phase whereas VSMC cell cycle re‐entry was attenuated in the CaCl₂‐treated infrarenal aorta of male rats. Thus, greater TIMP‐1 expression in the aorta of female BAV patients may prevent excessive elastin fragmentation and preferential neointimal growth following CaCl₂‐treatment of the infrarenal aorta of female rats represents a sex‐dependent biological event limiting vessel expansion secondary to a significant loss of the structural protein.

1. INTRODUCTION

An aortic aneurysm is a focal dilatation resulting from progressive weakening of the vessel wall translating to a potential life‐threatening disorder. ¹ , ² , ³ The incidence of thoracic (TAA) and abdominal aortic aneurysms (AAA) are greater in men versus women and affects 1%–5% of the population over the age of 65. ³ , ⁴ In ~30% of patients diagnosed with TAA, the disease is triggered by genetic mutations whereas sporadic‐driven is the most prevalent cause of TAA/AAAs and risk factors include age, male sex, smoking and hypertension. ³ , ⁴ , ⁵ , ⁶ However, despite sex‐related protection, women that develop a significant aortic aneurysm were associated with worse prognosis. ³ , ⁴ A recent study revealed that the growth rate of an aortic aneurysm was selectively greater in women diagnosed with a degenerative sporadic‐driven TAA. ⁷ The precipitating event leading to aortic vessel expansion and subsequent aneurysm formation is the fragmentation/loss of the structural protein elastin organized as laminae providing morphological integrity to elastic arteries. ⁶ , ⁸ Increased activity/expression of matrix metalloproteinases (MMPs) attributed in part to the secondary appearance of senescent vascular smooth muscle cells (VSMCs) during vessel expansion facilitates elastin fragmentation/loss by overriding the protective role of tissue inhibitors of matrix metalloproteinases (TIMPs). ⁹ , ¹⁰ , ¹¹ , ¹² , ¹³ , ¹⁴ Clinical studies have reported that the TIMP‐1 gene escaped X‐chromosome inactivation in females leading to biallelic activation and significantly greater protein expression versus males. ¹⁵ , ¹⁶ The reported cardioprotective role of TIMP‐1 in experimental models of vessel expansion supports the premise that greater expression of the protein in females may represent a seminal event reducing the risk of aortic vessel expansion and aneurysm formation. ¹⁴ , ¹⁷

Neointimal formation represents an additional biological paradigm reducing the risk of aortic vessel expansion in the presence of progressive elastin fragmentation/loss. ¹⁸ , ¹⁹ , ²⁰ , ²¹ , ²² Studies have reported that elastin loss triggered vascular smooth muscle cell (VSMC) proliferation. In the latter scenario, loss of the structural protein secondary to increased MMP activity and/or reduced TIMP expression provided the requisite environment for proliferating VSMCs to migrate to the damaged region leading to neointimal formation. ¹⁸ , ¹⁹ , ²⁰ , ²¹ , ²² The appearance of a neointimal region translates to an overall increase in vessel thickness thereby limiting or delaying expansion/dilation following damage. ¹⁸ , ¹⁹ , ²⁰ , ²¹ , ²² Clinical studies have revealed that neointimal formation may be influenced in part by sex as the growth and rate of in‐stent restenosis of coronary arteries were greater in female versus male patients. ²³ , ²⁴ , ²⁵ Despite the latter data, an experimental model recapitulating sex‐dependent neointimal formation has not been identified. Thus, the first series of experiments tested the hypothesis that the reported lower risk of aortic aneurysm formation in female versus male patients diagnosed with a BAV was associated with inherently greater expression of TIMP‐1 protein levels in the ascending aorta. ²⁶ , ²⁷ , ²⁸ , ²⁹ The latter premise will be further examined in rodents to ascertain whether sex‐dependent TIMP‐1 protein levels in the human aorta is conserved in adult rodents. Lastly, based on the clinical data that the rate of in‐stent restenosis was greater in the coronary artery of female patients, additional experiments tested the hypothesis that elastin fragmentation/loss of the infrarenal aorta of adult male and female rats following CaCl₂ (0.5 M) treatment preferentially triggered neointimal growth in female rats.

2. METHODS

2.1. Ethics approval for the use of human tissue and animals for experimental use

The use of human tissue was approved by the Montreal Heart Institute Scientific Ethics Committee on Human Research. Methods were performed in accordance with the guidelines and regulations stipulated by the Montreal Heart Institute Scientific Ethics Committee on Human Research. The guidelines and regulations were formulated by the Canadian Tri‐Council (Canadian Institutes of Health Research, Natural Sciences and Engineering Research Council of Canada Social Sciences and Humanities Research Council) policy of Ethical Conduct for Research Involving Humans. Guidelines and regulations were derived from various organizations including those indicated in the Helsinki Declaration on human research. Informed consent was obtained from each patient that participated in the study prior to surgery. The use and care of laboratory rodents was according to the Canadian Council for Animal Care and the protocol (Projects # 2021‐2915, 2020‐82‐02) was approved by the Animal Care Committee of the Montreal Heart Institute. Lastly, the data presented is in accordance with the ARRIVE guidelines.

2.2. Comparison between adult mice and rats, ovariectomy of adult female rats and CaCl₂ model of infrarenal aortic expansion in adult rats

Adult male and female Sprague–Dawley rats (9–11 weeks old; Charles Rivers, Canada) were anesthetized with isoflurane (2%) and the ascending aorta, aortic arch, thoracic and abdominal (suprarenal and infrarenal region) aortic regions were collected and rapidly frozen in liquid nitrogen for protein analysis. Adult male and female C57/Bl6 mice (10–12 weeks old; Charles Rivers, Canada) were anesthetized with isoflurane (2%) and the thoracic and abdominal aorta (suprarenal and infrarenal regions) were collected and rapidly frozen in liquid nitrogen for protein analysis.

Adult female Sprague–Dawley rats (9–11 weeks old; Charles Rivers, Canada) underwent sham operation or bilateral ovariectomy and sacrificed 4 weeks later, as previously described. ³⁰ Prior to sacrifice, body weight was determined, the uterus excised and weighed and the various regions of the aorta were collected and rapidly frozen in liquid nitrogen for protein analysis.

The infrarenal aorta of adult male and female Sprague–Dawley rats (9–11 weeks old; Charles Rivers, Canada) was treated with saline or CaCl₂ (0.5 M) for 15 min. ³¹ , ³² Prior to surgery, buprenorphine (0.5 mg/kg) was administered once and rats anesthetized with isoflurane (2%). Prior to the application of saline or CaCl₂, multiple measurements were made along the length of the infrarenal aorta with a digital electronic calliper (Fine Science Tools, British Columbia, Canada) to assess the in situ diameter. Thereafter, gauze was immersed in saline or 0.5 M CaCl₂ and applied to the ventral region of the infrarenal aorta for a period of 15 min. The infrarenal aorta was subsequently rinsed with saline and the abdomen sutured. Immediately after surgery, buprenorphine (0.5 mg/kg) was administered and repeated two times at 6‐h intervals. Four weeks following surgery, rats were anesthetized with isoflurane (2%) and the infrarenal aorta exposed. Prior to excision, multiple measurements were made along the length of the infrarenal aorta to assess the increase of the in situ diameter. The infrarenal aorta of saline‐treated male/female rats and a subpopulation of CaCl₂‐treated male (n = 7) and female (n = 7) rats was cut in two and one section was fixed in formalin for immunohistochemistry/immunofluorescence analysis and the remaining section was rapidly frozen in liquid nitrogen for protein analysis. The CaCl₂‐treated infrarenal aorta of the remaining male and female rats employed in the present study was excised and rapidly frozen in liquid nitrogen. With regard to the subpopulation of rats used for immunohistochemistry/immunofluorescence analysis, the data acquired from the CaCl₂‐treated infrarenal aorta of a single male and female rat was consistently weak, uninterpretable and excluded from the study. Lastly, a neointimal region was identified in the CaCl₂‐treated infrarenal aorta in six of seven female rats examined whereas a modest neointimal response was observed in the CaCl₂‐treated infrarenal aorta of a single male rat.

