Abstract
The incidence of polycystic ovary syndrome (PCOS) due to high-fat diet (HFD) consumption has been increasing significantly. However, the mechanism by which a HFD contributes to the pathogenesis of PCOS has not been elucidated. Proprotein convertase subtilisin/kexin type 9 (PCSK9) is a key protein that regulates cholesterol metabolism. Our previous study revealed abnormally high PCSK9 levels in serum from patients with PCOS and in serum and hepatic and ovarian tissues from PCOS model mice, suggesting that PCSK9 is involved in the pathogenesis of PCOS. However, the factor that induces high PCSK9 expression in PCOS remains unclear. In this study, Pcsk9 knockout mice were used to further explore the role of PCSK9 in PCOS. We also studied the effects of a HFD on the expression of PCSK9 and sterol regulatory element-binding protein 2 (SREBP2), a regulator of cholesterol homeostasis and a key transcription factor that regulates the expression of PCSK9, and the roles of these proteins in PCOS pathology. Our results indicated HFD may play an important role by inducing abnormally high PCSK9 expression via SREBP2 upregulation. We further investigated the effects of an effective SREBP inhibitor, fatostain, and found that it could reduce HFD-induced PCSK9 expression, ameliorate hyperlipidemia and improve follicular development in PCOS model mice. Our study thus further elucidates the important role of an HFD in the pathogenesis of PCOS and provides a new clue in the prevention and treatment of this disorder.
Introduction
Polycystic ovary syndrome (PCOS) is a common endocrine disease among women of reproductive age. This heterogeneous syndrome is characterized by polycystic ovaries, anovulation and hyperandrogenism (Conway et al. 2014). PCOS is often associated with obesity, disordered lipid metabolism and insulin resistance, consequently, patients with PCOS face a high risk of long-term complications, such as diabetes or cardiovascular disease. The current understanding indicates that PCOS is not only limited to female reproductive ovulatory and endocrine abnormalities but also includes systemic metabolic problems.
A high-fat diet (HFD) is an important cause of obesity and lipid metabolism disorders. In recent years, HFD has also been identified as one of the major contributors to the increased incidence of PCOS (de Melo et al. 2015). Hyperandrogenism is another important feature of PCOS. Many studies have shown that high androgen levels promote the growth of early follicles but hinder further follicular maturation and ovulation (Vendola et al. 1999, Steckler et al. 2007). Some clinical studies have demonstrated that HFD-induced abdominal obesity is not only closely related to serum hyperandrogenism but also aggravates the symptoms of hyperandrogenemia in susceptible people (McCartney et al. 2007, Franks 2008). These findings further suggest that HFD is an important contributor to PCOS. Correspondingly, reducing body weight and improving lipid metabolism are the main clinical treatment options for PCOS (Dokras et al. 2016, Legro et al. 2016). However, the role of HFD in the pathogenesis of PCOS has not yet been elucidated.
Proprotein convertase subtilisin/kexin type-9 (PCSK9) is a secreted protein that was first identified in 2003 (Seidah et al. 2003). Both diet and metabolic status may affect PCSK9 expression (Krysa et al. 2017). PCSK9 functions mainly as a negative regulator of lipid metabolism by binding to low-density lipoprotein receptors (LDLR) on cell membranes (Tavori et al. 2015). Studies have shown that an abnormally high serum concentration of PCSK9 is related to the occurrence and development of cardiovascular disease (Folsom et al. 2009). Other studies have suggested that PCSK9 is involved in regulating the proliferation and apoptosis of human glioma cells, umbilical vein endothelial cells and nerve cells (Bingham et al. 2006, Kysenius et al. 2012, Wu et al. 2012, Piao et al. 2015). In our previous study, we showed that PCSK9 is detected at abnormally high concentrations in serum from PCOS patients, as well as in serum and hepatic and ovarian tissues from PCOS model mice, which suggests that PCSK9 is involved in the pathogenesis of PCOS (Wang et al. 2019).
Sterol regulatory element-binding protein 2 (SREBP2) is an important regulator of cholesterol homeostasis in cells and a key transcription factor that regulates the expression of PCSK9 (Horton et al. 2003). Some drugs can reduce PCSK9 expression by downregulating SREBP2 in mice fed a HFD (Sui et al. 2018). Therefore, we believe that HFD induces the abnormal expression of PCSK9 by upregulating SREBP2, thereby playing a role in the pathogenesis of PCOS.
In this study, we first obtained a Pcsk9 knockout mouse strain and used this to provide further evidence for a role of PCSK9 in PCOS. We then explored the ability of a HFD to induce PCSK9 expression by upregulating SREBP2, the effects of androgen on PCSK9 expression and their roles in the pathological features of PCOS. Finally, we studied the effects of fatostain, an efficient SREBP inhibitor, on the HFD-induced expression of PCSK9 and pathological features of PCOS. This study aimed to confirm HFD as a key factor in the induction of abnormal PCSK9 expression in PCOS and demonstrate a role for SREBP2 in this process, providing new clues in the prevention and treatment of PCOS.
Materials and methods
Animals and ethics
Pcsk9 knockout (KO-PCOS) mice on the C57BL/6 background were provided by Biocytogen (SCXK (京) 2019-0001). C57BL/6 WT female mice were obtained from Chengdu Dashuo Biological Technology (SCXK (川) 2020-030). The mice were maintained under standard housing conditions with a 12 h light:12 h darkness cycle, an environment with controlled temperature and humidity and food and water ad libitum. All animal studies were conducted with the approval of the Animal Use and Care Committee of Sichuan University.
Animal experiments
In the first experiment, female mice were genotyped after birth. After weaning (day 25 of age), the mice were divided into WT and KO-PCSK9 groups (n = 10). Mice in both groups were fed a HFD (D12492, Research Diets, New Brunswick, USA) and administered DHEA (D400, Sigma; 6 mg per 100 g body weight, dissolved in 0.1 mL of sesame oil) once a day via s.c. injection for 20 consecutive days to generate the PCOS model mice according to the method of Lai et al. (2014). Body weight was recorded every 4 days. The estrous cycle stage was determined from day 10 after the first injection to the end of the experiment. We collected all samples from mice during estrum to eliminate any influence of variations in the estrous cycle on end points measured. Fasting serum samples were collected to evaluate lipid metabolism and ovarian hormone concentrations. The left ovary of each mouse was dissected and subjected to Oil Red O staining, histochemical staining with hematoxylin and eosin (H&E) and transmission electron microscopy (TEM). From the right ovary, RNA was extracted for qPCR and proteins for Western blotting.
In the second experiment, 25-day-old WT female C57BL/6 mice were randomly divided into four groups (n = 6 per group): control, DHEA (at the same dosage as in the previous experiment), HFD and HFD combined with DHEA (HFD+DHEA). The mice were treated for 20 consecutive days, and their body weights and estrous cycles were monitored. Serum samples were collected to analyze PCSK9 expression, lipid metabolism and ovarian hormone levels using biochemical methods and enzyme-linked immunosorbent assays (ELISAs). The left ovary of each mouse was dissected and subjected to histochemical staining with H&E; the right ovary was processed for qPCR and Western blotting. The liver tissues were dissected and subjected to Oil Red O staining, qPCR and Western blotting.
In the third experiment, 25-day-old female C57BL/6 mice were randomly divided into four groups (n = 6 per group): treated with a HFD alone (HFD group) or with the SREBP inhibitor fatostain (S9785, Selleck, TX, USA; HFD+fatostain group), both HFD and DHEA (HFD+DHEA group) or with fatostain (HFD+DHEA+fatostain group). Fatostain was administered intraperitoneally at a dose of 3 mg/100 g body weight once every 3 days for 20 consecutive days. Body weight and the estrous cycle were monitored in all mice. PCSK9 expression, lipid metabolism and ovarian hormone concentrations were analyzed in collected serum samples by biochemical methods and ELISAs.
