Polycystic Ovary Syndrome Fuels Cardiovascular Inflammation and Aggravates Ischemic Cardiac Injury

Abstract

BACKGROUND:

Reducing cardiovascular disease burden among women remains challenging. Epidemiologic studies have indicated that polycystic ovary syndrome (PCOS), the most common endocrine disease in women of reproductive age, is associated with an increased prevalence and extent of coronary artery disease. However, the mechanism through which PCOS affects cardiac health in women remains unclear.

METHODS:

Prenatal anti-Müllerian hormone treatment or peripubertal letrozole infusion was used to establish mouse models of PCOS. RNA sequencing was performed to determine global transcriptomic changes in the hearts of PCOS mice. Flow cytometry and immunofluorescence staining were performed to detect myocardial macrophage accumulation in multiple PCOS models. Parabiosis models, cell-tracking experiments, and in vivo gene silencing approaches were used to explore the mechanisms underlying increased macrophage infiltration in PCOS mouse hearts. Permanent coronary ligation was performed to establish myocardial infarction (MI). Histologic analysis and small-animal imaging modalities (eg, magnetic resonance imaging and echocardiography) were performed to evaluate the effects of PCOS on injury after MI. Women with PCOS and control participants (n=200) were recruited to confirm findings observed in animal models.

RESULTS:

Transcriptomic profiling and immunostaining revealed that hearts from PCOS mice were characterized by increased macrophage accumulation. Parabiosis studies revealed that monocyte-derived macrophages were significantly increased in the hearts of PCOS mice because of enhanced circulating Ly6C⁺ monocyte supply. Compared with control mice, PCOS mice showed a significant increase in splenic Ly6C⁺ monocyte output, associated with elevated hematopoietic progenitors in the spleen and sympathetic tone. Plasma norepinephrine (a sympathetic neurotransmitter) levels and spleen size were consistently increased in women with PCOS when compared with those in control participants, and norepinephrine levels were significantly correlated with circulating CD14⁺⁺CD16⁻ monocyte counts. Compared with animals without PCOS, PCOS animals showed significantly exacerbated atherosclerotic plaque development and post-MI cardiac remodeling. Conditional Vcam1 silencing in PCOS mice significantly suppressed cardiac inflammation and improved cardiac injury after MI.

CONCLUSIONS:

Our data documented previously unrecognized mechanisms through which PCOS could affect cardiovascular health in women. PCOS may promote myocardial macrophage accumulation and post-MI cardiac remodeling because of augmented splenic myelopoiesis.

Clinical Perspective.

What Is New?

• Polycystic ovary syndrome (PCOS) mice hearts were characterized by increased macrophage accumulation.

• Splenic monocytopoiesis was augmented by sympathetic activation in PCOS mice, leading to enhanced monocyte supply and macrophage differentiation in the myocardium.

• PCOS promoted remodeling after myocardial infarction and atherosclerotic plaque development, features alleviated by suppressing splenic monocytopoiesis.

What Are the Clinical Implications?

• The current findings suggest that PCOS fuels cardiac inflammation through splenic monocytopoiesis, independent of metabolic risk factors, providing novel mechanistic explanations for increased cardiovascular risk in women with PCOS.

• Suppression of splenic monocytopoiesis may afford a potential therapeutic option for cardiovascular protection in women with PCOS.

The epidemiology and clinical outcomes of coronary artery disease (CAD) vary between male and female patients.¹ Worse outcomes have been described after interventions for CAD in women, and tailored management strategies are lacking,¹ partly because of an incomplete understanding of underlying mechanisms. In addition to traditional CAD risk factors such as diabetes and dyslipidemia, certain risk factors are specific to women, including adverse pregnancy outcomes (eg, preterm delivery) and female reproductive endocrine disorders (eg, premature menopause and polycystic ovary syndrome [PCOS]). The well-studied effects of adverse pregnancy outcomes on cardiovascular health led to the proposal of adjunctive cardioprotective strategies for postpartum women.² The association between female reproductive endocrine disorders and CAD risk has been established, while more investigations are required.³

PCOS is the most common endocrine disorder in women of reproductive age, clinically defined by the presence of at least 2 of the following manifestations: polycystic ovaries, chronic anovulation, and hyperandrogenemia.⁴ PCOS affects ovarian function and can cause infertility,⁵ subsequently emerging as a CAD risk factor. Epidemiologic evidence indicates that PCOS is associated with an increased prevalence of CAD in women.^6,7 Moreover, PCOS is associated with more extensive coronary lesions in women with CAD.⁸ Women with PCOS have an increased prevalence of metabolic complications, including insulin resistance, obesity, and dyslipidemia,⁹ leading to the hypothesis that the increased CAD risk in women with PCOS may be attributable to these traditional risk factors. PCOS remains associated with cardiac impairments¹⁰ and increased CAD risk after adjusting for body mass index (BMI) or glucose metabolism measures,⁷ suggesting a potential involvement of nonmetabolic factors in the increased CAD risk in PCOS. However, the nonmetabolic mechanisms underlying the detrimental effects of PCOS on cardiac health in women remain underexplored.

We aimed to explore adverse pathologic changes in cardiac tissues of a PCOS animal model without detectable metabolic abnormalities to determine the contribution of these alterations to ischemic heart injury.

METHODS

Data used in this study are available upon reasonable request. RNA sequencing (RNA-seq) data are available in the database repository Gene Expression Omnibus (accession number GSE238075).