2.2.1. Histology

Formalin‐fixed vascular tissue sections (6–8 μm thick) of the ascending aorta of BAV patients and the infrarenal aorta of adult rats were stained with Masson's trichrome and/or Movat's pentachrome to assess collagen and elastin levels, respectively. ³³ Within each tissue examined, 4–5 arbitrary regions (area; ~0.31 mm²) were examined at 20× magnification to avoid biased sampling of the immunohistochemical analysis. The surface area of black‐stained elastin fibers or blue‐stained collagen fibers was measured using Image Pro Software and the data depicted as the average % of elastin ((mm²)/media (mm²) × 100) or % of collagen ((mm²)/media (mm²) × 100) content. The average medial area (mm²), lumen area (mm²), and medial thickness (mm) of the ascending aorta of male and female BAV patients and the infrarenal aorta of adult male and female rats were determined in Movat's pentachrome stained tissue using Image Pro Software. Lastly, the Von Kossa stain was used to detected calcium deposits in the infrarenal aorta of male and female rats following exposure to 0.5 M CaCl₂. ³²

2.2.2. Immunofluorescence

Tissue sections of the ascending aorta of BAV patients and infrarenal aorta of adult rats (6–8 μm thick) were formalin‐fixed and subjected to immunofluorescence, as previously described. ³³ Primary antibodies employed include mouse monoclonal anti‐nestin (1:200; RRID:AB_305313, Abcam), rabbit monoclonal anti‐non‐muscle myosin IIB (NMIIB; 1:150; catalogue #ab230823, Abcam), mouse monoclonal anti‐p16^INK4a (1:400; RRID:AB_881819, Abcam), rabbit polyclonal anti‐phosphorylated serine ¹⁰ residue of histone 3 (PHH3, 1:200; RRID:AB_304763, Abcam) and chicken polyclonal anti‐nestin (1:150, RRID:AB_2753197, Abcam). The nucleus was identified with 4′,6′‐diamidino‐2‐phenylindole (DAPI, Sigma‐Aldrich) staining. Secondary antibodies used were goat anti‐mouse IgG conjugated to Alexa‐555 or Alexa‐647 (1:600; Life Technologies) and a goat anti‐rabbit IgG conjugated to Alexa‐488 or Alexa‐647 (1:600; Life Technologies). Immunofluorescence was visualized with a 10X‐ or 63X‐oil 1.4 NA DIC plan apochromat objective mounted on a Zeiss Axiovert 100M confocal microscope. Non‐specific staining was determined following the addition of the conjugated secondary antibody alone. To determine cell density, each NMIIB‐ and nestin‐immunoreactive vascular smooth muscle cell that co‐expressed p16^INK4a or phosphorylated serine ¹⁰ residue of histone 3 was counted and normalized to the surface area of the region examined. The density of VSMCs was determined in 4–5 arbitrary regions of each vessel and averaged.

2.2.3. Western blot

Protein lysates were prepared from the ascending aorta of BAV patients, the ascending aorta, aortic arch, thoracic aorta and abdominal aorta of male/female mice and rats, ovariectomized female rats and the infrarenal aorta of adult rats, as previously described. ³³ In the subpopulation of CaCl₂‐treated infrarenal aorta of female rats used for western blot analysis, the presence of a neointimal region was not determined. Assuming that the neointimal response of the infrarenal aorta of female rats to CaCl₂ is conserved in the majority of rats examined, protein levels in this specific subpopulation of rats will include the medial and neointimal regions and compared exclusively to the medial region of the infrarenal aorta of normal female rats. Protein lysates were subjected to SDS‐electrophoresis and antibodies used include chicken polyclonal anti‐nestin (1:150, RRID:AB_2753197, Abcam), mouse monoclonal anti‐smooth muscle 22α (1:5000; Santa Cruz Biotechnology), rabbit polyclonal anti‐matrix metalloproteinase‐2 (MMP‐2; recognizes proenzyme/cleaved enzyme, 1:1000; Cell Signaling), rabbit polyclonal anti‐p27^kip1 (1:2000; Santa Cruz Biotechnology), rabbit monoclonal anti‐non‐muscle myosin IIB (NMIIB; 1:500; catalogue #ab230823, Abcam), mouse monoclonal anti‐tissue inhibitor of matrix metalloproteinases (TIMP‐1, 1:500; RRID:AB_10848565; Santa Cruz), mouse monoclonal anti‐TIMP‐2 (1:500; RRID:AB_628360; Santa Cruz), mouse monoclonal anti‐p53 (1:500; RRID:AB_628082, Santa Cruz), mouse monoclonal anti‐p53 (1:500, RRID:AB_331743; Cell Signaling), mouse monoclonal anti‐p21 (1:500; RRID:AB_628073, Santa Cruz) and a mouse monoclonal anti‐GAPDH (1:10,000, RRID:AB_437392; Ambion, Austin, TX). Following overnight incubation at 4°C, the appropriate secondary antibody‐conjugated to horseradish peroxidase (1:20,000, Jackson Immunoresearch, West Grove, PA) was added and bands visualized utilizing the ECL detection kit (Perkin Elmer, Waltham, MA). Films were scanned with Image J software® and the target protein signal was depicted as arbitrary light units.

2.2.4. Statistics

Data were presented as the mean ± SD, (n) represents the number of BAV patients or adult mice/rats used for each experimental study. In Table S1 comparing male and female BAV patients, statistical significance of the data was determined by t‐test, χ ² test or Fisher's exact test and p < .05 was considered statistically significant. Morphological data, immunohistochemistry staining of elastin and collagen, the density of non‐muscle myosin IIB⁽⁺⁾‐ and nestin⁽⁺⁾‐VSMCs expressing p16^INK4a or phosphorylated serine ¹⁰ residue of histone 3 were evaluated by a Students' unpaired t‐test and a significant difference determined by a value of p < .05 (Origin 2016; Northhampton, MA). The western blot data between male and female BAV patients and saline‐ and CaCl₂‐treated male and female infrarenal aorta were also evaluated by a Students' unpaired t‐test and a significant difference determined by a value of p < .05 (Origin 2016). The % change of the in situ diameter of the infrarenal aorta, the distribution and density of the perinuclear and nuclear p16^INK4a staining of non‐muscle myosin IIB⁽⁺⁾‐VSMCs of the saline‐ and CaCl₂‐treated infrarenal aorta within each female and male rat were evaluated with a Students' paired t‐test and a significant difference determined by a value of p < .05 (Origin 2016). A Students' paired t‐test was used to compare cellular remodeling of the medial and neointimal regions of the CaCl₂‐treated infrarenal aorta of each female rat. Data between the ascending aorta, aortic arch, thoracic aorta and abdominal aorta between normal male and female rats and between ovariectomized and normal female rats were evaluated by a one‐way ANOVA followed by a Fisher LSD post‐hoc test and a significant difference determined by a value of p < .05 (Origin 2016). Correlation between the % change of the in situ diameter of the infrarenal aorta and elastin content, neointimal area and the density of p16^INK4a‐VSMCs were determined with the program Origin.

