Kisspeptin mediates the impact of chronic psychological stress on reproductive and metabolic dysregulation in polycystic ovary syndrome: evidence from human and rat models Authors: Tingting Mo, Yue Qiu, Shuangshuang He, Guibin Wang, Bingbing Li, Qian Shen, Xiaowei Nie, Jiali Wei, Nachuan Li, Bo Pang, Yanfeng Liu, Xue Pan Published: 2025 ABSTRACT AbstractPolycystic ovary syndrome (PCOS) and mental health disorders have high rates of co-occurrence. Although the precise pathophysiological mechanisms remain unknown, the observed changes in those conditions may be modulated by kisspeptin (Kiss1), a protein that regulates energy metabolism. AimThe aim of this study was to investigate whether Kiss1 plays an important role in linking dysfunctional reproductive processes and abnormalities in energy metabolism in PCOS, both in humans and using a rat model of PCOS combined with chronic unpredictable mind stress (CUMS).MethodsIn 27 women with PCOS and 11 controls, fasting insulin (INS), blood glucose, Kiss1, and monoamine neurotransmitter levels were measured. For the animal experiments, the 16 rats were divided into control, PCOS, CUMS, and PCOS + CUMS groups to assess changes in behaviour, reproductive endocrine function, glucose metabolism, and granulosa cell apoptosis, as well as alterations in the expression levels of Kiss1, Kiss1 receptor (Kiss1R), and gonadotropin-releasing hormone (GnRH) in hypothalamic tissues and changes in INS-like growth factor 1 (IGF-1) and IGF-binding protein 1 (IGFBP-1) levels in ovarian tissues.ResultsThe body mass index, fasting INS levels, homeostatic model assessment of insulin resistance (HOMA-IR) index, and serum Kiss1 levels were significantly elevated in women with PCOS compared to the values in the healthy controls, whereas serum monoamine neurotransmitters levels were lower. Kiss1 levels negatively correlated with monoamine neurotransmitter concentrations and positively correlated with both INS levels and the HOMA-IR index. In the PCOS+CUMS model rats, reproductive endocrine dysfunction, abnormal glucose metabolism, and depressive-like behaviours were observed; these changes were accompanied by upregulated expression of Kiss1 and gonadotropin-releasing hormone (GnRH), downregulated IGF-1 levels, and increased ovarian granulosa cell apoptosis. ConclusionsCUMS exposure reduces monoamine neurotransmitter levels in PCOS, weakening Kiss1’s inhibitory effects., while regulating ovarian IGF-1 expression may promote granulosa cell apoptosis and reduce INS sensitivity, aggravating reproductive endocrine disorders and abnormal glucose metabolism in patients with PCOS-related psychological stress.Supplementary InformationThe online version contains supplementary material available at 10.1186/s13048-025-01918-6. Introduction Polycystic ovary syndrome (PCOS) is a common endocrine disorder that affects women of reproductive age [49], sometimes leading to infertility while negatively impacting metabolic health and increasing the risk of pregnancy-related complications [1]. Individuals with PCOS who are obese or have elevated free testosterone (FT) levels are also more susceptible to developing type 2 diabetes, as well as cardiovascular and cerebrovascular diseases [16, 76]. Owing to its diverse clinical manifestations, the diagnostic criteria of PCOS vary globally [22, 39, 43], affecting the reported prevalence. For example, according to the diagnostic criteria of the National Institutes of Health in the United States, the Rotterdam consensus, and the Androgen Excess and PCOS Society, the prevalence has been reported to be 6%, 10%, and 10%, respectively [6, 43]. Although the prevalence of PCOS is relatively high, its exact pathophysiology is not well understood. The clinical manifestations of PCOS include severe psychological distress and negative emotional modulation, and it is likely to be associated with mental health disorders [7, 11, 12, 15, 17, 74, 75], with many patients experiencing body image problems due to acne and hair loss,this psychological distress is exacerbated by the fear of infertility, diabetes, high blood pressure, coronary heart disease, and associated feelings of anxiety. A previous study showed that women with PCOS exhibited a higher prevalence of depression and anxiety-like symptoms compared with the rates in healthy controls, even after matching for various factors, such as the body mass index (BMI), the presence of infertility, and sociodemographic variables [15]. Another study reported that women with PCOS and comorbid obesity were more likely