Animal Studies

All experiment protocols complied with the National Institutes of Health Guidelines on the Care and Use of Laboratory Animals and were approved by the Institutional Animal Ethics Committee of Fudan University. C57BL/6 mice, CD45.1 allele-expressing C57BL/6 congenic mice, apoE^−/− mice, transgenic mice expressing ZsGreen, macrophage-specific Vcam1 knockout mice (Vcam1^cKO), and wild-type and apoE^−/− Sprague-Dawley rats¹¹ were housed under a 12-hour light/12-hour dark cycle. Two PCOS models were induced by different protocols: prenatal anti-Müllerian hormone (PAMH) treatment by peritoneal injection from gestational day 16.5 to 18.5 in pregnant mice once daily (0.12 mg/kg per day)¹²; or minipump-controlled continuous letrozole release (90 μg/day) for 8 weeks starting from 4 weeks of age. Permanent left coronary artery ligation was performed to establish myocardial infarction (MI) models. Atherosclerotic plaques were induced by partial ligation of renal and carotid arteries in apoE^−/− mice.¹³ Intraperitoneal injections of Clophosome-A clodronate liposome were administered twice weekly (0.1 mL/20 g; FormuMax) to deplete macrophages. The parabiosis model was established by joining the olecranon and knee joints and skins of paired mice. 6-Hydroxydopamine (6-OHDA; 250 mg/kg) was administered intraperitoneally once weekly for chemical ablation of sympathetic nerve fibers. Detailed Methods are provided in the Supplemental Material.

Human Studies

The study was approved by the ethics committee of the Obstetrics and Gynecology Hospital, Fudan University, and informed consent was obtained for all participants. We recruited 100 women with PCOS and 100 participants without PCOS to investigate the association of PCOS with monocyte number and spleen size. PCOS was diagnosed according to the Rotterdam criteria.¹⁴ Participants ≥18 years of age were included. Participants who were pregnant or had hematologic disease, autoimmune disease, infectious disease, portal hypertension or chronic liver disease, splenectomy, Cushing syndrome, nonclassic congenital adrenal hyperplasia, thyroid disease, hyperprolactinemia, or known cancer were excluded. Fasting blood was sampled to examine sex hormones, lipid and glucose metabolism measures, norepinephrine levels, and circulating monocyte numbers. Spleen size was measured by ultrasonic analyses. Detailed information is provided in the Supplemental Material.

Statistical Analysis

Continuous variables were tested for normality using the Q-Q plot and Shapiro-Wilk test, and homogeneity of variance was tested using the Levene test. Normally distributed variables are expressed as mean±SEM. Categorical variables are presented as frequencies (percentages). Differences in continuous data were compared using the Student t test or the Mann-Whitney U test, and categorical variables were compared using the χ² or Fisher exact test, when appropriate. For comparison among multiple groups, 1-way ANOVA with Tukey post hoc test or the Welch ANOVA test with the Games-Howell post hoc test was used, as appropriate. For comparisons involving 2 independent variables, 2-way ANOVA or 2-way repeated-measures ANOVA was used, as appropriate. The association between PCOS status and spleen size or monocyte count in humans was also assessed using multivariable linear regression. The adjusted variables included metabolic measures that significantly differed between women with PCOS and control participants, including BMI, triglycerides, high-density lipoprotein cholesterol level, and homeostatic model assessment of insulin resistance. Correlation analyses were performed using Spearman correlation coefficients. All statistical analyses were performed using R version 3.6.0 (R Foundation for Statistical Computing). A P value (2-sided) <0.05 was deemed significant.

The expanded Methods in the Supplemental Material include detailed descriptions of the following: generation of macrophage-specific Vcam1^cKO mice; PCOS, MI, atherosclerotic plaque, and parabiosis models; splenectomy; in vivo macrophage depletion; adoptive cell transfer; in vivo Vcam1 silencing¹⁵; chemical ablation of sympathetic nerve fibers; insulin and glucose tolerance testing; lipid levels and hormone assays; assessment of estrous cycle; body mass analysis by dual-energy X-ray absorptiometry; echocardiography and blood pressure assessment; cardiac magnetic resonance imaging; immunofluorescence and histologic staining¹⁶; TUNEL (terminal deoxynucleotidyl transferase dUTP nick-end labeling) assay; real-time quantitative polymerase chain reaction; flow cytometry; RNA-seq and bioinformatics analysis; culture of splenic macrophages¹⁷; luciferase assay; and human studies.

RESULTS

Characterization of Reproductive and Metabolic Phenotypes in a PCOS Mouse Model

To examine PCOS-related effects on cardiac health, we used PAMH-induced PCOS mice.¹² Compared with female control mice, PAMH mice exhibited significantly elevated serum testosterone levels (Figure 1A and 1B). PAMH mice displayed polycystic ovaries with a significant reduction of corpora lutea number (Figure 1C and 1D), indicating anovulation. Moreover, PAMH mice had notable estrous acyclicity when compared with female control mice (Figure 1E).

Figure 1.

Characterization of reproductive and metabolic phenotypes in a polycystic ovary syndrome mouse model. A, Schematic illustration showing the induction of polycystic ovary syndrome (PCOS) by prenatal anti-Müllerian hormone (PAMH) treatment. B, Serum testosterone levels in indicated groups (8 mice/group). C, Histologic analysis of ovaries. D, Number of corpora lutea in indicated groups (8 mice/group). E, PAMH mice exhibited acyclicity compared with control female mice. F, Body weight in indicated groups (8 mice/group). G, Body fat composition measured by dual-energy X-ray absorptiometry in indicated groups (8 mice/group). H, Representative images of hematoxylin & eosin staining of parametrial fat in indicated groups. I, Adipocyte cell sizes in indicated groups (100 cells from 5 mice/group). J, Parametrial fat weight in indicated groups (8 mice/group). K and L, Glucose tolerance test and insulin tolerance test results in indicated groups (8 mice/group). M, Serum lipid levels in indicated groups (8 mice/group). Data are expressed as mean±SEM. Statistical analysis was performed by Student t test (for D, F, G, J, K, L, and M), Mann-Whitney U test (for B), or a linear mixed model (for I). AUC indicates area under the curve; E, estrous stage; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; M/D, metestrus/diestrus stage; P, proestrous stage; TC, total cholesterol; and TG, triglycerides.