3. RESULTS

3.1. Sex‐dependent pattern of TIMP‐1 protein expression and density of p16^{INK4a (+)}‐vascular smooth muscle cells in patients diagnosed with a bicuspid aortic valve

The average age of male and female BAV patients examined in the present study was 55 and 52 years old, respectively (Table S1). The average age of natural menopause in Canadian women is 51 suggesting that the majority of female BAV patients examined in the present study were pre‐menopausal or menopausal. ³⁴ Furthermore, male and female BAV patients were diagnosed predominantly with aortic stenosis and the diameter of the aortic annulus and aortic root were significantly larger in males (Table S1). The medial width of the ascending aorta of female BAV patients was significantly smaller compared to male BAV patients whereas collagen and elastin content of the medial region were similar between sexes (Figure 1A,B). Despite significant vessel expansion of the ascending aorta, the presence of a focal dilatation indicative of an aortic aneurysm was absent in each patient examined. The absolute value of the diameter of the ascending aorta in the BAV patients examined at the time of surgery was characterized as moderate as surgical intervention is strongly recommended when the diameter is greater than 50 mm. ³⁵ , ³⁶

FIGURE 1.

Morphological and cellular remodeling of the ascending aorta of male and female bicuspid aortic valve (BAV) patients. (A) Medial elastin and collagen content were comparable in the ascending aorta of male (MBAV) and female (FBAV) patients whereas the medial width was significantly lower (Students' unpaired test; * denotes p < .05) in FBAV patients. (B) Nestin, MMP‐2, p27, and SM22α protein levels in the ascending aorta of FBAV and MBAV patients were not significantly different (Figure S1). By contrast, non‐muscle myosin IIB (NMIIB) protein levels were significantly lower whereas TIMP‐1 expression significantly increased (Students' unpaired test; * denotes p < .05) in the ascending aorta of FBAV versus MBAV patients. Protein expression was normalised to GAPDH protein levels. (C) In the ascending aorta of a subpopulation of MBAV and FBAV patients, immunostaining of the cell cycle inhibitor p16^INK4a (grey fluorescence) revealed a predominant nuclear signal in NMIIB⁽⁺⁾‐vascular smooth muscle cells (VSMCs, green fluorescence) whereas a modest number of VSMCs was associated with perinuclear p16^INK4a staining. The density of NMIIB⁽⁺⁾‐VSMCs in the ascending aorta of a subpopulation of MBAV and FBAV patients was comparable whereas the density of NMIIB/p16^INK4a(+)‐VSMCs staining was significantly lower (* denotes p < .05) in FBAV patients.

Nestin and smooth muscle 22α protein levels in the ascending aorta were similar between sexes whereas non‐muscle myosin IIB expression was significantly lower in the population of female BAV patients compared to male BAV patients (Figures 1B and S1). Protein expression of the matrix metalloproteinase MMP‐2 and the cell cycle inhibitor p27^kip1 in the medial region of the ascending aorta were similar between sexes (Figures 1B and S1). Previous studies have reported that the tissue inhibitor of matrix metalloproteinase‐1 (TIMP‐1) gene escaped X‐Chromosome inactivation in females providing an inherent sex‐dependent mechanism limiting elastin fragmentation/loss. ¹⁵ , ¹⁶ TIMP‐1 protein levels were significantly higher in the population of female BAV patients compared to male BAV patients (Figure 1B). Despite statistical significance, TIMP‐1 protein levels in the ascending aorta at least 2–3 female BAV patients were comparable to male BAV patients suggesting a potential greater risk of elastin fragmentation/loss in this female BAV subpopulation. To examine the density of p16^INK4a(+)‐VSMCs in the ascending aorta, a subpopulation of male and female BAV patients with comparable expression of non‐muscle myosin IIB protein levels were examined (Figure 1B). Consistent with the latter approach, the density of non‐muscle myosin IIB⁽⁺⁾‐VSMCs in the ascending aorta of a subpopulation of male and female BAV patients was not significantly different (Figure 1C). In the ascending aorta of male and female BAV patients, p16^INK4a immunoreactivity was detected predominantly in the nucleus of non‐muscle myosin IIB⁽⁺⁾‐VSMCs whereas as a modest number was associated with an apparent perinuclear stain (Figure 1C). The total density of non‐muscle myosin IIB⁽⁺⁾‐VSMCs associated with p16^INK4a immunoreactivity (regardless the cellular localization of the cell cycle inhibitor) in the ascending aorta was significantly lower in the subpopulation of female BAV patients (Figure 1C).

3.2. Sex‐dependent expression of TIMP‐1, TIMP‐2 and p53 identified in the aorta of adult rats whereas the paradigm was not observed in adult C57/Bl6 mice

The aorta of normal adult male and female Sprague–Dawley rats (9 weeks old) was divided into four distinct regions as the ascending aorta and aortic arch are neural crest derived whereas the thoracic and abdominal aorta (consists of suprarenal and infrarenal regions) originate from the mesoderm during embryogenesis. ³⁷ , ³⁸ Despite the absence of a difference between sexes, non‐muscle myosin IIB protein expression was significantly greater in the ascending aorta and aortic arch compared to the thoracic and abdominal aorta of adult male and female rats (Figure 2A). An analogous paradigm was identified in the normal male rat aorta as protein levels of the intermediate filament protein nestin were greater in the aortic arch versus thoracic and abdominal regions of the aorta. ³⁹ Matrix metalloproteinase (MMP‐2) protein levels in the ascending aorta, thoracic aorta and abdominal aorta of adult female rats was not significantly different from male rats but a significant difference between sexes was observed in the aortic arch (Figure 2A). In parallel, greater TIMP‐1 protein expression identified in the ascending aorta of female BAV patients versus male patients (Figure 1A) was recapitulated in normal adult Sprague–Dawley rats as protein levels in each region of the aorta of the female rat were significantly greater versus male rats (Figure 2B). In parallel, TIMP‐2 protein levels were also higher in each region of the adult female rat aorta compared to adult male rats (Figure 2B). However, in contrast to TIMP‐1, the TIMP‐2 gene is located on chromosome 10 in rats (National Library of Medicine) and not subjected to X‐chromosome inactivation. In parallel, protein levels of the cell cycle inhibitor p53 were higher in each region of the aorta of adult female rats compared to adult male rats and the gene is also located on chromosome 10 in rats (National Library of Medicine) (Figure 2C). Expression of p21^cip1 is directly influenced by p53 and protein levels of the cell cycle inhibitor were significantly higher in each region of the aorta of adult female rats compared to male rats (Figure 2C). ⁴⁰ In the thoracic and abdominal regions of the aorta of C57/Bl6 mice, TIMP‐1 protein levels were significantly lower in normal adult female versus adult male mice (Figures 2D and S2). By contrast, p53 protein levels in the thoracic and abdominal regions of the aorta of C57/Bl6 female mice was variable but not significantly different from adult male mice (Figures 2D and S2). However, protein levels of the cell cycle inhibitor p21^cip1, a downstream target of p53, was significantly reduced in the abdominal aorta of female adult mice compared to normal adult male mice (Figure 2D).

FIGURE 2.