Cell culture experiments
Cultured normal hepatocytes (WRL) or ovarian granulosa cancer (KGN) cell lines were each divided into four groups: treated with dihydrotestosterone (DHT, 10−4 M; abs47034291, Absin, Shanghai, China), sodium oleate (0.2 µM; CAS#[143-19-1], BBI life sciences, Shanghai, China), both sodium oleate and DHT (sodium oleate+DHT) or only control solution (control). After 24 h, the cells were collected, and the expression of PCSK9 mRNA and protein and the accumulation of lipid droplets were measured.
Analyses of the estrous cycle and serum hormone concentrations
The estrous cycle stage was determined by microscopic analysis of the predominant cell types obtained via vaginal smears, according to a previous study (Marcondes et al. 2002).
The serum testosterone, estradiol (E2) and progesterone (P4) concentrations in mice from the first animal experiment described above were measured using commercial Iodine [125I] Radioimmunoassay kits (North Institute of Biological Technology, Beijing, China). For all assays, the intra- and interassay errors were less than 10 and 15%, respectively. The limits of sensitivity were 0.02 ng/mL for testosterone, 5 pg/mL for E2 and 0.2 ng/mL for P4, and the intra- and interassay coefficients of variability (CV) were less than 10 and 15%, respectively, for all assays. In the other two animal experiments, the serum testosterone, E2 and P4 concentrations were measured using sandwich ELISAs according to the manufacturer’s instructions (JL25196/11790/20678, Jianglai Biotechnology, Shanghai, China).
Lipid profile analysis
The serum samples from mice were analyzed using biochemical analysis kits (A113/111/112, Nanjing Jiancheng Bioengineering Institute, Nanjing, China) to determine the serum lipid profiles, including LDL-C, total cholesterol (TC) and HDL-C.
Histological staining and ovarian morphology analysis
The approach used for estimating follicle numbers is based on a histological sampling of all sections of ovary for the number of preantral ,antral follicle, corpora lutea, cyst-like and atretic follicles. This approach essentially entails that paraffin-embedded ovaries were longitudinally and serially sectioned at 4 μm, mounted on a glass slide and stained with H&E. Ovarian follicles at different stages of development were classified according to a previously published method (Luo et al. 2008, Caldwell et al. 2014). Histological analysis was performed by two researchers independently to avoid duplicate counting, and the results were confirmed by a pathologist.
Oil Red O staining
Frozen 10-μm tissue sections were stained with Oil Red O to detect neutral lipid accumulation according to a standard procedure (G1260, Solarbio, Beijing, China), and lipid droplets in the cells were quantified using ImageJ software (https://imagej.nih.gov/ij/). The tissue sections were observed and imaged using an Olympus microscope (Olympus).
Transmission electron microscopy
Ovarian tissue samples were cut into 1-mm3 sections and fixed in 2.5% glutaraldehyde (pH 7.4) for 4 h. The samples were then flushed with 0.1 M phosphate buffer (pH 7.2) three times and fixed in osmic acid for 2 h. Next, the samples were subjected to acetone dehydration and embedded in Epon-Araldite resin (18030, Ted Pella, CA, USA). Semi-thin sections were cut into ultrathin sections and counterstained with 3% uranyl acetate and 0.3% lead citrate. Finally, the ovarian tissues were observed using a TEM (JEM1230, JEOL, Tokyo, Japan).
Immunohistochemistry (IHC)
IHC was performed according to the kit instructions (SP-9001, ZSGBBIO, Beijing, China). Ovarian and liver tissues were stained with a primary antibody against PCSK9 (55206-1-AP; Proteintech, Wuhan, China) at the respective dilutions of 1:100 and 1:600 in PBS. After 3,3′-diaminobenzidine staining (#ZLI9107, ZSGB-BIO, Beijing, China) and hematoxylin counterstaining, the tissues were photographed using an Olympus microscope (Olympus). As a negative control, duplicate sections were subjected to IHC without primary antibodies.
ELISA for serum PCSK9
The PCSK9 concentrations in mouse sera were measured using a sandwich ELISA (MPC900, R&D Systems, Minneapolis, MN, USA) according to the manufacturer’s instructions. Each sample was measured in duplicate. The intra- and interassay CV were 5.4 and 7.1%, respectively.
RNA preparation and quantitative RT-PCR
RNA isolation was performed using TRIzol reagent (10296010, Invitrogen) according to the manufacturer’s instructions. RT-PCR was performed using HiScript II Reverse Transcriptase (#R211-01, Vazyme, Nanjing, China) and SYBR Green probes (#R123-01, Vazyme, Nanjing, China) according to the manufacturers’ protocols. The sequences of primers used for measuring mRNA expression levels are shown as follows: mouse Ldlr, 5′-GCATCAGCTTGGACAAGGTGT-3′ and 5′-GGGAACAGCCACCATTGTTG-3′; mouse Casp3 (caspase-3), 5′-CCATACATGGGAGCAAGTCAGT-3′ and 5′-GGCCCATGAATGTCTCTCTGA-3′; mouse Pcsk9, 5′-GAGCACATTGCATCC-3′ and 5′-TGCAAAATCAAGGAGCATGGG-3′; mouse Actb (β-actin), 5′-ATCTGGCACCACACCTTC-3′ and 5′-AGCCAGGTCCAGACGCA-3′; human Pcsk9, 5′-ACGATGCCTGCCTCTACTCC-3′ and 5′-GCCTGTGATGTCCCACTCTGT-3′; and human GAPDH, 5′-TGCACCACCAACTGCTTAGC-3′ and 5′-GGCATGGACTGTGGTCATGAG-3′. All measurements were performed in triplicate. Actb or GAPDH was used as a reference to normalize gene expression. Relative gene expression was quantified using the 2−ΔΔCt method.
Western blotting
Proteins were extracted using a total protein extraction kit (#KGP2100, KeyGEN BioTECH, Nanjing, China), and protein concentrations were measured using an Enhanced BCA Protein Assay Kit (#P0010, Beyotime Biotechnology, Shanghai, China). Forty micrograms of protein per extract were loaded into the lanes. The following primary antibodies were used: anti-PCSK9 rabbit polyclonal antibody (1:2000 dilution; ab31762; Abcam) (1:500; 55206-1-AP, Proteintech, Wuhan, China), anti-LDLR rabbit polyclonal antibody (1:5000; ab52818; Abcam), anti-caspase-3 rabbit polyclonal antibody (1:1000, 19677-1-AP, Proteintech), anti-SREBP2 rabbit polyclonal antibody (1:400; ab30682; Abcam) and anti-β-actin rabbit polyclonal antibody (1: 1000; bs-0061R; Bioss, Beijing, China). The secondary antibody was a horseradish peroxidase-labeled goat anti-rabbit IgG (1: 1000; bs-0295G-HRP; Bioss, Beijing, China). A chemiluminescence reagent (#P0018S, Beyotime Biotechnology) was used to visualize the blots. All experiments were repeated at least three times. The protein band densities were quantified using ImageJ software (https://imagej.nih.gov/ij/).
Evaluation of ovulatory function
After treatment, the animals were superovulated by administering 5 IU of pregnant mare serum gonadotropin (M2620, Aibei Biotechnology, Nanjing, China) via intraperitoneal injection, followed by 5 IU of human chorionic gonadotropin (M2520, Aibei Biotechnology) 48 h later. Cumulus–oocyte complexes were collected 16 h after hCG (human chorionic gonadotrophin) injection by puncturing the oviductal ampulla with a fine pair of forceps and a syringe and needle under a stereomicroscope, and the cumulus cells were removed by brief incubation in 0.2% hyaluronidase (M2215, Aibei Biotechnology). Denuded oocytes were observed under a stereomicroscope. The total number of ovulated oocytes in each mouse was used as the ovulation rate.
Statistical analysis
All statistical analyses were performed using Prism software (GraphPad). Student’s t-test was used to compare normal or parametric variables between two groups, and an ANOVA followed by the Bonferroni post-test was used to compare multiple groups. Correlations were calculated using Pearson’s correlation coefficient. Data are presented as the mean ± s.d.. A P value < 0.05 was considered statistically significant.