We then evaluated metabolic measures and found no significant differences in body weight, body fat composition, or adipocyte cell size between PAMH and control mice (Figure 1F through 1J). Glucose and lipid homeostasis were not overtly impaired in PAMH mice (Figure 1K through 1M). Liver weight and intrahepatic lipid content did not significantly differ between the 2 groups (Figure S1A through S1D). Daily food intake and mean blood pressure were similar between PAMH and control mice (Figure S1E and S1F). The PAMH-induced PCOS model mimicked PCOS features without detectable confounding metabolic disruptions 12 weeks postnatally.

Hearts of Female PCOS Mice Are Characterized by Increased Macrophage Accumulation

To explore pathologic changes in cardiac tissues of PAMH-induced PCOS mice, we performed unbiased genome-wide RNA-seq analysis. There were 85 differentially expressed genes in cardiac tissues of PCOS and control mice (57 upregulated and 28 downregulated; Figure 2A). The top 10 upregulated genes were validated by quantitative polymerase chain reaction analysis (Figure 2B and 2C). Gene set enrichment analysis indicated that the most upregulated gene set was inflammatory response in the PCOS group (Figure 2D), with the most enriched terms relating to immune response (Figure 2E and 2F). PCOS mice had downregulated androgen response, estrogen response, and bile acid metabolism (Figure 2G), previously documented in noncardiac samples from women with PCOS (eg, skeletal muscle tissues).^18,19 These results indicate that cardiac tissues from PCOS mice recapitulate transcriptomic alterations observed in women with PCOS.

Figure 2.

Hearts of polycystic ovary syndrome mice are characterized by increased macrophage accumulation. A, Volcano plot of the RNA sequencing results obtained from hearts of polycystic ovary syndrome (PCOS) mice and control mice (5 or 6 mice/group). The red dots represent significantly upregulated genes (nominal P<0.05; fold change ≥2) and the blue dots represent significantly downregulated genes (nominal P<0.05; fold change ≤0.5). To avoid skewing of the plot, genes with >16-fold downregulation were omitted from the plot. B, Heatmap plot showing the expression levels of the top 10 upregulated genes in PCOS hearts. C, Quantitative polymerase chain reaction analysis results validated the upregulated genes in RNA sequencing analysis (5 mice/group). D through G, Results of the gene set enrichment analysis, Kyoto Encyclopedia of Genes and Genomes (KEGG), and Gene Ontology (GO) enrichment analysis of the RNA sequencing data. H, t-Distributed stochastic neighbor embedding plots showing that the top upregulated genes in PCOS mouse heart were enriched in the cardiac macrophage cluster. I, Flow cytometry analysis indicated that cardiac macrophages were significantly increased in hearts of PCOS mice (8 mice/group). Data are expressed as mean±SEM. Statistical analysis was performed by Student t test (for I); Student t test or Mann-Whitney U test was used for different genes in C.

The top enriched terms included antigen processing and presentation, monocyte chemotaxis, and CCR chemokine-receptor binding (Figure 2E and 2F), suggesting the potential active involvement of monocytes or macrophages, or both, in the PCOS mouse heart. To confirm this hypothesis, we explored whether genes upregulated in PCOS hearts were markers for specific immune cell types. In 2 publicly available single-cell sequencing data sets of heart cell suspensions,^20,21 all top-10 upregulated genes in the PCOS mouse heart were predominantly expressed in the cardiac macrophage cluster (Figure 2H and Figure S2). Flow cytometry revealed a significant increase in cardiac macrophages in PCOS mouse hearts (Figure 2I); this finding was further supported by immunofluorescence staining (Figure S3). To confirm that these cardiac phenotypes were specifically related to the PCOS status in the PAMH-treated model, we examined phenotypes in PAMH-treated male mice. In contrast to their female counterparts, the hearts of male PAMH mice showed minimal enrichment of innate immune-related gene sets (Figure S4A through S4F). Cardiac macrophage numbers did not significantly differ between male PAMH and control mice (Figure S4G through S4H).

To confirm increased cardiac macrophage accumulation in PCOS, we established another PCOS model using peripubertal letrozole treatment.²² After an 8-week infusion, letrozole-treated mice manifested typical PCOS phenotypes, including polycystic ovaries and elevated circulating testosterone levels (Figure S5A through S5C). Contrary to the PAMH-induced PCOS model, the letrozole-induced model exhibited metabolic abnormalities compared with control mice (Figure S5D through S5G). Despite these differences, most upregulated genes in PAMH mouse hearts were confirmed in the letrozole-induced PCOS model (Figure S5H). The letrozole-induced PCOS model also showed an increased number of cardiac macrophages (Figure S5I through S5J). The number of cardiac macrophages was not significantly increased in letrozole-treated male mice (Figure S6). Overall, the increased macrophage accumulation and inflammatory burden may represent key pathologic characteristics of PCOS mouse hearts.

Given the prevalence of metabolic abnormalities (eg, glucose intolerance and obesity) in women with PCOS,²³ we further investigated whether metabolic abnormalities could induce cardiac molecular changes similar to those mediated by PCOS. We established a high-fat diet–induced obese mouse model. Compared with normal feeding, 8-week high-fat diet administration increased body weight and impaired glucose tolerance (Figure S7A and S7B), whereas serum testosterone levels were not significantly altered (Figure S7C). Different from the pathway changes in hearts from the PAMH-induced PCOS model, fatty acid metabolism, adipogenesis, and interferon-α response pathways were highly activated in high-fat diet–treated mouse hearts (Figure S7D and S7E). Contrary to the high enrichment in PCOS mouse hearts (Figure 2E and 2F), Gene Ontology terms related to monocyte or macrophage biology were not enriched in high-fat diet–treated mouse hearts (Figure S7F). These results indicate that PCOS-induced pathologic changes in the heart are distinct from those induced by metabolic abnormalities.