Sex‐dependent phenotypic differences in the aorta of the normal adult rodent. (A) Non‐muscle myosin IIB (NMIIB) protein levels in the ascending aortic (AA) and aortic arch (AAr) were significantly greater (one‐way ANOVA; * denotes p < .05 vs. AA/AAr in males and *p < .05 vs. AA/AAr in females) versus the thoracic (TA) and abdominal aorta (AbA) within each normal adult male (n = 3) and female (n = 3) rat. MMP‐2 protein levels were significantly lower (one‐way ANOVA; * denotes p < .05) in the AAr of male (n = 3) vs. female (n = 3) rats. Protein expression was normalised to GAPDH protein levels. (B) TIMP‐1 protein levels in each region of the female rat aorta (n = 3–4) were significantly greater (one‐way ANOVA; * denotes p < .05) compared to the identical region of the aorta of normal male rats (n = 4). TIMP‐2 protein levels (AA, 1.33 ± 0.08; AAr, 1.43 ± 0.05; TA, 1.82 ± 0.18; AbA, 3.43 ± 0.80) in each region of the female rat aorta (n = 2) were greater compared to the identical region of the normal male rat aorta (n = 2) (AA, 0.24 ± 0.014; AAr, 0.38 ± 0.015; TA, 0.21 ± 0.014; AbA, 0.495 ± 0.15). (C) Protein levels of p53 and p21^cip1 in each region of the female rat aorta (n = 3–4) were significantly greater (one‐way ANOVA; * denotes p < .05) compared to the identical region of the aorta of normal male rats (n = 3–4). Protein expression was normalised to GAPDH protein levels. (D) TIMP‐1, p53 and p21^cip1 protein levels in the abdominal aorta of normal male C57/Bl6 mice (n = 3) were higher compared to normal female C57/Bl6 mice (n = 3) and a significance difference (Students' unpaired test; * denotes p < .05) was observed for TIMP‐1 and p21^cip1. Protein expression was normalised to GAPDH protein levels.

3.3. The impact of ovariectomy on the pattern of protein expression in the aorta of adult female rats

Escape of the TIMP‐1 gene from X‐chromosome inactivation supports the premise that ovarian sex hormones did not influence protein expression in the adult female rat. However, ovarian sex hormones may have contributed in part to the greater expression of TIMP‐2 and p53 in the aorta of adult female rats. Thus, adult female rats were ovariectomized and TIMP‐1, TIMP‐2, and p53 protein levels in each region of the aorta examined 4‐weeks later. The loss of circulating ovarian hormones was confirmed via a significant increase in body weight gain and concomitant uterine atrophy of ovariectomized adult female rats (Figure 3A). In the aorta of ovariectomized female rats, TIMP‐1 protein levels were significantly upregulated in the ascending, thoracic and abdominal regions as compared to normal adult female rats (Figure 3B). TIMP‐2 protein levels were likewise upregulated in the aorta of ovariectomized adult female rats and a significance difference was observed in the ascending, thoracic and abdominal regions compared to normal adult female rats (Figure 3B). By contrast, p53 protein levels in each region of the aorta of ovariectomized adult female rats were similar to normal adult female rats (Figure 3B).

FIGURE 3.

Cellular remodeling of the aorta of the ovariectomized adult female rat. (A) The delta change in the body weight of ovariectomized (OVX) female rats (n = 4) was significantly higher (Students' unpaired test; * denotes p < .05) compared to sham female rats (n = 4) whereas the uterine weight was significantly reduced (Students' unpaired test; * denotes p < .05) in the OVX female rat. (B) TIMP‐1 and TIMP‐2 protein levels were significantly greater (one‐way ANOVA; * denotes p < .05 comparing the identical region of the aorta) of various regions of the aorta (AbA, abdominal; AAr, aortic arch; AA, ascending aortic; TA, thoracic) of OVX female rats (n = 3) versus sham female rats (n = 3). The protein levels of p53 within each aortic region of the OVX female rat (n = 3) were comparable to sham female rats (n = 3).

3.4. Sex‐dependent differences in the cell cycle re‐entry of medial VSMCs and expression of the cell cycle inhibitor p16^INK4a in the infrarenal aorta normal adult rats

The total density of non‐muscle myosin IIB⁽⁺⁾‐ (male, 2027 ± 368 cells/mm², n = 6; female, 1863 ± 352 cells/mm², n = 6) and nestin⁽⁺⁾‐VSMCs (male, 2479 ± 289 cells/mm², n = 6; female, 2289 ± 355 cells/mm², n = 6) in the medial region of the infrarenal aorta of normal adult male and female rat were comparable. Despite a comparable density of VSMCs in the medial region of the infrarenal aorta of normal adult rats, the density of nestin⁽⁺⁾‐VSMCs that re‐entered the G₂‐M phase of the cell cycle characterized by the nuclear appearance of the phosphorylated serine ¹⁰ residue of histone‐3 (PHH3) was significantly higher (p = 8.9E−05) in males (1055 ± 271 cells/mm²; n = 6) versus females (262 ± 74 cells/mm²; n = 6). ³⁹

Immunofluorescent staining revealed that the cell cycle inhibitor p16^INK4a was identified in the perinuclear or nuclear region of medial non‐muscle myosin IIB⁽⁺⁾‐VSMCs of the infrarenal aorta of normal male and female rats (Figure 4A). The density of medial non‐muscle myosin IIB⁽⁺⁾‐VSMCs associated with perinuclear p16^INK4a staining was significantly higher versus the density of nuclear p16^INK4a staining of the infrarenal aorta of normal male (non‐muscle myosin IIB/perinuclear p16^INK4a, 884 ± 249 vs. non‐muscle myosin IIB/nuclear p16^INK4a, 658 ± 265 cells/mm²; p = .044, n = 6; Student's paired t‐test) and female rats (non‐muscle myosin IIB/perinuclearp16^INK4a, 595 ± 132 vs. non‐muscle myosin IIB/nuclear p16^INK4a, 452 ± 54 cells/mm²; p = .0018, n = 6; Student's paired t‐test). However, the density of medial non‐muscle myosin IIB⁽⁺⁾‐VSMCs associated with nuclear p16^INK4a (p = .1284, n = 6) or perinuclear p16^INK4a (p = .0534, n = 6) in the infrarenal aorta between normal male and female rats were not significantly different. Lastly, the nuclear/perinuclear ratio of p16^INK4a staining of non‐muscle myosin IIB⁽⁺⁾‐VSMCs in the medial region of the infrarenal aorta of normal adult male (0.79 ± 0.28; n = 6) and female rats (0.78 ± 0.10; n = 6) was identical.

FIGURE 4.

Cellular localization of the cell cycle inhibitor p16^INK4a in vascular smooth muscle cells (VSMCs) of saline and CaCl₂‐ treated infrarenal aorta of male rats. (A) In the saline‐treated infrarenal aorta of male rats, staining of p16^INK4a (red fluorescence) was identified in the perinuclear (indicated by arrow) and nuclear regions (indicated by asterisk) of medial residing non‐muscle myosin IIB (NMIIB; green fluorescence)⁽⁺⁾‐VSMCs. In the CaCl₂‐treated infrarenal aorta of male rats, a redistribution of p16^INK4a (red fluorescence) from the perinuclear (indicated by arrow) to nuclear (indicated by asterisk) region was observed in NMIIB⁽⁺⁾‐VSMCs (green fluorescence). Furthermore, a modest neointimal region was observed in the CaCl₂‐treated infrarenal aorta of a single male rat characterized by the appearance of NMIIB⁽⁺⁾‐VSMCs predominantly associated with nuclear p16^INK4a staining. The density of NMIIB⁽⁺⁾‐cells expressing p16^INK4a residing in the adventitial region was not determined in the present study. The nucleus was identified with 4′,6′‐diamidino‐2‐phenylindole (DAPI) staining (blue fluorescence).