Results
Pcsk9 knockout could partially reverse the abnormal lipid metabolism and ovarian function in PCOS mice
In the first experiment, mice in both the WT-PCOS and KO-PCOS groups exhibited irregular estrous cycles (Fig. 1A) and polycystic changes (Fig. 1F). Serum concentrations of reproductive hormones were tested, and ovarian follicles were observed to evaluate ovarian function. No significant differences were observed in the serum testosterone, E2 and P4 concentrations between KO-PCOS and WT-PCOS mice (Fig. 1B, C and D). Compared with WT-PCOS mice, KO-PCOS mice had a significantly reduced ratio of atretic follicles (Fig. 1G). KO-PCOS mice also had slight increases in preantral follicles, cystic follicles and the corpora lutea, but these differences were not statistically significant (Fig. 1G). In addition, we subjected both groups of mice to artificial ovulation experiments. The results showed that the number of ovulations in KO-PCOS mice was significantly less than that in WT-PCOS mice (Fig. 1E), indicating that in the context of PCOS, knocking out Pcsk9 partly restored follicular development.
Comparison of ovarian function in WT-PCOS and KO–PCOS mice. (A) Representative estrous cycle of one mouse from each group. D, diestrus; M, metestrus; E, estrus; P, proestrus. (B, C and D) Serum testosterone , estradiol (E2) and progesterone (P4). (E) quantitative analysis the induced oocyte of mice. (F) Representative hematoxylin eosin (H&E) staining of mice ovarian sections from each group. (G) Percentage of follicles at some levels. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a, and b, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Comparison of ovarian function in WT-PCOS and KO–PCOS mice. (A) Representative estrous cycle of one mouse from each group. D, diestrus; M, metestrus; E, estrus; P, proestrus. (B, C and D) Serum testosterone , estradiol (E2) and progesterone (P4). (E) quantitative analysis the induced oocyte of mice. (F) Representative hematoxylin eosin (H&E) staining of mice ovarian sections from each group. (G) Percentage of follicles at some levels. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a, and b, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Comparison of ovarian function in WT-PCOS and KO–PCOS mice. (A) Representative estrous cycle of one mouse from each group. D, diestrus; M, metestrus; E, estrus; P, proestrus. (B, C and D) Serum testosterone , estradiol (E2) and progesterone (P4). (E) quantitative analysis the induced oocyte of mice. (F) Representative hematoxylin eosin (H&E) staining of mice ovarian sections from each group. (G) Percentage of follicles at some levels. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a, and b, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Body weight and the serum lipid profile are key indicators in lipid metabolism evaluations. Our results showed that knocking out Pcsk9 significantly reduced the serum concentrations of TC, LDL-C and HDL-C in PCOS mice but had no significant effect on body weight (Fig. 2A, B, C and D). Oil Red O staining and TEM revealed a greater accumulation of lipid droplets in the ovarian cells of KO-PCOS mice relative to that in WT-PCOS mice (Fig. 2E and F). Meanwhile, Western blotting revealed a significantly higher level of LDLR protein in the ovaries of KO-PCOS mice relative to that in WT-PCOS mice; however, there was no difference in Ldlr mRNA expression between the two groups (Fig. 2G and H). In addition, a significantly lower level of CASPASE3 was detected in the ovaries of KO-PCOS mice relative to that in WT-PCOS mice (Fig. 2I and J). These results indicated that knocking out Pcsk9 could partially reverse the abnormal lipid metabolism associated with PCOS in the mouse model.
Comparison of the body weight, serum lipid profiles and liver lipid deposition, LDLR and caspase3 in WT-PCOS and KO-PCOS mice. (A) Body weight. (B, C and D) serum TC, LDL-C, HDL-C level. (E) Representative photomicrographs of ovary sections stained with Oil Red O in the mice. (F) Representative electron micrographs from WT-PCOS and KO-PCOS mice, taken with a TEM. LD, lipid droplets. (G and I) Gene expression of Ldlr and Casp3 in the ovaries by RT-PCR analysis. (H and J) Western blot analysis of LDLR and CASP3 in the ovaries (left) and densitometry quantification (right). Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a and b, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Comparison of the body weight, serum lipid profiles and liver lipid deposition, LDLR and caspase3 in WT-PCOS and KO-PCOS mice. (A) Body weight. (B, C and D) serum TC, LDL-C, HDL-C level. (E) Representative photomicrographs of ovary sections stained with Oil Red O in the mice. (F) Representative electron micrographs from WT-PCOS and KO-PCOS mice, taken with a TEM. LD, lipid droplets. (G and I) Gene expression of Ldlr and Casp3 in the ovaries by RT-PCR analysis. (H and J) Western blot analysis of LDLR and CASP3 in the ovaries (left) and densitometry quantification (right). Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a and b, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Comparison of the body weight, serum lipid profiles and liver lipid deposition, LDLR and caspase3 in WT-PCOS and KO-PCOS mice. (A) Body weight. (B, C and D) serum TC, LDL-C, HDL-C level. (E) Representative photomicrographs of ovary sections stained with Oil Red O in the mice. (F) Representative electron micrographs from WT-PCOS and KO-PCOS mice, taken with a TEM. LD, lipid droplets. (G and I) Gene expression of Ldlr and Casp3 in the ovaries by RT-PCR analysis. (H and J) Western blot analysis of LDLR and CASP3 in the ovaries (left) and densitometry quantification (right). Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a and b, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
A high-fat diet increased PCSK9 in the serum and hepatic and ovarian tissues of mice
Since DHEA could be converted to testosterone and then the more active DHT in some tissues of body in vivo. In vitro, we used non-aromatizable androgen-DHT for its most direct action to explore the factor promoting increased PCSK9 expression. We treated normal WRL human hepatocytes and KGN human ovarian granulosa cells with DHT, sodium oleate or a combination thereof for 24 h. Although all three treatments significantly induced PCSK9 expression in both cell lines, sodium oleate and sodium oleate+DHT induced significantly stronger responses than DHT alone, indicating that sodium oleate more effectively induced PCSK9 expression than DHT (Fig. 3A, B, C and D). Similarly, Oil Red O staining revealed that all three treatments increased lipid accumulation in both WRL and KGN cells; however, sodium oleate and sodium oleate+DHT had a stronger effect on lipid accumulation (Fig. 3E and F). These results suggest that although both DHT and sodium oleate can induce PCSK9 expression and lipid metabolism alterations in vitro, sodium oleate had a stronger effect.
Effects of DHT or/and sodium oleate on PCSK9 expression and lipid droplets in WRL and KGN cells. (A and C) Gene expression of Pcsk9 in WRL and KGN cells by RT- PCR analysis. (B) Western blot analysis (left) and densitometry quantification (right) of PCSK9 in WRL cells. (D) Western blot analysis (left) and densitometry quantification (right) of PCSK9 in KGN cells. (E and F) Respresentative photomicrographs of treated WRL and KGN cells stained with Oil Red O. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a and b, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Effects of DHT or/and sodium oleate on PCSK9 expression and lipid droplets in WRL and KGN cells. (A and C) Gene expression of Pcsk9 in WRL and KGN cells by RT- PCR analysis. (B) Western blot analysis (left) and densitometry quantification (right) of PCSK9 in WRL cells. (D) Western blot analysis (left) and densitometry quantification (right) of PCSK9 in KGN cells. (E and F) Respresentative photomicrographs of treated WRL and KGN cells stained with Oil Red O. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a and b, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Effects of DHT or/and sodium oleate on PCSK9 expression and lipid droplets in WRL and KGN cells. (A and C) Gene expression of Pcsk9 in WRL and KGN cells by RT- PCR analysis. (B) Western blot analysis (left) and densitometry quantification (right) of PCSK9 in WRL cells. (D) Western blot analysis (left) and densitometry quantification (right) of PCSK9 in KGN cells. (E and F) Respresentative photomicrographs of treated WRL and KGN cells stained with Oil Red O. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a and b, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
In the second animal experiment, compared with the control group, the serum concentration of PCSK9 was significantly decreased in the HFD group and significantly increased in the HFD+DHEA group but did not change significantly in the DHEA group (Fig. 4A). Immunohistochemical staining revealed increased PCSK9 expression in the hepatic and ovarian tissues of mice in the HFD and HFD+DHEA groups relative to the control group, with no significant differences between the two treatment groups. In contrast, no significant differences were observed between the DHEA and control groups (Fig. 4B and E). Similar results were obtained with qPCR and Western blotting (Fig. 4C, D, F and G). No obvious differences in liver morphology were observed between the groups (Fig. 5A); however, Oil Red O staining revealed a significant increase in hepatic lipid droplet accumulation in both the HFD and HFD+DHEA groups compared with the control group, but no significant change in the DHEA group (Fig. 5B). Correspondingly, significant increases in body weight were observed in mice in the HFD and HFD+DHEA groups but not in mice in the DHEA and control groups (Fig. 5C). Mice in the HFD and HFD+DHEA groups also exhibited increases in serum TC, LDL-C and HDL-C concentrations, whereas these changes were not observed in the DHEA group (Fig. 5D, E and F ).