than women with PCOS without obesity to experience depression and anxiety-like symptoms [35]. In addition to these common symptoms, several studies have reported an elevated likelihood of eating disorders, attention-deficit/hyperactivity disorder, and obsessive–compulsive disorder in women with PCOS [7, 19, 30, 60]. Despite the increased risk of mental health disorders in such women, the underlying aetiology remains unclear. Moreover, a small number of studies have reported that the rates of PCOS have increased significantly in patients with mental health disorders, such as autism spectrum disorder, bulimia nervosa, and post-traumatic stress disorder [19, 75]. Collectively, these findings suggest that PCOS and mental health disorders are likely to co-occur,therefore, it is important to further explore the underlying mechanisms. Dysregulation of the hypothalamic–pituitary–ovarian (HPO) axis has been identified as an important mechanism driving the hormonal disorders observed in PCOS [37]. More specifically, the increased frequency and amplitude of gonadotropin-releasing hormone (GnRH) pulses can induce luteinizing hormone (LH) synthesis and secretion, with high LH levels promoting androgen synthesis in ovarian membrane cells, leading to hyperandrogenaemia [37]. Kiss1 is an upstream regulator of gonadotropin-releasing GnRH that primarily acts on hypothalamic GnRH neurons to stimulate GnRH secretion and thereby modulate the HPO axis. Early experimental studies demonstrated that the administration of a GnRH antagonist could block the Kiss1-induced GnRH release [23]. Insulin (INS) resistance (IR) is another common concern in patients with PCOS. In a study that used a hyperinsulinaemic euglycaemic clamp, the gold standard method for measuring INS action, INS dysfunction was observed in 74.9% of women with the disease [65]. These individuals have been shown to have 27% poorer INS sensitivity than otherwise healthy controls, independent of their BMI [9]. IR plays an important role in the pathogenesis of PCOS [57], as it can induce endocrine abnormalities, leading to excessively elevated androgen levels, which, in turn, lead to impaired INS function, creating a vicious cycle [42]. Like PCOS, IR is also strongly associated with the occurrence of mental health disorders, with some studies suggesting that individuals with metabolic disorders have an increased risk of developing depression [8, 63]. In a study involving women with PCOS, INS-resistant individuals who had an elevated homeostatic model assessment of insulin resistance (HOMA-IR) index were more than twice as likely to experience depression that those who did not [29]. Kiss1 also plays a regulatory role in glucose metabolism. In vitro studies have shown that elevated levels of Kiss1 can enhance glucose-stimulated insulin secretion, and under high-glucose conditions, kisspeptin-10 and −13 directly activate pancreatic β-cells to promote insulin release [4]. In women with PCOS, the correlations between serum Kiss1 levels and metabolic parameters such as body weight, BMI, and blood glucose remain inconclusive. Given the unclear nature of these relationships, the aim of this study was to investigate the link between mental health problems and IR in patients with PCOS. Psychological stress is known to induce adverse effects that lead to reproductive disorders in patients with the disease; however, it is unknown if such changes are related to metabolic abnormalities. A small-scale, preliminary clinical investigation was conducted to quantify levels of serum monoamine neurotransmitters, which are closely associated with the occurrence of psychological disorders; in addition, fasting INS and blood glucose levels were assessed, as was the HOMA-IR index. The effects of Kiss1 on the relationship between these factors was also evaluated. Kiss1 expression is related to the function of the hypothalamic–pituitary–adrenal (HPA) and HPO axes, serving as a bridge between the regulation of emotional behaviour and reproductive processes and playing a role in the modulation of energy metabolism. Based on the clinical findings, animal experiments were also conducted. CUMS model exposes animals to long-term, mild, and unpredictable stressors that mimic the daily psychological challenges experienced by humans. This paradigm can induce depression-like symptoms, including anhedonia and anxiety, as well as monoaminergic dysfunction, HPA axis disturbances, and structural alterations in the brain [56]. This expeiments used PCOS and CUMS models to further investigate