3.5. Sex‐dependent morphological remodeling of the CaCl₂ ‐infrarenal rat aorta

The infrarenal aorta of adult male and female Sprague–Dawley rats was treated with saline or 0.5 M CaCl₂ for 15 min. Four weeks later, the medial region of the CaCl₂‐treated infrarenal aorta of male and female rats was distinguished by the appearance of calcium deposits (Figure S3). The in situ diameter of the saline‐treated infrarenal aorta of male and female adult rats did not significantly change (Figure 5A). Four weeks following the acute exposure to CaCl₂, the in situ diameter of the infrarenal aorta of adult male and female rats was significantly increased compared to saline‐treated rats (Figure 5A–C). Moreover, the in situ diameter of the CaCl₂‐treated infrarenal aorta of female rats was significantly smaller compared to the CaCl₂‐treated infrarenal aorta of adult male rats (Figure 5A–C). Immunohistochemical analysis of the male rat infrarenal aorta revealed that the lumen and medial areas were significantly larger following CaCl₂ treatment whereas medial thickness was significantly reduced compared to the saline‐treated infrarenal aorta (Figure 5D). The lumen area and medial thickness of the CaCl₂‐treated infrarenal aorta of female rat were variable within the group and did not reach a statistical significance whereas the medial area was significantly higher compared to saline‐treated female rats (Figure 5D). The variable lumen area and the smaller increase of the in situ diameter of the infrarenal aorta reported in CaCl₂‐treated adult female rats was attributed at least in part to a sex‐dependent appearance of a neointimal region in six of seven female rats examined (Figure 5A,C). The magnitude of the neointimal response of the CaCl₂‐treated infrarenal of adult female rats correlated positively and significantly with the % change of the in situ diameter of the infrarenal aorta (Figure 5A). By contrast, a modest neointimal response was observed in the CaCl₂‐treated infrarenal aorta of a single male rat among the population of seven adult male rats examined (Figure 4B).

FIGURE 5.

Morphological remodeling of the CaCl₂‐treated infrarenal aorta of adult male and female rats. (A) Four weeks after CaCl₂ (0.5 M) treatment of the infrarenal aorta of adult male (n = 11) and female (n = 11) rats, significant vessel expansion was observed (Students' paired test; * denotes p < .05) whereas exposure to saline (males, n = 7; females n = 7) had no significant effect. The % change of the in situ diameter of the CaCl₂‐treated infrarenal aorta of female rats was significantly smaller (Students' unpaired test; * denotes p < .05) compared to that observed in male rats. The latter paradigm was attributed in part to the appearance of a neointimal region of the CaCl₂‐treated infrarenal aorta in 6 of 7 female rats and the magnitude of the response positively and significantly correlated with the % change of the in‐situ diameter of the infrarenal aorta. (B & C) Movat's pentachrome stained revealed expansion of the CaCl₂‐treated infrarenal aorta of male and female rats versus saline‐treated rats. Furthermore, a neointimal region was evident in the CaCl₂‐treated infrarenal aorta of female rats. (D) The lumen and medial area were significantly increased (Students' unpaired test; * denotes p < .05) whereas the medial thickness was significantly reduced following CaCl₂ treatment of the infrarenal aorta of male rats versus the saline‐treated male rats. The lumen area and medial thickness area were not significantly different whereas medial area was significantly increased (Students' unpaired test; * denotes p < .05) following CaCl₂ treatment of the infrarenal aorta of female rats versus the saline‐treated infrarenal aorta of female rats.

In the absence of atherosclerosis, neointimal formation was induced following selective elastin fragmentation/loss in the medial region of a vessel. ²⁰ , ²¹ Elastin content of the medial region of the CaCl₂‐treated infrarenal aorta of female rats was significantly reduced compared to the saline‐treated infrarenal aorta and density of the structural protein correlated negatively and significantly with the % increase in the in situ diameter (Pearson's coefficient = −0.64; p = .017) (Figure 6). In the newly formed neointimal region of the CaCl₂‐treated infrarenal aorta of female rats, elastin immunostaining was not detected. Despite the absence of a neointimal region in the CaCl₂‐treated infrarenal aorta of male rats, elastin loss in the medial region and the extent of the reduction was analogous to that observed in female rats and the density of the structural protein negatively correlated and significantly with the % change of the in situ diameter (Pearson's coefficient = −0.67; p = .011) (Figure 6).

FIGURE 6.

Elastin content of the saline‐ and CaCl₂‐treated infrarenal aorta of adult male and female rats. Medial elastin content of the infrarenal aorta of saline‐treated (open circle) and CaCl₂‐treated (solid circle) male and female rats negatively and significantly correlated with the % change of the in situ diameter of the infrarenal aorta. Medial elastin content of the CaCl₂‐treated infrarenal aorta of male and female rats was significantly reduced (Students' unpaired test; * denotes p < .05) compared to the saline‐treated infrarenal aorta.

3.6. Sex‐dependent differences in the cell cycle re‐entry of VSMCs and expression of the cell cycle inhibitor p16^INK4a in the CaCl₂ ‐treated infrarenal aorta

In the medial region of the CaCl₂‐treated infrarenal aorta of female rats, the density of nestin⁽⁺⁾‐VSMCs characterized by nuclear PHH3 staining was significantly increased compared to the saline‐treated infrarenal aorta of female rats supporting a robust proliferative VSMC response (Figure 7A,C,D). Furthermore, the total density of nestin⁽⁺⁾‐ and non‐muscle myosin IIB⁽⁺⁾‐VSMCs identified in the neointimal region were significantly greater compared to medial region of the CaCl₂‐treated infrarenal aorta of female rats (Figure 7B). Consistent with the latter data, the total density of nestin⁽⁺⁾‐VSMCs associated with nuclear PHH3 staining was significantly higher in the neointimal versus the medial region of the CaCl₂‐treated infrarenal aorta of female rats (Figure 7B). Lastly, the nuclear/perinuclear ratio of p16^INK4a staining of non‐muscle myosin IIB⁽⁺⁾‐VSMCs in the medial and neointimal regions of the CaCl₂‐treated infrarenal aorta of female rats were analogous, despite the presence of a robust proliferative VSMC response (Figure 7B).

FIGURE 7.