Effects of DHEA or/and HFD on PCSK9 expression in mice. (A) Serum expression PCSK9 level by the ELISA assay. (B and E) Immunohistochemical staining of livers and ovaries in the mice of different groups; Brown staining represents the positive expression of PCSK9. (C and F) Expression of Pcsk9 mRNA in the livers and ovaries. (D and G) Western blot analysis (left) and densitometry quantification (right) of PCSK9 protein in the livers and ovaries. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a and b, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Effects of DHEA or/and HFD on PCSK9 expression in mice. (A) Serum expression PCSK9 level by the ELISA assay. (B and E) Immunohistochemical staining of livers and ovaries in the mice of different groups; Brown staining represents the positive expression of PCSK9. (C and F) Expression of Pcsk9 mRNA in the livers and ovaries. (D and G) Western blot analysis (left) and densitometry quantification (right) of PCSK9 protein in the livers and ovaries. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a and b, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Effects of DHEA or/and HFD on PCSK9 expression in mice. (A) Serum expression PCSK9 level by the ELISA assay. (B and E) Immunohistochemical staining of livers and ovaries in the mice of different groups; Brown staining represents the positive expression of PCSK9. (C and F) Expression of Pcsk9 mRNA in the livers and ovaries. (D and G) Western blot analysis (left) and densitometry quantification (right) of PCSK9 protein in the livers and ovaries. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a and b, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Effects of DHEA or/and HFD on liver lipid droplets accumulation, body weight and serum lipid profile. (A and B) Representative photomicrographs of liver sections stained with hematoxylin & eosin and Oil Red O staining. (C, D, E and F) Body weight, TC, LDL-C and HDL-C. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a and b, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Effects of DHEA or/and HFD on liver lipid droplets accumulation, body weight and serum lipid profile. (A and B) Representative photomicrographs of liver sections stained with hematoxylin & eosin and Oil Red O staining. (C, D, E and F) Body weight, TC, LDL-C and HDL-C. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a and b, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Effects of DHEA or/and HFD on liver lipid droplets accumulation, body weight and serum lipid profile. (A and B) Representative photomicrographs of liver sections stained with hematoxylin & eosin and Oil Red O staining. (C, D, E and F) Body weight, TC, LDL-C and HDL-C. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a and b, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Ovarian morphology and function were evaluated in each group of mice. Those in the control group showed a complete and regular estrus cycle, whereas those mice in the DHEA and HFD+DHEA groups lost their cycles, with most remaining in estrus. Although mice in the HFD group continued to exhibit a periodic estrous cycle, the cycle length was longer than that in the control group (Fig. 6A). In addition, the serum testosterone and E2 concentrations of mice in the HFD group were higher than those in the control group but lower than those in the DHEA and HFD+DHEA groups (Fig. 6B and C ). There were no differences in the serum concentration of P4 between the four groups of mice (Fig. 6D). These results indicate that DHEA is a main factor influencing the secretion of ovarian sex hormones to disrupt the estrous cycle in PCOS mice. Although a HFD can also affect the estrous cycle and sex hormone secretion in mice to a certain extent, its effect is much weaker than that of DHEA. H&E staining of paraffin-embedded sections of mouse ovaries showed normal ovarian morphology in the control and HFD groups. In contrast, ovarian tissues from the DHEA and HFD+DHEA groups contained typical cystic follicles, with significantly increased ratios of cystic follicles and atresia follicles and a significant decrease in the ratio of corpora lutea (Fig. 6E and F ). These results indicated that DHEA is a main factor affecting ovarian morphology and function in PCOS mice.
Effects of DHEA or/and HFD on ovarian function in mice. (A) Representative estrous cycle of one mouse from each group. (B, C and D) Serum testosterone, E2 and P4. (E) Representative H&E staining of ovarian section of mice for each group. (F) Percentage of follicles at some levels. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a, b and c, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Effects of DHEA or/and HFD on ovarian function in mice. (A) Representative estrous cycle of one mouse from each group. (B, C and D) Serum testosterone, E2 and P4. (E) Representative H&E staining of ovarian section of mice for each group. (F) Percentage of follicles at some levels. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a, b and c, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Effects of DHEA or/and HFD on ovarian function in mice. (A) Representative estrous cycle of one mouse from each group. (B, C and D) Serum testosterone, E2 and P4. (E) Representative H&E staining of ovarian section of mice for each group. (F) Percentage of follicles at some levels. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a, b and c, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
In summary, our findings suggest that a HFD induces increased PSCK9 levels in the serum and tissues of PCOS mice and leads to abnormal lipid metabolism. Although a HFD also affects the levels of serum sex hormones in PCOS mice, it is not the main factor affecting follicular development.
SREBP2 upregulation is associated with the HFD-induced increase in PCSK9 expression
To explore how a HFD induces high PCSK9 expression, we evaluated the expression of SREBP2 in WRL and KGN cells treated with DHT, sodium oleate or a combination thereof. Notably, although the expression of SREBP2 was significantly increased in all treated cell groups, its expression was significantly higher in cells treated with sodium oleate or sodium oleate+DHT relative to that in cells treated with DHT alone. There was no difference in SREBP2 expression between the two sodium oleate-treated groups (Fig. 7A and B ). Similar results were obtained in vivo, as significantly higher levels of SREBP2 were observed in the hepatic and ovarian tissues of mice in the HFD only and combined with DHEA treatment groups relative to those in the control group, but no significant change was observed in the DHEA group (Fig. 7C and D ). The above results indicate that a HFD induces high PCSK9 expression via SREBP2 upregulation.
Expression of SREBP2 with the treatment of different factors in vitro and in vivo. (A, B, C and D) Western blot analysis (left) and densitometry quantification (right) of SREBP2 protein in the WRL, KGN cells, livers of mice and ovaries of mice. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a and b, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance, n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Expression of SREBP2 with the treatment of different factors in vitro and in vivo. (A, B, C and D) Western blot analysis (left) and densitometry quantification (right) of SREBP2 protein in the WRL, KGN cells, livers of mice and ovaries of mice. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a and b, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance, n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Expression of SREBP2 with the treatment of different factors in vitro and in vivo. (A, B, C and D) Western blot analysis (left) and densitometry quantification (right) of SREBP2 protein in the WRL, KGN cells, livers of mice and ovaries of mice. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a and b, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance, n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Inhibiting SREBPs partly reverses increased PCSK9 expression and restores abnormal lipid metabolism and ovarian function in PCOS mice
In the third animal experiment, we used fatostain to confirm that the effect of HFD on PCSK9 expression was mediated by SREBP2. Three-week-old mice were divided into four groups and treated with a HFD alone or in combination with fatostain, HFD and DHEA or in combination with fatostain for 20 days. Using an ELISA, we showed that fatostain significantly inhibited the increase in serum PCSK9 concentration (Fig. 8A). Similarly, fatostain significantly reduced the expression of PCSK9 mRNA or protein in hepatic and ovarian tissues (Fig. 8B, C , D , E , F and G ). In addition, fatostain significantly reduced both sodium oleate induced lipid accumulation in cells and HFD-induced lipid accumulation in the livers of model mice (Fig. 9A, B and D ), and led to reductions in body weight (Fig. 9E) and significant reductions in serum TC, LDL-C and HDL-C concentrations (Fig. 9F, G and H ). These results indicate that the HFD-induced increase in PCSK9 expression in model mice was associated with SREBPs induction. Inhibition of SREBPs not only blocked the HFD-induced upregulation of PCSK9, but also improved abnormal lipid metabolism in PCOS model mice.