Circulating Monocytes Contribute to Enhanced Cardiac Macrophage Accumulation in PCOS Mice

Cardiac macrophages may increase because of augmented resident cell proliferation or increased monocyte infiltration and differentiation.²⁴ We first measured in vivo macrophage proliferation in the heart and detected no significant differences between PAMH-induced PCOS and control mice (Figure 3A). Thus, we established a parabiosis model combining CD45.1 and CD45.2 mice to assess the contribution of circulating monocytes to the increased macrophages in PCOS hearts (Figure 3B). Chimerism of circulating Ly6C⁺ monocytes was comparable between the 2 parabiosis groups (Figure 3C). However, the percentage of CD45.1 monocyte-derived macrophages was significantly higher in paired CD45.2 mouse hearts of the PCOS group than in the control group (Figure 3D), suggesting that circulating monocytes contribute to the increased cardiac macrophages in PCOS mice. We further established a parabiosis model by combining ZsGreen⁺ mice and ZsGreen⁻ mice to confirm these results. Paired ZsGreen⁻ mouse hearts in the PCOS group consistently exhibited significantly higher ZsGreen⁺ monocyte-derived macrophages than the control group (Figure 3E and 3F). Circulating Ly6C⁺ monocytes were significantly increased in PAMH-induced PCOS mice compared with control mice (Figure 3G). On the contrary, circulating Ly6C⁺ monocytes did not differ significantly between PAMH-treated male mice and control male mice (Figure S8). Taken together, these data suggest that increased cardiac macrophages in PCOS mice are derived from circulating monocytes.

Figure 3.

Circulating monocytes contribute to increased macrophage accumulation in hearts of polycystic ovary syndrome mice. A, 5-Bromodeoxyuridine (BrdU)⁺ incorporation in heart macrophages in indicated groups measured at 2 hours after BrdU injection (8 mice/group). B, Illustration showing the parabiosis models joining CD45.1 mice and CD45.2 mice to assess the contribution of circulating monocytes to increased cardiac macrophages in polycystic ovary syndrome (PCOS) mice. C, Flow cytometry analysis indicated that the chimerism of Ly6C⁺ monocytes in peripheral blood was comparable between PCOS pairs and control pairs (8 mice/group). D, The chimerism of donor-derived CD45.1⁺ cardiac macrophages was significantly increased in the PCOS parabiotic pairs compared with control pairs (8 mice/group). E and F, The parabiosis models joining ZsGreen⁺ mice and ZsGreen⁻ mice were used to confirm the contribution of circulating monocytes to increased cardiac macrophages in PCOS mice. Immunofluorescence staining results indicated that ZsGreen⁺ monocyte-differentiated cardiac macrophages were significantly increased in PCOS mice versus control mice (5 mice/group). G, Dot plots and quantification of Ly6C⁺ monocytes in the blood of PCOS and control mice (8 mice/group). Data are expressed as mean±SEM. Statistical analysis was performed by the Student t test (for A, D, F, and G) or Mann-Whitney U test (for C).

Splenic Monocytopoiesis Is Enhanced in PCOS Mice

We next explored the source of increased monocyte supply in PAMH-induced PCOS mice. Assessing hematopoiesis in the bone marrow, the number of bone marrow Ly6C⁺ monocytes did not significantly differ between PCOS and control mice (Figure S9A). We analyzed lineage⁻c-kit⁺Sca-1⁺ (LSK) stem cells and progenitors, including common myeloid progenitors and granulocyte-macrophage progenitors (GMPs), in the bone marrow, which are progenitors responsible for generating monocytes.²⁵ Compared with control mice, the numbers of LSK cells, GMPs, and common myeloid progenitors were decreased in the bone marrow of PCOS mice (Figure S9B and S9C). PCOS mice exhibited a reduction in progenitor retention factors in the bone marrow, including C-X-C motif chemokine ligand 12 (Cxcl12) and stem cell factor (Scf; Figure S9D through S9F).

We next examined splenic hematopoiesis in PAMH-induced PCOS mice. Compared with control mice, PCOS mice showed significantly elevated splenic Ly6C⁺ monocyte abundance (Figure S9G). LSK cells and GMPs were significantly increased in the spleen of PCOS mice (Figure 4A and 4B), accompanied by increased spleen weight (Figure 4C). These results indicate increased splenic monocytopoiesis in PCOS mice. Together with the findings in the bone marrow, we speculate that PCOS may trigger hematopoietic progenitor cell egress from the bone marrow and engraftment in the spleen. To test this hypothesis, we used parabiosis models combining CD45.1 and CD45.2 mice. The PCOS group showed a significantly higher GMP cell number in peripheral blood than the control group (Figure 4D), indicating an increased egress of GMPs into the peripheral blood. Moreover, CD45.1⁺ GMP cells in the spleen of paired CD45.2 mice (ie, GMP cells originating from outside the spleen) were significantly increased in the PCOS parabiotic pairs compared with the control pairs (Figure 4E and 4F), indicating increased progenitor engraftment in the spleen of PCOS mice. Increased sympathetic tone reportedly contributes to hematopoietic progenitor cell mobilization from the bone marrow.²⁶ Women with PCOS show increased sympathetic nerve activity in different tissues.²⁷ We observed that tyrosine hydroxylase expression (an index of sympathetic tone) was increased in the bone marrow of PCOS mice compared with control mice (Figure 4G). After 6-OHDA–mediated chemical sympathetic ablation, PCOS mice recovered GMPs in the bone marrow compartment, resulting in reduced levels in the blood and spleen; these alterations were not observed in control mice (Figure 4H through 4K). Moreover, 6-OHDA treatment significantly decreased the number of splenic and circulating monocytes and cardiac macrophages in PCOS mice but not in control mice (Figure 4L through 4N). This suggests that increased sympathetic tone contributes to stem or progenitor cell mobilization from the bone marrow and engraftment in the spleen of PCOS mice.

Figure 4.