Cell cycle response of vascular smooth muscle cells (VSMCs) of the saline‐ and CaCl₂‐treated infrarenal aorta of adult male and female rats. (A) Nuclear staining of the G₂‐M cell cycle marker phosphorylated serine ¹⁰ residue of histone 3 (PHH3) of medial non‐muscle myosin IIB (NMIIB)⁽⁺⁾‐VSMCs was significantly increased (Students' unpaired test; * denotes p < .05) in the CaCl₂‐treated infrarenal aorta of female rats versus the saline‐treated infrarenal aorta. The density of medial NMIIB⁽⁺⁾‐VSMCs associated with nuclear PHH3 staining was significantly reduced (Students' unpaired test; * denotes p < .05) in the CaCl₂‐treated infrarenal aorta of male rats versus the saline‐treated infrarenal aorta. (B) In the CaCl₂‐treated infrarenal aorta of female rats, the density of neointimal NMIIB⁽⁺⁾‐VSMCs and nestin⁽⁺⁾‐VSMCs was significantly higher (Students' paired test; * denotes p < .05) versus the respective population identified in the medial region. Furthermore, the density of neointimal NMIIB⁽⁺⁾‐VSMCs associated with nuclear PHH3 staining was significantly higher (Students' paired test; * denotes p < .05) versus medial NMIIB⁽⁺⁾‐VSMCs that re‐entered the cell cycle. Lastly, the nuclear/perinuclear ratio of neointimal NMIIB⁽⁺⁾‐VSMCs expressing the cell cycle inhibitor p16^INK4a was analogous to the population observed in the medial region of the CaCl₂‐treated infrarenal aorta of female rats. (C) In the saline‐treated infrarenal aorta of female rats, staining of the PHH3 (green fluorescence) was identified in the nuclear region of medial nestin⁽⁺⁾‐VSMCs (green fluorescence). In the CaCl₂‐treated infrarenal aorta of female rats, nuclear PHH3 staining was identified in nestin⁽⁺⁾‐VSMCs residing in the neointimal and medial regions. The density of nestin‐immunoreactive cells expressing nuclear PHH3 residing in the advential region was not determined in the present study. The nucleus was identified with 4′,6′‐diamidino‐2‐phenylindole (DAPI) staining (blue fluorescence).

In the CaCl₂‐treated infrarenal aorta of male rats, a redistribution from the perinuclear to the nuclear region was observed as the density of nuclear p16^INK4a staining of medial non‐muscle myosin IIB⁽⁺⁾‐VSMCs was significantly higher compared to perinuclear p16^INK4a staining of non‐muscle myosin IIB⁽⁺⁾‐VSMCs (non‐muscle myosin IIB/perinuclear p16^INK4a, 629 ± 303 vs. non‐muscle myosin IIB/nuclear p16^INK4a, 1100 ± 322 cells/mm²; p = .007, n = 6; Students' paired t‐test) (Figures 4A,B and 8). An analogous redistribution paradigm was identified in the CaCl₂‐treated infrarenal aorta of female rats as the density of nuclear p16^INK4a staining of medial non‐muscle myosin IIB⁽⁺⁾‐VSMCs was greater than the population associated with perinuclear staining but a significant difference was not achieved (non‐muscle myosin IIB/perinuclear p16^INK4a, 863 ± 327 vs. non‐muscle myosin IIB/nuclear p16^INK4a, 1117 ± 332 cells/mm²; p = .19, n = 6; Students' paired t‐test). The greater redistribution from perinuclear to nuclear region led to a significant increase in the nuclear/perinuclear ratio of p16^INK4a staining of non‐muscle myosin IIB⁽⁺⁾‐VSMCs in the medial region of the CaCl₂‐treated versus saline‐treated infrarenal aorta of male and female rats (Figure 8). Moreover, a significant positive correlation was identified between the nuclear/perinuclear ratio of p16^INK4a staining of non‐muscle myosin IIB⁽⁺⁾‐VSMCs and the % change of the in situ diameter of the saline‐ and CaCl₂‐treated infrarenal aorta of male and female rats (Figure 8). Despite the transition to a higher density of nuclear/perinuclear non‐muscle myosin IIB/p16^INK4a‐VSMCs in the CaCl₂‐treated infrarenal aorta of male and female rats, a proliferative response persisted in female rats whereas the density of nestin/PHH3⁽⁺⁾‐medial VSMCs was significantly reduced in male rats (Figure 7).

FIGURE 8.

Expression of the cell cycle inhibitor p16^INK4a in vascular smooth muscle cells (VSMCs) of the saline‐ and CaCl₂‐treated infrarenal aorta of male and female rats. The nuclear/perinuclear ratio of the cell cycle inhibitor p16^INK4a expressed in medial non‐muscle myosin IIB (NMIIB)⁽⁺⁾‐VSMCs positively and significantly correlated with the % change of the in situ diameter of the infrarenal aorta of saline‐ (open circle) and CaCl₂‐treated (solid circle) male and female rats. Consistent with the latter data, the density of the nuclear/perinuclear ratio of the cell cycle inhibitor p16^INK4a expressed in medial NMIIB⁽⁺⁾‐VSMCs was significantly upregulated (Students' unpaired test; * denotes p < .05) in the CaCl₂‐treated infrarenal aorta of male and female rats versus saline‐treated infrarenal aorta of both sexes.

3.7. TIMP‐1/2 and p53 expression in the CaCl_{2 −}treated infrarenal aorta

In the medial region of the CaCl₂‐treated infrarenal aorta of male rats, TIMP‐1 (p = .083 vs. saline‐treated), TIMP‐2 (p = .111 vs. saline‐treated), and p53 (p = .0886 vs. saline‐treated) protein levels were variable within the population and a significant difference was not observed as compared to the saline‐treated infrarenal aorta (Figure 9A,B). A correlation was also not observed between TIMP‐1, TIMP‐2 or p53 protein levels and the % change of the in situ diameter of the infrarenal male rat aorta following CaCl₂ exposure (data not shown).

FIGURE 9.

Cellular remodeling of the CaCl₂‐treated infrarenal aorta of male and female rats. (A & B) TIMP‐1, TIMP‐2 and p53 protein levels in the saline‐ and CaCl₂‐treated infrarenal aorta of male rats were not significantly different and TIMP‐1 and TIMP‐2 protein levels were likewise not significantly different between the saline‐ and CaCl₂‐treated infrarenal aorta of female rats. A significant upregulation (Students' unpaired test; * denotes p < .05) of p53 protein levels was observed in the CaCl₂‐treated infrarenal aorta of female rats versus saline‐treated female rats.

Examining protein expression in the vessel wall of the saline‐treated versus the CaCl₂‐treated infrarenal aorta of female rats is problematic as a neointimal region is prevalent in the latter group. Nonetheless, nestin⁽⁺⁾‐ and NMIIB⁽⁺⁾‐VSMCs were identified in the medial and neointimal regions of the CaCl₂‐treated infrarenal aorta of female rats supporting the premise that at least a comparable VSMC phenotype was prevalent in both groups. In the CaCl₂‐treated infrarenal aorta of female rats, TIMP‐1 (p = .0581 vs. saline‐treated) and TIMP‐2 (p = .0648 vs. saline‐treated) protein levels were variable within the population and a significant difference was not observed compared to the saline‐treated infrarenal aorta (Figure 9A,B). By contrast, p53 protein levels in the medial and neointimal regions of the CaCl₂‐treated infrarenal aorta of female rats were significantly upregulated compared to the saline‐treated infrarenal aorta (Figure 9A,B). Similar to that observed in male rats, a correlation was not observed between TIMP‐1, TIMP‐2 or p53 protein levels and the % change of the in situ diameter of the infrarenal female rat aorta following CaCl₂ exposure (data not shown).