Effects of fatostain on PCSK9 expression in mice. (A) Serum expression PCSK9 level by the ELISA assay. (B and E) Immunohistochemical staining of livers and ovaries of the mice of different groups; Brown staining represents the positive expression of PCSK9. (C and F) The expression of Pcsk9 mRNA in the livers and ovaries. (D and G) Western blot analysis (left) and densitometry quantification (right) of PCSK9 protein in the livers and ovaries. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a and b, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Effects of fatostain on PCSK9 expression in mice. (A) Serum expression PCSK9 level by the ELISA assay. (B and E) Immunohistochemical staining of livers and ovaries of the mice of different groups; Brown staining represents the positive expression of PCSK9. (C and F) The expression of Pcsk9 mRNA in the livers and ovaries. (D and G) Western blot analysis (left) and densitometry quantification (right) of PCSK9 protein in the livers and ovaries. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a and b, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Effects of fatostain on PCSK9 expression in mice. (A) Serum expression PCSK9 level by the ELISA assay. (B and E) Immunohistochemical staining of livers and ovaries of the mice of different groups; Brown staining represents the positive expression of PCSK9. (C and F) The expression of Pcsk9 mRNA in the livers and ovaries. (D and G) Western blot analysis (left) and densitometry quantification (right) of PCSK9 protein in the livers and ovaries. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a and b, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Effects of fatostain on cell and liver lipid droplets accumulation and serum lipid metabolism. (A and B) Respresentative photomicrographs of treated WRL and KGN cells stained with Oil Red O. (C and D) Representative photomicrographs of liver section stained with H&E and Oil Red O. (E, F, G and H) Body weight, TC, LDL-C and HDL-C. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a and b, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Effects of fatostain on cell and liver lipid droplets accumulation and serum lipid metabolism. (A and B) Respresentative photomicrographs of treated WRL and KGN cells stained with Oil Red O. (C and D) Representative photomicrographs of liver section stained with H&E and Oil Red O. (E, F, G and H) Body weight, TC, LDL-C and HDL-C. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a and b, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Effects of fatostain on cell and liver lipid droplets accumulation and serum lipid metabolism. (A and B) Respresentative photomicrographs of treated WRL and KGN cells stained with Oil Red O. (C and D) Representative photomicrographs of liver section stained with H&E and Oil Red O. (E, F, G and H) Body weight, TC, LDL-C and HDL-C. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a and b, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
We also observed the ovarian function of mice above. Mice in the HFD group had a prolonged but regular estrus cycle, whereas those in the HFD+DHEA group had irregular estrous cycles. Fatostain had no effect on the estrous cycle in model mice (Fig. 10A). In terms of sex hormones, fatostain treatment led to a significant reduction in the serum concentrations of testosterone and E2 in model mice (Fig. 10B and C) but had no effect on that of P4 (Fig. 10D). In addition, mice in both the HFD and HFD+fatostain groups had relatively normal ovarian morphology with no cystic follicles, and fatostain significantly reduced the high ratios of cystic follicles and atretic follicles caused by HFD+DHEA (Fig. 10E and F). These results indicate that SREBP2 inhibition partly restored ovarian morphology and function in PCOS model mice.
Effects of fatostain on ovarian function in mice. (A) Representative estrous cycle of one mouse from each group. (B, C and D) Serum testosterone, E2 and P4. (E) Representative H&E staining of ovarian section of mice for each group. (F) Percentage of follicles at some levels. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a, b and c, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Effects of fatostain on ovarian function in mice. (A) Representative estrous cycle of one mouse from each group. (B, C and D) Serum testosterone, E2 and P4. (E) Representative H&E staining of ovarian section of mice for each group. (F) Percentage of follicles at some levels. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a, b and c, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Effects of fatostain on ovarian function in mice. (A) Representative estrous cycle of one mouse from each group. (B, C and D) Serum testosterone, E2 and P4. (E) Representative H&E staining of ovarian section of mice for each group. (F) Percentage of follicles at some levels. Data are presented as mean ± s.d.; different lower-case letters above the columns, such as a, b and c, indicate P < 0.05, and if two columns have the same lowercase letter, it indicates no statistical significance. At least n = 3 per group.
Citation: Reproduction 162, 6; 10.1530/REP-21-0164
Discussion
According to clinical studies, greater than 70% of PCOS cases exhibit abnormal lipid profiles (Rocha et al. 2011). However, the cause of abnormal lipid metabolism and its role in the pathogenesis of PCOS remain to be clarified. PCSK9 is a known negative regulator of lipid metabolism, but few articles have described its relationship with PCOS. We previously confirmed that abnormal high expression of PCSK9 contributed to the pathogenesis of PCOS by affecting lipid metabolism and ovarian function, and that a PCSK9 inhibitor partially rescued the pathological changes associated with PCOS (Wang et al. 2019). In this study, we used Pcsk9 knockout mice to further implicate PCSK9 in the occurrence and development of PCOS.
The estrous cycle of a mouse directly reflects the endocrine state of its ovaries. In our experiments, both WT and KO-PCOS mice had disordered estrous cycles. Correspondingly, we observed no difference in serum testosterone, E2 and P4 concentrations between the two groups, suggesting that Pcsk9 knockout may not significantly improve ovarian endocrine function in PCOS model mice. Polycystic ovaries are characterized by increased number of follicles in the earliest stage, which associate with hyperandrogenism (Hughesdon 1982, Maciel et al. 2004). Animal experiments have shown that androgen can increase ovulation rates due to its complex and crucial role in anovulation (Cardenas et al. 2002, Ehrmann et al. 2006). A clinical study observed that the number of follicles obtained on the day of oocyte retrieval after ovarian stimulation was much higher in women with PCOS than in those without PCOS (Kumar et al. 2013). In this study, we observed a significantly lower number of ovulations in KO-PCOS mice than in WT-PCOS mice, suggesting that knocking out Pcsk9 led to the partial reversal of excessive early follicular development.
The close correlation between lipid metabolism and ovarian functions has been reported widely (Miettinen et al. 2001, Yesilaltay et al. 2014). In this study, the significantly reduced serum TC, LDL-C and HDL-C concentrations observed in KO-PCOS mice were consistent with the results of previous studies (Cunningham et al. 2007, Zaid et al. 2008, Abifadel et al. 2009) indicating that Pcsk9 knockout can effectively improve lipid metabolism and ovarian function in PCOS model mice. In addition, the presence of adipose tissue around the ovaries and the accumulation of lipid droplets in the ovaries contribute to follicular development and ovulation (Cherian-Shaw et al. 2009). In our study, Oil Red O staining and electron microscopy revealed significant increases in the deposition of neutral lipid droplets in the ovarian tissue of KO-PCOS mice. Although this increased lipid accumulation may contribute to the increased follicular development observed in KO-PCOS mice, the specific mechanism needs to be further explored. In PCOS, abnormal follicular development is associated with granulosa cell apoptosis (Cui et al. 2017, Wu et al. 2017, Song et al. 2018), as well as systemic and local lipid metabolism. In this study, we observed a significant reduction in the level of the pro-apoptotic protein CASP3 in KO-PCOS mice, providing further experimental evidence of the correlation between PCSK9 and apoptosis.