Hematopoietic stem and progenitor cells are increased in the spleen of polycystic ovary syndrome mice. A and B, Dot plots and quantification of hematopoietic stem and progenitor cells (lineage⁻c-kit+Sca-1+ cells [LSK] and granulocyte macrophage progenitors [GMPs]) in the spleen of polycystic ovary syndrome (PCOS) and control mice (8 mice/group). C, Spleen weight in indicated groups (8 mice/group). D, Quantification of GMPs in the blood of PCOS and control mice (8 mice/group). E and F, Dot plots and quantification of CD45.1⁺ partner-derived splenic GMP cells in PCOS pairs and control pairs (8 mice/group). G, Representative images and quantification of tyrosine hydroxylase staining in the bone marrow of PCOS mice and control mice (5 mice/group). H, Dot plots showing GMP (in the bone marrow, blood, and spleen), Ly6C⁺ monocytes (in the spleen and blood), and cardiac macrophages in indicated groups. I through K, Quantification of GMPs in the bone marrow, blood, and spleen in indicated groups (8 mice/group). L through N, Quantification of Ly6C⁺ monocytes in the spleen and blood and cardiac macrophages in indicated groups (8 mice/group). Data are expressed as mean±SEM. Statistical analysis was performed by Student t test (for B, C, F, and G) or Mann-Whitney U test (for D) or 2-way ANOVA with Tukey multiple comparisons test (for I, J, K, L, M, and N). Log transformation was done for blood GMP numbers to pass the heteroscedasticity test. 6-OHDA indicates 6-hydroxydopamine hydrobromide.

To further investigate whether the progenitor cell–retaining capacity of the spleen was altered in PAMH-induced PCOS mice, we performed adoptive transfer of ZsGreen⁺ progenitor cells. Six hours after ZsGreen⁺ LSK or GMP injection, splenic engraftment of LSK cells and GMPs was significantly higher in PCOS mice than in control mice (Figure S10A and S10B). The number of splenic macrophages was significantly higher in PCOS mice than in control mice (Figure S10C). Splenic macrophages reportedly play an important role in obtaining progenitor cells through direct adhesion.²⁸ Thus, we examined whether splenic macrophages participate in the augmented retention of progenitor cells in PCOS spleen. After macrophage depletion (Figure S10D), progenitor cell numbers were similar in the spleen of PCOS and control mice (Figure S10E).

We found that Vcam1 (vascular cell adhesion molecule 1), a progenitor cell retention factor, was significantly increased in splenic macrophages from PAMH-induced PCOS mice compared with control mice (Figure S10F). 6-OHDA–mediated chemical sympathetic ablation (which depletes norepinephrine) significantly decreased Vcam1 expression in splenic macrophages (Figure S10G). Compared with control mice, PCOS mice showed significantly increased splenic norepinephrine levels (Figure S10H), suggesting a potential role of norepinephrine in Vcam1 upregulation. In cultured splenic macrophages, norepinephrine dose-dependently induced Vcam1 expression (Figure S10I). In RAW 264.7 macrophages, norepinephrine treatment significantly induced nuclear factor–κB activation (Figure S10J). Cotreatment with JSH23 (a nuclear factor–κB inhibitor) significantly suppressed the norepinephrine-induced increase in Vcam1 expression (Figure S10K). Thus, increased sympathetic tone may contribute to the upregulated macrophage Vcam1 expression. To elucidate the role of Vcam1 expression in splenic monocytopoiesis in PCOS, Vcam1 was silenced by siRNA delivery using a nanoparticle with high enrichment in the spleen¹⁵ (Figure S10L and S11). In vivo Vcam1 silencing significantly decreased splenic retention of adoptively transferred ZsGreen⁺ GMPs in PCOS mice (Figure S10M and S10N). Vcam1 silencing decreased the number of splenic GMPs and monocytes, circulating monocytes, and cardiac macrophages in PCOS mice, which was not observed in control mice (Figure S10O through S10R). These results collectively indicate that splenic monocytopoiesis could be enhanced in PCOS mice by means of increased Vcam1 expression, leading to increased circulating monocytes and accumulation of myocardial macrophages.

Plasma Norepinephrine Levels Are Elevated and Correlate With Circulating Inflammatory Monocyte Counts in Women With PCOS

To explore the clinical relevance of the findings in mice, control participants and women with PCOS (n=200) were recruited on the basis of the Rotterdam criteria. Table S1 summarizes the baseline characteristics of all participants. Women with PCOS had higher BMI (25.3 kg/m² [interquartile range, 21.3–28.9 kg/m²] versus 23.7 kg/m² [interquartile range, 21.2–26.2 kg/m²]; P=0.019) than control participants, along with significantly altered sex hormonal status (Table S1). Consistent with findings in mice, the number of total circulating monocytes, inflammatory CD14⁺⁺CD16⁻ monocytes, and spleen size were significantly increased in women with PCOS compared with those in control participants (Figure S12A through S12D). After adjusting for significantly different metabolic measures, multivariable linear regression revealed that PCOS remained significantly associated with increased total monocytes (adjusted β=0.11; 95% CI, 0.07–0.16), inflammatory monocytes (adjusted β=0.09; 95% CI, 0.05–0.13), and spleen size (adjusted β=9.0; 95% CI, 5.01–12.98). Women with PCOS had significantly elevated plasma norepinephrine levels (Figure S12E); these levels significantly correlated with total circulating monocytes and inflammatory monocytes (Figure S12F and S12G).