4. DISCUSSION

Clinical studies have reported that the risk of aortic vessel expansion and subsequent aneurysm formation is significantly lower in age‐matched female versus male patients. ² , ³ , ⁴ A reduced risk of aortic aneurysm was also reported in female patients diagnosed with a BAV. ²⁶ , ²⁷ , ²⁸ , ²⁹ The latter paradigm may be attributed in part to the escape of the TIMP‐1 gene from X‐chromosome inactivation in females limiting elastin fragmentation/loss by antagonising the biological action of MMPs. ¹ , ¹⁵ , ¹⁶ Consistent with a role of TIMP‐1, the lower risk of aortic aneurysm formation was abolished in female BAV patients diagnosed with Turner Syndrome characterized by the partial/complete loss of the second X‐chromosome. ²⁶ , ²⁷ The present study has revealed that despite the presence of an underlying genetic abnormality, TIMP‐1 protein levels in the ascending aorta of female BAV patients diagnosed with aortic stenosis were significantly higher versus male BAV patients. However, greater TIMP‐1 protein levels in the female aorta may be offset by the uncontrolled synthesis and release of MMPs following the appearance of senescent VSMCs within the damaged vessel. Experimental and clinical studies have reported that higher expression of cell cycle inhibitors including p21^cip1, p27^kip1, p53, and p16^INK4a were capable of eliciting a senescent VSMC phenotype leading to a higher expression of MMPs translating to greater elastin fragmentation/loss. ¹² , ⁴¹ , ⁴² Indeed, in the aneurysmal aorta of male BAV patients, p21^cip1 and p16^INK4a protein levels were upregulated in medial VSMCs. ⁴² In the present study, the density of medial non‐muscle myosin IIB‐VSMCs characterized by p16^INK4a staining was significantly lower in the ascending aorta of female versus male BAV patients. Thus, at least in female BAV patients, the lower risk of significant elastin fragmentation/loss may be attributed in part to higher TIMP‐1 protein levels and a concomitant smaller density of VSMCs expressing the cell cycle inhibitor p16^INKa translating to a potentially reduced population of senescent VSMCs.

To the best of our knowledge, an experimental animal model recapitulating the sex‐dependent phenotype of greater TIMP‐1 protein levels in the aorta of female versus male patients has not been reported. In the present study, TIMP‐1 protein levels were examined in four distinct regions of the male and female rodent aorta to assess whether the sex‐dependent phenotype observed in humans was recapitulated. Previous work from our lab reported that the intermediate filament protein nestin, a marker of neural crest cells was identified in VSMCs and expression was greater in the ascending aorta/aortic arch versus the thoracic/abdominal aortic regions of the adult male rat. ³⁹ In the ascending aorta/aortic arch regions of the adult female and male rat, non‐muscle myosin IIB protein levels were significantly greater versus the thoracic/abdominal aortic regions suggesting that this protein may likewise represent an additional marker of neural crest cells. By contrast, TIMP‐1 protein levels in each region of the aorta of female rats were significantly higher compared to male rats and complemented by greater TIMP‐2 protein expression. In contrast to the reported escape of the TIMP‐1 gene from X‐chromosome inactivation leading to higher protein levels in females, the TIMP‐2 gene is expressed on chromosome 17 in humans and 10 in rats (National Library of Medicine). Clinical studies have also reported that the cell cycle inhibitor p53 provided females greater protection against the risk of cancer of non‐reproductive tissues. ⁴³ , ⁴⁴ In each region of the female rat aorta, p53 protein levels were significantly higher compared to normal adult male rats and as reported for TIMP‐2, the gene is expressed on chromosome 17 in humans and 10 in rats (National Library of Medicine). Furthermore, p53 directly promoted p21^cip1 protein upregulation and expression of the cell cycle inhibitor in each region of the normal adult female rat was significantly higher compared to normal adult male rats. By contrast, in the thoracic and abdominal aorta of normal adult C57/BL6 female mice, TIMP‐1, p53, and p21^cip1 protein levels were consistently lower compared to normal adult male mice. It remains presently unknown if the latter paradigm persists in other mouse strains. Thus, the sex‐dependent phenotype observed in male and female BAV patients was recapitulated in part in the aorta of normal adult Sprague–Dawley rats and further distinguished by a greater expression of TIMP‐2, p53, and p21^cip1 protein levels.

In contrast to TIMP‐1, TIMP‐2, and p53 genes are not subjected to X‐chromosome inactivation suggesting that higher protein expression of one or both genes in females may be influenced in part by circulating ovarian hormones. A previous study reported that TIMP‐2 protein levels were modestly increased in the left ventricle of the ovariectomized adult female rat. ⁴⁵ Furthermore, in breast and endometrial cancer cell lines, estrogen activation of the estrogen receptor‐α subtype significantly influenced p53 protein levels. ⁴⁶ , ⁴⁷ In the present study, ovariectomy of normal adult female rats was confirmed by uterine atrophy and concomitant increase in body weight. TIMP‐1 and TIMP‐2 protein levels were significantly elevated in several distinct aortic regions whereas p53 expression was unchanged in the aorta of ovariectomized versus normal female rats. Increased protein expression of TIMP isoforms in the aorta of ovariectomized female rats may be attributed in part to the reported increase of the mean arterial pressure secondary to the loss of circulating ovarian sex hormones. ³⁰ Indeed, TIMP‐1 protein levels were significantly upregulated in the hypertensive lung following monocrotaline treatment and the aorta of the adult male rat secondary to L‐NAME induced hypertension. ⁴⁸ , ⁴⁹

Vascular calcification is a complication of chronic kidney disease, diabetes, and hypertension characterized by abnormal accumulation of calcium deposits in the medial region leading to vessel damage. ³¹ , ³² Studies have reported that exposure of the rodent aorta to CaCl₂ recapitulated seminal features of human aortic aneurysm formation including elastin degradation and vessel expansion. ³¹ , ³² In the absence of atherosclerosis, progressive elastin fragmentation/loss triggered neointimal growth of elastic/muscular arteries characterized by the proliferation of medial VSMCs and migration to the damaged vascular region. ²⁰ , ²¹ In the latter scenario, neointimal growth represents a physiological response attempting to limit further expansion by increasing vessel thickness at the site of reduced elastin content. ¹⁸ , ¹⁹ , ²⁰ , ²¹ , ²² Direct support for the latter premise was reported in a saccular aortic aneurysm rat model as stent implantation at the site of the vessel bulge triggered a robust neointimal response that completely healed the compromised region. ¹⁹ Moreover, clinical studies have highlighted a sex‐dependent response as neointimal growth and the rate of in‐stent restenosis of diseased coronary arteries were greater in female versus male patients. ²³ , ²⁴ , ²⁵ Based on the latter observations, a series of experiments tested the hypothesis that CaCl₂‐mediated elastin loss in the medial region of the infrarenal aorta of male and female adult rats may trigger a greater or selective neointimal response in female rats. A measure of the in situ diameter of the CaCl₂‐treated infrarenal aorta of male and female rats revealed significant vessel expansion and the % change was significantly smaller in female versus male rats. The latter paradigm was attributed at least in part to the selective appearance of a neointimal region of the CaCl₂‐treated infrarenal aorta of female rats and the % change of the in situ diameter positively correlated with the magnitude of the neointimal response. Moreover, despite the apparent sex‐dependent neointimal response to CaCl₂ treatment, the magnitude of elastin loss in the medial region of the infrarenal aorta of male and female rats was comparable. Thus, progressive elastin fragmentation/loss in the medial region of the infrarenal aorta of male rat was apparently insufficient to trigger a neointimal response following exposure to CaCl₂.