PCSK9 in serum and obesity are positively correlated with symptom severity in PCOS women (Gambineri et al. 2002, Salehi et al. 2004). Consequently, metabolic and dietary factors that regulate PCSK9 have gained much attention (Krysa et al. 2017). Here, we aimed to explore the influence of HFD on PCSK9 expression in a DHEA-induced mouse model of PCOS. According to the literature, interventions associated with negative effects on lipid and glucose metabolism, including HFD and high-fructose diet, result in an elevated serum PCSK9 concentration (Dong et al. 2015). Dong et al. showed that hamsters fed a HFD had dyslipidemia and an elevated serum PCSK9 concentration (Dong et al. 2015), consistent with our findings. Limited and even conflicting results of DHEA on metabolic were both reported, which was once thought to be associated with the complex metabolic effects of DHEA. In our study, DHEA treatment alone had no significant influence on serum lipid concentrations, suggesting another perspective wherein DHEA treatment does not lead to an increased serum PCSK9 concentration. Interestingly, although PCSK9 downregulates LDLR expression, resulting in reduced LDL-C uptake by cells, we found that an increase in PCSK9 was always accompanied by an increase in lipid deposition in cells and tissues, consistent with the findings of previous studies (Yang et al. 2018, Chang et al. 2019). We speculate that PCSK9 can directly and independently affect intracellular lipid metabolism, leading to the deposition of lipid droplets in cells or tissues.
Our results also showed that DHEA had a great influence on the estrous cycle, follicular development and serum hormone concentrations in mice, again consistent with previous reports (Roy et al. 1962, van Houten & Visser 2014). In PCOS mice, a HFD had very little effect on indicators of ovarian function, indicating that the follicular development disorder associated with this condition was mainly related to androgen rather than dietary fat. However, clinical studies have shown that HFD-induced obesity can aggravate the pathology of PCOS in humans (Gambineri et al. 2002, Carmina et al. 2007), whereas weight loss can reduce the pathological features of this condition (Moran et al. 2009). Animal studies have shown that feeding mice a HFD with 60% fat for 16 weeks (Brothers et al. 2010) or with 40% for 26 weeks (Patel & Shah 2018) successfully led to the generation of obese mice with a PCOS-like condition. To explain the discrepancy between our and previously reported results, we suggest that it may have resulted from a too-short HFD feeding time in our study; in this case, changes in ovarian function may not have been obvious. Taken together, these results suggest that a HFD is the main cause of the observed increases in PCSK9 and abnormal lipid metabolism in PCOS model mice, whereas follicular development disorders are mainly attributable to high androgen levels.
SREBP2 transcriptionally regulates PCSK9 expression (Jia et al. 2014, Sui et al. 2018). We found that HFD was associated with increased PCSK9 levels in the serum and hepatic and ovarian tissues of model mice. These changes were also accompanied by an increase in SREBP2 expression, consistent with the experimental data from studies on other cells and animal models (Eberle et al. 2004, Jeong et al. 2008). Therefore, we speculate that a HFD may indeed induce PCSK9 expression by upregulating SREBP2, but the specific mechanism remains to be further explored.
Inhibiting SREBPs was shown to improve lipid metabolism disorders in obese OB/OB mice by promoting weight loss, reducing lipid accumulation in the liver and significantly reducing blood concentrations of TC, LDL and HDL (Gao et al. 2018). The findings in our model were mostly consistent with those earlier findings. The proximal promoter of Pcsk9 contains a functional sterol regulatory element targeted by SREBPs (Horton et al. 2003). We also observed reduced PCSK9 expression in mice in both the HFD and HFD+DHEA groups after treatment with the SREBPs inhibitor fatostain. Furthermore, the effects of fatostain on PCOS model mice were consistent with the effects of direct inhibition of PCSK9 (Wang et al. 2019). Thus, we speculate that SREBPs inhibitor could improve lipid metabolism disorders in PCOS mice at least partly by reducing PCSK9 expression.
Based on the above observations, we first hypothesized that inhibiting SREBPs would reduce the HFD-induced expression of PCSK9, and thus improving abnormal lipid metabolism in PCOS model mice. In further experiments, we observed that fatostain could reduce HFD- or sodium oleate-induced PCSK9 expression in vivo and in vitro, respectively. SREBPs are important regulators of intracellular lipid and cholesterol homeostasis and affect the expression of several genes in addition to Pcsk9, including Ldlr and Hmgcr(Horton et al. 2003), the protein products of which are relevant to cholesterol management and steroidogenesis in ovarian cells. Cholesterol is essential for ovarian function; it is the precursor of all steroid hormones and provides material for follicular growth (Azhar et al. 2003). In our research, we indeed found that fatostain affected serum testosterone and E2 concentrations but cannot exclude the potential effects of SREBPs on factors other than PCSK9. However, we mainly focused on PCSK9, and based on the literature and our observations in this study, we conclude that fatostain mitigates increased serum testosterone and E2 concentrations partly by reducing SREBP2 action and PCSK9 expression. In turn, we speculate that a HFD induces SREBP2 expression, which further increases PCSK9 expression to affect cholesterol metabolism and, ultimately, ovarian hormones. SREBPs targeting drugs intended to improve lipid metabolism have been shown not only to reduce cholesterol synthesis in cells but also to improve the reproductive and metabolic disorders associated with PCOS (Banaszewska et al. 2007, Sathyapalan et al. 2009), indirectly proving a link between abnormal lipid metabolism and ovarian function. However, the pathogenic factors affecting PCOS are complex; although the increased expression of PCSK9, which is induced by a HFD and mediated via SREBP2, is involved in the pathology of PCOS, this protein cannot induce all of the pathological features of PCOS. This may explain why we found that HFD was related to abnormal lipid metabolism but not to abnormal reproductive function in our PCOS model mice.
In conclusion, we provide evidence that a HFD is an important pathogenic factor influencing PCOS. Specifically, a HFD induces abnormal PCSK9 expression, leading to abnormal lipid metabolism and disordered ovarian follicular development partly by upregulating SREBP2. Our findings may be helpful for the prevention and treatment of PCOS.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This work did not receive any specific grant from any funding agency in the public, commercial, or not-for-profit sector.
Acknowledgements
The authors gratefully acknowledge support from the National Natural Science Foundation of China (No. 81771542, URL: http://www.nsfc.gov.cn/).