Cardiac Inflammation and Ventricular Remodeling After MI Are Augmented in PCOS Models

MI is associated with an inflammatory response, and immune hyperactivation promotes cardiac injury after MI.²⁹ We evaluated whether the augmented immune status in PAMH-induced PCOS mice could affect cardiac injury after MI. Compared with control mice, PCOS mice had significantly increased LSK cells and GMPs in the spleen on day 3 after MI (Figure S13A and S13B), along with significantly increased splenic numbers of total monocytes, Ly6C⁺ monocytes, and neutrophils (Figure S13C through S13F). Circulating and cardiac inflammatory cell numbers after MI were significantly increased in PCOS mice compared with control mice (Figure S13G through S13J and Figure 5A through 5E). Infarcted hearts of PCOS mice exhibited significantly higher expression of cytokines, including tumor necrosis factor–α, interleukin-1β, and interleukin-6, than did those of control mice (Figure 5F through 5H). However, cardiac inflammation after MI did not significantly differ between PAMH-treated male mice and control mice (Figure S14). PCOS mice showed significantly enhanced myocardial apoptosis in the infarct border zone (Figure 5I and 5J). PCOS mice displayed increased left ventricular circumference and decreased infarct wall thickness compared with control mice 3 weeks after MI (Figure 5K and 5L). Echocardiographic results revealed that PCOS mice had a significantly reduced ejection fraction after MI (Figure 5M). We further validated these findings in a prenatal androgenization-induced rat model of PCOS.³⁰ Consistent with findings in mice, PCOS rats had significantly elevated cardiac inflammatory cell numbers when compared with control rats after MI (Figure S15A through S15C). Sirius red staining revealed significantly reduced infarct wall thickness in PCOS rats 3 weeks after MI (Figure S15D and S15E), suggesting impaired scar formation. Cardiac magnetic resonance imaging revealed exaggerated end-systolic volume and contractile dysfunction in PCOS rats compared with control rats after MI (Figure S15F through S15H and Video S1).

Figure 5.

Cardiac inflammation and ventricular remodeling after myocardial infarction are augmented in polycystic ovary syndrome mice. A through D, Dot plots and quantification of monocyte, macrophage, and neutrophils in the myocardium of polycystic ovary syndrome (PCOS) mice and control mice at 3 days after myocardial infarction (MI; 10 mice/group). E, Immunofluorescence staining and quantification of macrophages and neutrophils in infarcted hearts from PCOS mice and control mice (5 mice/group). F through H, Transcriptional levels of proinflammatory cytokines in indicated groups (10 mice/group). I and J, Myocardial apoptosis in the infarct border zone was measured by TUNEL (terminal deoxynucleotidyl transferase dUTP nick-end labeling) staining (5 mice/group). K, Representative images of Masson trichrome staining in indicated groups at 3 weeks after MI. L, Quantification of scar circumference and scar thickness in indicated groups at 3 weeks after MI (10 or 12 mice/group). M, Representative images of small animal echocardiography and quantitative analysis of left ventricular ejection fraction (LVEF) in indicated groups at 3 weeks after MI (10 or 12 mice/group). Data are expressed as mean±SEM. Statistical analysis was performed by Student t test (for B, D, E, G, H, I, L, and M); Mann-Whitney U test was used for C and F. IL1β indicates interleukin-1β; IL6, interleukin-6; and TNFα, tumor necrosis factor–α.

To determine whether the augmented splenic hematopoietic response accounted for enhanced cardiac inflammation after MI in PCOS mice, PAMH-induced PCOS and control mice were subjected to splenectomy before MI induction. As shown in Figure S16, splenectomy reduced differences in circulating and cardiac inflammatory cell numbers between the 2 groups. These results indicate that increased splenic hematopoiesis in PCOS facilitated myocardial inflammation after MI.

As demonstrated previously, in vivo Vcam1 inhibition in splenic macrophages suppressed the augmented immune response in PCOS mice at baseline (Figure S10). We further examined the effects of Vcam1 inhibition on injury after MI in PAMH-induced PCOS mice. In vivo Vcam1 silencing significantly decreased splenic production of total monocytes, Ly6C⁺ monocytes, and neutrophils, accompanied by reduced circulating inflammatory cells, in both PCOS and control mice at 3 days after MI (Figure S17A and S17B). Meanwhile, Vcam1 silencing significantly reduced myocardial inflammation and cardiomyocyte apoptosis after MI (Figure S17C through S17G). Vcam1 silencing alleviated left ventricular dilation, increased infarct wall thickness (Figure S17H), and significantly improved contractile function (Figure S17I). To further confirm the cardioprotective effects of Vcam1 silencing in the splenic macrophages, we generated a mouse strain with macrophage-specific Vcam1 depletion (Vcam1^cKO; Figure S18A through S18D). Vcam1^cKO and Vcam1^flox/flox mice were treated with PAMH to induce PCOS. Compared with Vcam1^flox/flox PCOS mice, Vcam1^cKO PCOS mice exhibited decreased splenic leukocytosis and circulating inflammatory cells after MI (Figure S18E and 18F). Cardiac inflammation and remodeling after MI were significantly improved in Vcam1^cKO PCOS mice (Figure S18G through S18M). In addition, macrophage-specific Vcam1 depletion significantly improved injury and remodeling after MI in non-PCOS mice (Figure S18E through S18M).

Accelerated Atherosclerotic Plaque in PCOS Models

Atherosclerotic plaque development, 1 of the most common causes of MI, is pivotally driven by chronic immune response.³¹ Therefore, we investigated whether atherosclerotic plaque development is affected in PCOS mice. PCOS was induced by PAMH treatment in apoE^−/− mice, followed by establishment of vulnerable atherosclerotic plaques.¹³ Eight weeks after plaque induction, PCOS mice exhibited significantly larger lesions than control mice (Figure 6A and 6B). Moreover, PCOS mice displayed significantly increased necrotic plaque cores, intraplaque lipid accumulation, and decreased fibrous cap thickness (Figure 6A and 6B), which are known hallmarks of rupture-prone lesions.³² These plaque phenotypes were not accompanied by differences in blood pressure or circulating lipid levels between the 2 groups (Figure S19). The lesion phenotypes in PCOS mice were associated with significantly increased numbers of intraplaque monocytes, macrophages, and neutrophils (Figure 6C through 6G) and elevated inflammatory cytokine expression (Figure 6H through 6J). We validated these findings using apoE^−/− rats with prenatal androgenization-induced PCOS (Figure S20A). Rat models generally developed less advanced plaques (ie, smaller lesion size and necrotic core, less lipid accumulation) than mice models (Figure S20B), consistent with previous studies that considered rats an atherosclerosis-resistant species.³³ However, consistent with findings in mice, PCOS rats exhibited significantly larger lesion sizes, necrotic plaque cores, and intraplaque lipid accumulation and reduced fibrous cap thickness than control rats (Figure S20C through S20F).