A requisite event of neointimal formation is the migration of proliferating medial VSMCs to the elastin‐depleted region of the damaged vessel. ¹⁸ , ¹⁹ , ²⁰ , ²¹ , ²² The neointimal response documented in the CaCl₂‐treated infrarenal aorta of female rats was associated with a significant increase in the density of medial VSMCs that re‐entered the G₂‐M phase of the cell cycle characterized by the nuclear appearance of the phosphorylated serine ¹⁰ residue of histone‐3 (PHH3). ³⁹ Moreover, the density of VSMCs expressing nuclear PHH3 identified in the neointimal region was significantly higher compared to the medial region of the CaCl₂‐treated infrarenal aorta of female rats; highlighting a robust cellular proliferative response. The latter paradigm persisted despite the inherently higher expression of the cell cycle inhibitor p53 in the normal female rat aorta and further upregulation of p53 protein levels in the CaCl₂‐treated versus the saline‐treated infrarenal aorta of female rats. By contrast, the absence of a neointimal response of the CaCl₂‐treated infrarenal aorta of male rats was associated with a significant reduction of the density of VSMCs that re‐entered the G₂‐M phase of the cell cycle. The latter response was not attributed to an upregulation of p53 protein levels in the CaCl₂‐treated infrarenal aorta of male rats. In this regard, alternative cell cycle inhibitors may have played a seminal role in the antiproliferative response observed in the CaCl₂‐treated male infrarenal aorta. Indeed, protein levels of the cell cycle inhibitor p16^INK4a were upregulated in VSMCs of the aneurysmal aorta of BAV patients translating to a senescent phenotype. ⁴² In the infrarenal aorta of normal adult male and female rats, p16^INK4a immunostaining was unexpectedly detected in the perinuclear or nuclear region of medial VSMCs and a significantly higher density was observed in the perinuclear region within both sexes. Following CaCl₂‐mediated vessel expansion and concomitant elastin fragmentation/loss of the infrarenal aorta of both male and female rats, p16^INK4a preferentially redistributed to the nuclear region of medial VSMCs. The latter phenotype led to an increase in the nuclear/perinuclear ratio of the density of p16^INK4a‐immunoreactive medial VSMCs that positively and significantly correlated with the % change of the in situ diameter of the infrarenal aorta of both sexes. Consistent in part with our data, previous studies have reported p16^INK4a shuttling between subcellular compartments in various disease states including cancer. ⁵⁰ , ⁵¹ In female patients diagnosed with squamous cell carcinoma of the cervix, preferential shuttling of nuclear p16^INK4a to the cytoplasm was associated with a significantly predicted poorer outcome. ⁵¹ The latter conclusion is consistent with nuclear p16^INK4a promoting G₁ phase cell cycle arrest and inducing cellular senescence. ⁴¹ Thus, despite a disparate proliferative response of VSMCs following CaCl₂‐treatment of the male and female infrarenal aorta, redistribution of the cell cycle inhibitor p16^INK4a from the perinuclear to nuclear region of VSMCs was comparable between sexes. In this regard and in contrast to that reported in the literature, upregulation of p53 protein levels and greater nuclear localization of p16^INK4a in medial VSMCs were apparently insufficient to suppress the neointimal response of the CaCl₂‐treated infrarenal aorta of adult female rats. Lastly, a paucity of non‐muscle myosin IIB⁽⁺⁾‐VSMCs identified in the ascending aorta of male and female BAV patients were also associated with perinuclear p16^INK4a staining whereas the cell cycle inhibitor was preferentially detected in the nuclear region. Unfortunately, the absence of aortic tissue from normal patients precluded determining whether BAV patients were associated with a significant redistribution of perinuclear p16^INK4a to the nuclear region of medial VSMCs.

The increased synthesis/release of MMPs by senescent VSMCs and/or a concomitant reduced expression of TIMP isoforms translates to elastin fragmentation/loss leading to vessel expansion and potential aneurysm formation. ¹³ , ¹⁴ , ¹⁷ , ³¹ However, compromised elastin content in the medial region also represents a seminal event driving neointimal formation providing the requisite environment of proliferating VSMCs to migrate to the damaged region thereby limiting vessel expansion. ¹⁸ , ¹⁹ , ²⁰ , ²¹ , ²² In the vessel wall of the CaCl₂‐treated infrarenal aorta of female rats, TIMP‐1 and TIMP‐2 protein levels were predominantly elevated but did not reach a statistical significance compared to the saline‐treated infrarenal aorta of female rats. An analogous paradigm was observed in the CaCl₂‐treated infrarenal aorta of male rats as TIMP‐1 and TIMP‐2 protein levels were not significantly elevated.

In conclusion, the present study has revealed that despite the presence of an underlying genetic abnormality, female BAV patients were associated with higher TIMP‐1 protein levels and a significantly smaller density of p16^INK4a(+)‐VSMCs in the ascending aorta compared to male BAV patients. Greater sex‐dependent aortic TIMP‐1 protein levels observed in female BAV patients was selectively recapitulated in the female rat aorta complemented by higher TIMP‐2 and p53 protein expression independent of ovarian hormones. Following CaCl₂‐treatment of the male and female rat infrarenal aorta, a comparable loss of medial elastin content in both sexes led to preferential neointimal growth in female rats that may have contributed in part to the significantly smaller in situ expansion of the vessel diameter. Neointimal growth of the CaCl₂‐treated infrarenal aorta of female rats occurred despite protein upregulation of cell cycle inhibitor p53 and predominant redistribution of p16^INK4a from the perinuclear to nuclear region of medial and neointimal VSMCs. By contrast, absence of a neointimal response secondary to the reduced density of proliferating VSMCs of the CaCl₂‐treated male rat infrarenal aorta may be attributed in part to the male sex hormone testosterone. Experimental studies have revealed that testosterone suppressed the neointimal response of the damaged vessel in males in part via the upregulation of cell cycle inhibitors. ⁵² , ⁵³ , ⁵⁴ Thus, inherently greater TIMP‐1 expression in the aorta of female BAV patients may prevent excessive elastin fragmentation/loss. Moreover, it is tempting to speculate that higher TIMP‐1 and TIMP‐2 protein levels in the infrarenal aorta of adult female rats may have delayed the rapid loss of the structural protein elastin following CaCl₂‐treatment thereby providing an opportunity of VSMCs to re‐enter the cell cycle and migrate to the damaged region translating to the preferential formation of a neointimal region.

Supporting information

FIGURE S1. Cellular remodeling of the ascending aorta of male and female bicuspid aortic valve (BAV) patients. Nestin, MMP‐2, p27, and SM22α protein levels in the ascending aorta of female BAV (FBAV; n = 8–11) and male BAV (MBAV; n = 8–12) patients were not significantly different. Protein expression was normalized to GAPDH protein levels.

FIGURE S2. Sex‐dependent cellular differences in the thoracic aorta of normal mice. TIMP‐1 and p53 and p21^cip1 protein levels in the thoracic aorta of male C57/Bl6 mice (n = 3) were higher compared to female C57/Bl6 mice (n = 4) and a significance difference (Students' unpaired test; * denotes p < .05) was observed for TIMP‐1. Protein expression was normalized to GAPDH protein levels.

FIGURE S3. Morphological remodeling of the CaCl₂‐treated infrarenal aorta of adult male and female rats. The infrarenal aorta of adult male and female Sprague–Dawley rats was treated with saline or 0.5 M CaCl₂ for 15 min. Four weeks later, the medial region of the CaCl₂‐treated infrarenal aorta of male and female rats was distinguished from saline‐treated infrarenal aorta by the selective appearance of calcium deposits (brownish regions) following Von Kossa staining.

TABLE S1. Patient demographics and clinical characteristics.

Al‐Katat A, Boudreau L, Gagnon E, Assous I, Villeneuve L, Leblanc CA, et al. Greater TIMP‐1 protein levels and neointimal formation represent sex‐dependent cellular events limiting aortic vessel expansion in female rats. IUBMB Life. 2024;76(12):1356–1376. 10.1002/iub.2916

DATA AVAILABILITY STATEMENT

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.