References
Abifadel M, Rabes JP, Devillers M, Munnich A, Erlich D, Junien C, Varret M & Boileau C 2009 Mutations and polymorphisms in the proprotein convertase subtilisin kexin 9 (PCSK9) gene in cholesterol metabolism and disease. Human Mutation 30 520–529. (https://doi.org/10.1002/humu.20882)
Azhar S, Leers-Sucheta S & Reaven E 2003 Cholesterol uptake in adrenal and gonadal tissues: the SR-BI and ‘selective’ pathway connection. Frontiers in Bioscience 8 s998–s1029. (https://doi.org/10.2741/1165)
Banaszewska B, Pawelczyk L, Spaczynski RZ, Dziura J & Duleba AJ 2007 Effects of simvastatin and oral contraceptive agent on polycystic ovary syndrome: prospective, randomized, crossover trial. Journal of Clinical Endocrinology and Metabolism 92 456–461. (https://doi.org/10.1210/jc.2006-1988)
Bingham B, Shen R, Kotnis S, Lo CF, Ozenberger BA, Ghosh N, Kennedy JD, Jacobsen JS, Grenier JM & Distefano PS et al.2006 Proapoptotic effects of NARC 1 (= PCSK9), the gene encoding a novel serine proteinase. Cytometry: Part A 69 1123–1131. (https://doi.org/10.1002/cyto.a.20346)
Brothers KJ, Wu S, Divall SA, Messmer MR, Kahn CR, Miller RS, Radovick S, Wondisford FE & Wolfe A 2010 Rescue of obesity-induced infertility in female mice due to a pituitary-specific knockout of the insulin receptor. Cell Metabolism 12 295–305. (https://doi.org/10.1016/j.cmet.2010.06.010)
Caldwell AS, Middleton LJ, Jimenez M, Desai R, Mcmahon AC, Allan CM, Handelsman DJ & Walters KA 2014 Characterization of reproductive, metabolic, and endocrine features of polycystic ovary syndrome in female hyperandrogenic mouse models. Endocrinology 155 3146–3159. (https://doi.org/10.1210/en.2014-1196)
Cardenas H, Herrick JR & Pope WF 2002 Increased ovulation rate in gilts treated with dihydrotestosterone. Reproduction 123 527–533. (https://doi.org/10.1530/rep.0.1230527)
Carmina E, Bucchieri S, Esposito A, Del Puente A, Mansueto P, Orio F, Di Fede G & Rini G 2007 Abdominal fat quantity and distribution in women with polycystic ovary syndrome and extent of its relation to insulin resistance. Journal of Clinical Endocrinology and Metabolism 92 2500–2505. (https://doi.org/10.1210/jc.2006-2725)
Chang HY, Wu JR, Gao WY, Lin HR, Chen PY, Chen CI, Wu MJ & Yen JH 2019 The cholesterol-modulating effect of methanol extract of pigeon pea (Cajanus cajan (L.) Millsp.) leaves on regulating LDLR and PCSK9 expression in HepG2 cells. Molecules 24 493. (https://doi.org/10.3390/molecules24030493)
Cherian-Shaw M, Puttabyatappa M, Greason E, Rodriguez A, Vandevoort CA & Chaffin CL 2009 Expression of scavenger receptor-BI and low-density lipoprotein receptor and differential use of lipoproteins to support early steroidogenesis in luteinizing macaque granulosa cells. Endocrinology 150 957–965. (https://doi.org/10.1210/en.2008-0619)
Conway G, Dewailly D, Diamanti-Kandarakis E, Escobar-Morreale HF, Franks S, Gambineri A, Kelestimur F, Macut D, Micic D & Pasquali R et al.2014 The polycystic ovary syndrome: a position statement from the European Society of Endocrinology. European Journal of Endocrinology 171 P1–P29. (https://doi.org/10.1530/EJE-14-0253)
Cui X, Jing X, Wu X, Bi X, Liu J, Long Z, Zhang X, Zhang D, Jia H & Su D et al.2017 Abnormal expression levels of BMP15/Smad1 are associated with granulosa cell apoptosis in patients with polycystic ovary syndrome. Molecular Medicine Reports 16 8231–8236. (https://doi.org/10.3892/mmr.2017.7658)
Cunningham D, Danley DE, Geoghegan KF, Griffor MC, Hawkins JL, Subashi TA, Varghese AH, Ammirati MJ, Culp JS & Hoth LR et al.2007 Structural and biophysical studies of PCSK9 and its mutants linked to familial hypercholesterolemia. Nature Structural and Molecular Biology 14 413–419. (https://doi.org/10.1038/nsmb1235)
de Melo AS, Dias SV, Cavalli Rde C, Cardoso VC, Bettiol H, Barbieri MA, Ferriani RA & Vieira CS 2015 Pathogenesis of polycystic ovary syndrome: multifactorial assessment from the foetal stage to menopause. Reproduction 150 R11–R24. (https://doi.org/10.1530/REP-14-0499)
Dokras A, Sarwer DB, Allison KC, Milman L, Kris-Etherton PM, Kunselman AR, Stetter CM, Williams NI, Gnatuk CL & Estes SJ et al.2016 Weight loss and lowering androgens predict improvements in health-related quality of life in women with PCOS. Journal of Clinical Endocrinology and Metabolism 101 2966–2974. (https://doi.org/10.1210/jc.2016-1896)
Dong B, Singh AB, Azhar S, Seidah NG & Liu J 2015 High-fructose feeding promotes accelerated degradation of hepatic LDL receptor and hypercholesterolemia in hamsters via elevated circulating PCSK9 levels. Atherosclerosis 239 364–374. (https://doi.org/10.1016/j.atherosclerosis.2015.01.013)
Eberle D, Hegarty B, Bossard P, Ferre P & Foufelle F 2004 SREBP transcription factors: master regulators of lipid homeostasis. Biochimie 86 839–848. (https://doi.org/10.1016/j.biochi.2004.09.018)
Ehrmann DA, Liljenquist DR, Kasza K, Azziz R, Legro RS, Ghazzi MN & PCOS/Troglitazone Study Group 2006 Prevalence and predictors of the metabolic syndrome in women with polycystic ovary syndrome. Journal of Clinical Endocrinology and Metabolism 91 48–53. (https://doi.org/10.1210/jc.2005-1329)
Folsom AR, Peacock JM, Boerwinkle E & Atherosclerosis Risk in Communities (ARIC) Study Investigators 2009 Variation in PCSK9, low LDL cholesterol, and risk of peripheral arterial disease. Atherosclerosis 202 211–215. (https://doi.org/10.1016/j.atherosclerosis.2008.03.009)
Franks S 2008 Polycystic ovary syndrome in adolescents. International Journal of Obesity 32 1035–1041. (https://doi.org/10.1038/ijo.2008.61)
Gambineri A, Pelusi C, Vicennati V, Pagotto U & Pasquali R 2002 Obesity and the polycystic ovary syndrome. International Journal of Obesity and Related Metabolic Disorders 26 883–896. (https://doi.org/10.1038/sj.ijo.0801994)
Gao S, Shi Z, Li X, Li W, Wang Y, Liu Z & Jiang J 2018 Fatostatin suppresses growth and enhances apoptosis by blocking SREBP-regulated metabolic pathways in endometrial carcinoma. Oncology Reports 39 1919–1929. (https://doi.org/10.3892/or.2018.6265)
Horton JD, Shah NA, Warrington JA, Anderson NN, Park SW, Brown MS & Goldstein JL 2003 Combined analysis of oligonucleotide microarray data from transgenic and knockout mice identifies direct SREBP target genes. PNAS 100 12027–12032. (https://doi.org/10.1073/pnas.1534923100)
Hughesdon PE 1982 Morphology and morphogenesis of the Stein-Leventhal ovary and of so-called ‘hyperthecosis’. Obstetrical and Gynecological Survey 37 59–77. (https://doi.org/10.1097/00006254-198202000-00001)
Jeong HJ, Lee HS, Kim KS, Kim YK, Yoon D & Park SW 2008 Sterol-dependent regulation of proprotein convertase subtilisin/kexin type 9 expression by sterol-regulatory element binding protein-2. Journal of Lipid Research 49 399–409. (https://doi.org/10.1194/jlr.M700443-JLR200)
Jia YJ, Xu RX, Sun J, Tang Y & Li JJ 2014 Enhanced circulating PCSK9 concentration by berberine through SREBP-2 pathway in high fat diet-fed rats. Journal of Translational Medicine 12 103. (https://doi.org/10.1186/1479-5876-12-103)
Krysa JA, Ooi TC, Proctor SD & Vine DF 2017 Nutritional and lipid modulation of PCSK9: effects on cardiometabolic risk factors. Journal of Nutrition 147 473–481. (https://doi.org/10.3945/jn.116.235069)
Kumar P, Nawani N, Malhotra N, Malhotra J, Patil M, Jayakrishnan K, Kar S, Jirge PR & Mahajan N 2013 Assisted reproduction in polycystic ovarian disease: a multicentric trial in India. Journal of Human Reproductive Sciences 6 49–53. (https://doi.org/10.4103/0974-1208.112382)
Kysenius K, Muggalla P, Matlik K, Arumae U & Huttunen HJ 2012 PCSK9 regulates neuronal apoptosis by adjusting ApoER2 levels and signaling. Cellular and Molecular Life Sciences 69 1903–1916. (https://doi.org/10.1007/s00018-012-0977-6)
Lai H, Jia X, Yu Q, Zhang C, Qiao J, Guan Y & Kang J 2014 High-fat diet induces significant metabolic disorders in a mouse model of polycystic ovary syndrome. Biology of Reproduction 91 127. (https://doi.org/10.1095/biolreprod.114.120063)
Legro RS, Dodson WC, Kunselman AR, Stetter CM, Kris-Etherton PM, Williams NI, Gnatuk CL, Estes SJ, Allison KC & Sarwer DB et al.2016 Benefit of delayed fertility therapy with preconception weight loss over immediate therapy in obese women with PCOS. Journal of Clinical Endocrinology and Metabolism 101 2658–2666. (https://doi.org/10.1210/jc.2016-1659)
Luo LL, Huang J, Fu YC, Xu JJ & Qian YS 2008 Effects of tea polyphenols on ovarian development in rats. Journal of Endocrinological Investigation 31 1110–1118. (https://doi.org/10.1007/BF03345661)
Maciel GA, Baracat EC, Benda JA, Markham SM, Hensinger K, Chang RJ & Erickson GF 2004 Stockpiling of transitional and classic primary follicles in ovaries of women with polycystic ovary syndrome. Journal of Clinical Endocrinology and Metabolism 89 5321–5327. (https://doi.org/10.1210/jc.2004-0643)
Marcondes FK, Bianchi FJ & Tanno AP 2002 Determination of the estrous cycle phases of rats: some helpful considerations. Brazilian Journal of Biology 62 609–614. (https://doi.org/10.1590/s1519-69842002000400008)
McCartney CR, Blank SK, Prendergast KA, Chhabra S, Eagleson CA, Helm KD, Yoo R, Chang RJ, Foster CM & Caprio S et al.2007 Obesity and sex steroid changes across puberty: evidence for marked hyperandrogenemia in pre- and early pubertal obese girls. Journal of Clinical Endocrinology and Metabolism 92 430–436. (https://doi.org/10.1210/jc.2006-2002)
Miettinen HE, Rayburn H & Krieger M 2001 Abnormal lipoprotein metabolism and reversible female infertility in HDL receptor (SR-BI)-deficient mice. Journal of Clinical Investigation 108 1717–1722. (https://doi.org/10.1172/JCI13288)
Moran LJ, Pasquali R, Teede HJ, Hoeger KM & norman RJ 2009 Treatment of obesity in polycystic ovary syndrome: a position statement of the Androgen Excess and Polycystic Ovary Syndrome Society. Fertility and Sterility 92 1966–1982. (https://doi.org/10.1016/j.fertnstert.2008.09.018)
R & Shah G 2018 High-fat diet exposure from pre-pubertal age induces polycystic ovary syndrome (PCOS) in rats. Reproduction 155 141–151. (https://doi.org/10.1530/REP-17-0584)
Piao MX, Bai JW, Zhang PF & Zhang YZ 2015 PCSK9 regulates apoptosis in human neuroglioma u251 cells via mitochondrial signaling pathways. International Journal of Clinical and Experimental Pathology 8 2787–2794.