Figure 6.

Atherosclerotic plaque development is accelerated in polycystic ovary syndrome mice. A, Representative images of the plaque sections stained with hematoxylin & eosin (H&E), Masson trichrome, or oil red O. B, Quantification of the atherosclerotic lesion size, necrotic core size, lipid content, and fibrous cap thickness in indicated groups (20 mice/group). C through F, Dot plots and quantification of intraplaque monocytes, macrophages, and neutrophils measured by flow cytometry in indicated groups (10 mice/group). G, Immunofluorescence staining of intraplaque macrophages and neutrophils in indicated groups (5 mice/group). H through K, Transcriptional levels of inflammatory markers in indicated groups (8 mice/group). Data are expressed as mean±SEM. Statistical analysis was performed by the Student t test (for B, D, F, and J) or Mann-Whitney U test (for E, H, and I). Ly6C indicates lymphocyte antigen 6 complex, locus C1; and PCOS, polycystic ovary syndrome.

We examined the effects of in vivo Vcam1 silencing on atherosclerosis development in mice with PAMH-induced PCOS and control mice. Vcam1 silencing significantly attenuated atherosclerotic plaque inflammation (Figure S21A through S21E), reduced lesion size, and improved markers of plaque vulnerability, including necrotic core size, intraplaque lipid, and fibrous cap thickness, in both PCOS and control mice (Figure S21F and 21G).

DISCUSSION

We documented previously unrecognized mechanisms through which PCOS, a prevalent female endocrine disorder, affects cardiovascular health through mechanisms unrelated to metabolic abnormalities. First, unbiased genome-wide RNA-seq analysis and immunostaining in mice indicated that PCOS hearts were characterized by increased macrophage accumulation. Second, cardiac macrophage abundance in PCOS can be attributed to excessive monocyte supply and infiltration, caused by augmented splenic monocytopoiesis owing to sympathetic activation. Women with PCOS consistently had higher plasma norepinephrine levels and increased spleen size, with notable correlations between norepinephrine levels and inflammatory monocyte counts. Third, PCOS predisposes animal models to increased injury after MI and atherosclerotic plaque development by means of increased tissue immune infiltration. Fourth, suppressing splenic myelopoiesis by conditional Vcam1 knockout in PCOS mice significantly improved ventricular remodeling after MI (Figure 7). Our findings provide novel mechanistic insights into PCOS-mediated detrimental effects on female cardiovascular health.

Figure 7.

Illustrations of the mechanisms underlying the negative cardiovascular effects of polycystic ovary syndrome. Polycystic ovary syndrome (PCOS) mobilized hematopoietic progenitor cells in the bone marrow, leading to increased circulating hematopoietic progenitor cells; meanwhile, splenic norepinephrine content is increased in PCOS mice, contributing to increased expression of a hematopoietic progenitor retention factor Vcam1 (vascular cell adhesion molecule 1) in splenic macrophages, subsequently leading to increased splenic retention of hematopoietic progenitor cells and monocytopoiesis; increased circulating Ly6C⁺ monocyte supply enhanced inflammatory burden in heart and atherosclerotic plaques, promoting remodeling after myocardial infarction (MI) and atherosclerotic plaque instability.

Reducing cardiovascular disease (CVD) burden in women remains an ongoing challenge. Recent studies have estimated that >270 million women worldwide have CVD, with nearly 9 million CVD-related deaths in 2019 alone.³ Investigations of female-specific biological factors, risk factor prevalence, and sociopsychological factors have resulted in substantial declines in cardiovascular deaths among women.³⁴ Nevertheless, women with CAD still experience worse outcomes than their male counterparts, with a trend toward increased incidence of acute coronary syndrome among young women in recent years.^35,36 Understanding the pathologic mechanisms underlying female-specific CVD risk factors may open new therapeutic avenues for cardiovascular protection in women.² PCOS, an infertility-causing endocrine condition affecting 10% of women of reproductive age worldwide, has emerged as a female-specific CVD risk factor.^3,37 PCOS has been associated with increased subclinical CVD, such as arterial stiffness, carotid intima-media thickness, and coronary artery calcification.³⁸ Accumulating evidence suggests an increased incidence of overt CVD events in women with PCOS.^6,7 A recent large-population study (174 660 participants) reported an increased prevalence of atherosclerotic CVD in women with PCOS (eg, a 40% increased MI risk and an ≈30% increased stroke risk).^7,39 However, the mechanisms underlying PCOS–CVD associations remain largely unknown. To our knowledge, the current study provides the first experimental evidence that PCOS promotes remodeling after MI and atherosclerotic plaque vulnerability, independent of metabolic risk factors, by a novel mechanism involving augmented immune cell infiltration. These findings offer new insights into the female-specific mechanisms underlying CVD, along with potential novel targets for reducing CVD burden in women. Our findings may help raise awareness of CVD risk among young women, an age group frequently affected by PCOS and facing the greatest decline in CVD risk awareness during the past decade.⁴⁰

Initial clinical studies exploring PCOS–CVD associations observed prevalent comorbid metabolic conditions in women with PCOS. Insulin resistance, dyslipidemia, obesity, and metabolic syndrome are frequently observed in women with PCOS. These comorbid metabolic conditions likely contribute to the increased CVD risks in PCOS; however, whether PCOS affects cardiovascular health independent of these traditional risk factors remains unknown.^3,37 Here, we used a PAMH-induced PCOS model with no detectable alterations in body weight, glucose or lipid metabolism, or blood pressure. This PCOS model more closely resembles the human lean PCOS phenotype, a minor phenotype presenting with normal BMI and comparable or less obvious clinical manifestations compared with women with overweight or obese PCOS.⁴¹ Our findings suggest that PCOS exacerbates adverse outcomes in clinically relevant models of heart failure after MI and vulnerable atherosclerotic plaques because of an augmented immune response. These data suggest that PCOS may negatively affect cardiovascular health independent of classical risk factors. In an epidemiologic study, PCOS was consistently associated with increased risks of coronary heart disease after adjusting for BMI and diabetes.⁷ Women with PCOS may benefit from adjunctive anti-inflammatory interventions for cardioprotection in addition to managing traditional CAD risk factors. The current findings highlight the need for future cohort studies or reanalysis of existing cohort data sets to establish the contributing roles of immune measures in the increased CVD risk in women with PCOS.

Despite numerous clinical studies concerning PCOS–CVD associations, mechanistic explorations remain limited. Using genome-wide RNA-seq analysis and different immunostaining modalities, we identified the presence of increased macrophage accumulation as a major pathologic change in PCOS mouse hearts. These findings have both biologic and clinical significance. Increased cardiac macrophage accumulation reportedly exerts net detrimental effects in cardiac diseases of various pathogeneses, including ischemic heart disease and diastolic cardiomyopathy.⁴² Expansion in cardiac macrophages continuously contributes to the progression of heart failure.⁴³ Our findings suggest that alleviating cardiac immune burden may afford cardioprotective effects in women with PCOS. Our findings are consistent with the notion that PCOS is a state of chronic inflammation.⁴⁴ Elevated levels of inflammatory markers have been documented in women with PCOS, including tumor necrosis factor–α, C-reactive protein, and interleukin-6.⁴⁵ The correlation between serum levels of these markers and insulin resistance and fatty mass in women with PCOS⁴⁶ highlights the potential contribution of comorbid metabolic conditions to the inflammatory state in PCOS. However, the precise mechanism responsible for increased inflammation in women with PCOS remains unclear. Using a PAMH PCOS model without overt metabolic abnormalities, we revealed that augmented splenic monocytopoiesis in PCOS, caused by sympathetic activation, represents a mechanism responsible for increased circulating monocytes and cardiac inflammation burden. Consistent with these animal findings, women with PCOS had increased circulatory monocyte counts and spleen sizes compared with control participants, and plasma norepinephrine levels were increased and correlated with inflammatory monocyte counts in PCOS. The mechanisms underlying the increased sympathetic tone in PCOS warrant further investigation. Androgen excess may represent 1 potential mechanism, considering that testosterone levels were elevated in current PCOS models. Androgen excess can intensify sympathetic tone in different tissues (eg, adipose tissue).^47,48 The current study provides the first evidence that PCOS mouse hearts are characterized by increased macrophage accumulation through augmented splenic myelopoiesis, contributing to exaggerated cardiac injury and remodeling after MI.

Limitations

Our study had several limitations. First, the current human data are limited by the inability to validate increased myocardial macrophage accumulation observed in PCOS animal models, given the difficulties in collecting cardiac tissues from women with PCOS. Second, the recruited women with PCOS more closely represented the lean PCOS subtype. Future studies with larger sample sizes are warranted to explore the association between different PCOS subtypes and CAD prognosis. Third, all female mice used here were premenopausal. It is important to investigate whether heightened cardiac inflammation in PCOS models persists after menopause, a stage that witnesses suppression of PCOS symptoms.⁴⁹ In addition, the animal models used failed to mimic the complex pathogenesis of PCOS in humans comprehensively, and extrapolating these data to humans requires caution.

Conclusion and Perspectives

PCOS promotes macrophage accumulation in cardiac tissues by increasing extramedullary myelopoiesis, exacerbating cardiac inflammation and ventricular remodeling after MI. These findings highlight the potential significance of PCOS screening for prognostic evaluation in women with CAD. Suppression of splenic myelopoiesis may represent a potential therapeutic option for cardioprotection in women with PCOS.

ARTICLE INFORMATION

Sources of Funding

The study was supported by the Basic Science Center Project of National Natural Science Foundation of China (grant 82088102), National Key Research and Development Program of China (grants 2021YFC2700701, 2021YFC2700603, 2022YFC2703803, and 2022YFC2703001), National Natural Science Foundation of China (grants 82171613, 82071731, 82270342, 81601238, 82192864, and 81800307), Chinese Academy of Medical Sciences Innovation Fund for Medical Sciences (grant 2019-I2M-5-064), the Science and Technology Commission of Shanghai Municipality (grants 21Y11907600 and 22QA1405400), Shanghai Municipal Commission of Health and Family Planning (grants 20215Y0216 and 20234Y0048), Collaborative Innovation Program of Shanghai Municipal Health Commission (grant 2020CXJQ01), Clinical Research Plan of Shanghai Hospital Development Center (grant SHDC2020CR1008A), Shanghai Clinical Research Center for Gynecological Diseases (grant 22MC1940200), Shanghai Urogenital System Diseases Research Center (grant 2022ZZ01012), Shanghai Frontiers Science Research Base of Reproduction and Development, The Science and Technology Commission of Quzhou Municipality (grant 2022K54), Open Fund Project of Key Laboratory of Reproductive Genetics, Ministry of Education, Zhejiang University (grant KY2022035), and Open Fund Project of Guangdong Academy of Medical Sciences (grant YKY-KF202202).

Disclosures

None.

Supplemental Material

Methods

Video S1

Tables S1–S3

Figures S1–S21

Supplementary Material

cir-148-1958-s001.pdf ^{(6.7MB, pdf)}

Download video file ^{(878KB, mp4)}