Rocha MP, Marcondes JA, Barcellos CR, Hayashida SA, Curi DD, Da Fonseca ÂM, Bagnoli VR & Baracat EC 2011 Dyslipidemia in women with polycystic ovary syndrome: incidence, pattern and predictors. Gynecological Endocrinology 27 814–819. (https://doi.org/10.3109/09513590.2010.508852)
Roy S, Mahesh VB & Greenblatt RB 1962 Effect of dehydroepiandrosterone and delta4-androstenedione on the reproductive organs of female rats: production of cystic changes in the ovary. Nature 196 42–43. (https://doi.org/10.1038/196042a0)
Salehi M, Bravo-Vera R, Sheikh A, Gouller A & Poretsky L 2004 Pathogenesis of polycystic ovary syndrome: what is the role of obesity? Metabolism: Clinical and Experimental 53 358–376. (https://doi.org/10.1016/j.metabol.2003.10.005)
Sathyapalan T, Kilpatrick ES, Coady AM & Atkin SL 2009 The effect of atorvastatin in patients with polycystic ovary syndrome: a randomized double-blind placebo-controlled study. Journal of Clinical Endocrinology and Metabolism 94 103–108. (https://doi.org/10.1210/jc.2008-1750)
Seidah NG, Benjannet S, Wickham L, Marcinkiewicz J, Jasmin SB, Stifani S, Basak A, Prat A & Chretien M 2003 The secretory proprotein convertase neural apoptosis-regulated convertase 1 (NARC-1): liver regeneration and neuronal differentiation. PNAS 100 928–933. (https://doi.org/10.1073/pnas.0335507100)
Song WJ, Shi X, Zhang J, Chen L, Fu SX & Ding YL 2018 Akt-mTOR signaling mediates abnormalities in the proliferation and apoptosis of ovarian granulosa cells in patients with polycystic ovary syndrome. Gynecologic and Obstetric Investigation 83 124–132. (https://doi.org/10.1159/000464351)
Steckler T, Manikkam M, Inskeep EK & Padmanabhan V 2007 Developmental programming: follicular persistence in prenatal testosterone-treated sheep is not programmed by androgenic actions of testosterone. Endocrinology 148 3532–3540. (https://doi.org/10.1210/en.2007-0339)
Sui GG, Xiao HB, Lu XY & Sun ZL 2018 Naringin activates AMPK resulting in altered expression of SREBPs, PCSK9, and LDLR to reduce body weight in obese C57BL/6J mice. Journal of Agricultural and Food Chemistry 66 8983–8990. (https://doi.org/10.1021/acs.jafc.8b02696)
Tavori H, Rashid S & Fazio S 2015 On the function and homeostasis of PCSK9: reciprocal interaction with LDLR and additional lipid effects. Atherosclerosis 238 264–270. (https://doi.org/10.1016/j.atherosclerosis.2014.12.017)
van Houten EL & Visser JA 2014 Mouse models to study polycystic ovary syndrome: a possible link between metabolism and ovarian function? Reproductive Biology 14 32–43. (https://doi.org/10.1016/j.repbio.2013.09.007)
Vendola K, Zhou J, Wang J, Famuyiwa OA, Bievre M & Bondy CA 1999 Androgens promote oocyte insulin-like growth factor I expression and initiation of follicle development in the primate ovary. Biology of Reproduction 61 353–357. (https://doi.org/10.1095/biolreprod61.2.353)
Wang M, Zhao D, Xu L, Guo W, Nie L, Lei Y, Long Y, Liu M, Wang Y & Zhang X et al.2019 Role of PCSK9 in lipid metabolic disorders and ovarian dysfunction in polycystic ovary syndrome. Metabolism: Clinical and Experimental 94 47–58. (https://doi.org/10.1016/j.metabol.2019.02.002)
Wu CY, Tang ZH, Jiang L, Li XF, Jiang ZS & Liu LS 2012 PCSK9 siRNA inhibits HUVEC apoptosis induced by ox-LDL via Bcl/Bax-caspase9-caspase3 pathway. Molecular and Cellular Biochemistry 359 347–358. (https://doi.org/10.1007/s11010-011-1028-6)
Wu XQ, Wang YQ, Xu SM, Liu JF, Bi XY, Wang ZQ & Zhang JP 2017 The WNT/beta-catenin signaling pathway may be involved in granulosa cell apoptosis from patients with PCOS in North China. Journal of Gynecology Obstetrics and Human Reproduction 46 93–99. (https://doi.org/10.1016/j.jgyn.2015.08.013)
Yang SH, Xu RX, Cui CJ, Wang Y, Du Y, Chen ZG, Yao YH, Ma CY, Zhu CG & Guo YL et al.2018 Liraglutide downregulates hepatic LDL receptor and PCSK9 expression in HepG2 cells and db/db mice through a HNF-1a dependent mechanism. Cardiovascular Diabetology 17 48. (https://doi.org/10.1186/s12933-018-0689-9)
Yesilaltay A, Dokshin GA, Busso D, Wang L, Galiani D, Chavarria T, Vasile E, Quilaqueo L, Orellana JA & Walzer D et al.2014 Excess cholesterol induces mouse egg activation and may cause female infertility. PNAS 111 E4972–E4980. (https://doi.org/10.1073/pnas.1418954111)
Zaid A, Roubtsova A, Essalmani R, Marcinkiewicz J, Chamberland A, Hamelin J, Tremblay M, Jacques H, Jin W & Davignon J et al.2008 Proprotein convertase subtilisin/kexin type 9 (PCSK9): hepatocyte-specific low-density lipoprotein receptor degradation and critical role in mouse liver regeneration. Hepatology 48 646–654. (https://doi.org/10.1002/hep.22354)