the underlying mechanisms driving the interaction between mental health problems and IR in PCOS. Materials and methods Human subjects The present study included 27 women with PCOS and 11 healthy age- and sex-matched controls. The diagnosis of PCOS was based on the diagnostic criteria of the Rotterdam consensus, which included the presence of two or more of the following conditions: clinical and/or biochemical hyperandrogenism, menstrual disorders, and polycystic ovaries [40], excluding other hyperandrogenic causes. The inclusion criteria were an age of 19–40 years and a willingness to cooperate with the investigators. The exclusion criteria were as follows: (1) similar symptoms resulting from diseases other than PCOS, such as ovarian functional decline and hyperprolactinaemia; (2) the presence of various malignancies, severe diseases of other systems, and mental disorders; (3) unwillingness to cooperate with the investigators; and (4) simultaneous participation in other clinical studies that would have affected the present results. This study was approved by the Ethics Committee of Dongzhimen Hospital, Beijing University of Chinese Traditional Medicine (ethical approval number: 2020DZMEC-141–01) and was performed in accordance with the tenets of the Declaration of Helsinki. All participants signed informed consent documents. All participants fasted for more than 8 h prior to morning venous blood collection. Blood samples were left at room temperature for 2 h, after which they underwent centrifugation at 3,000 RPM for 15 min to isolate the supernatant; this was followed by freezing at −80 °C until later testing. Animals Female Sprague–Dawley rats (specific-pathogen-free, aged 6 weeks, weighing 180–200 g) were purchased from Beijing Hfk Bioscience Co., Ltd. (Beijing, China; animal license: SCXK (jing) 2019–0008). The rats were group-housed in the animal laboratory of Beijing University of Chinese Medicine in an appropriate environment. The study was approved by the Institutional Animal Care and Use Committee of Beijing University of Chinese Medicine (No. BUCM-2023032103–1160), and all procedures conformed with the ARRIVE guidelines. All rats underwent vaginal smear examinations to assess the status of a regular oestrous cycle. Sixteen rats with a regular oestrous were selected after observing the status over two consecutive cycles. The 16 rats were divided into four groups: a control group (CON, n = 4), a PCOS model group (PCOS, n = 4), a CUMS model group (CUMS, n = 4), and a combined PCOS and CUMS model group (PCOS + CUMS, n = 4). Drug administration and model establishment Letrozole, an aromatase inhibitor, was purchased in powder form from Hisun Pharmaceutical Co., Ltd. (Zhejiang, China; catalogue number H20133109), dissolved in 1% carboxymethylcellulose-Na (0.125 mg/mL), and stored at 4 °C. Rats in the PCOS and PCOS + CUMS groups were administered 1 mg/kg letrozole solution via oral gavage once daily for 42 days, whereas the CON group received an equivalent volume of carboxymethylcellulose-Na solution. During the 6-week administration period, the CUMS and PCOS + CUMS groups were exposed to stressors in the CUMS paradigm. CUMS paradigm Rats in the CUMS and PCOS + CUMS groups were housed in isolation and exposed to a CUMS paradigm that included the following nine different stressors: (1) water or food deprivation for 24 h; (2) tail clamping for 1 min; (3) exposure to a wet cage for 24 h; (4) forced swimming in cold water (0–4 °C) for 5 min; (5) 45-degree cage tilting for 24 h; (6) restricted movement for 3 h; (7) inversion of the light/dark cycle for 24 h; (8) heat exposure at 45 °C for 5 min; and (9) exposure to noise exceeding 85 dB in amplitude for 30 min. Rats were randomly exposed to one stressor every day for six weeks, and the same stressor was never applied consecutively. Rats in the CON and PCOS groups were pair-housed and left undisturbed, except for necessary procedures. Glucose tolerance test (GTT) and insulin tolerance test (ITT) After the models were established, all animals were subjected to a GTT. After fasting for 16 h, the rats were transferred to a quiet procedure room and permitted to acclimatise for 30 min. A 20% glucose solution was prepared and administered by oral gavage (2 g/kg). Blood glucose levels were measured before gavage and 30, 60, 90, and 120 min after. All animals underwent 6 h of fasting before an ITT was conducted. INS was dissolved in 0.9% sodium chloride solution to a concentration of 0.1 U/mL, which was injected intraperitoneally (0.5 U/kg). Blood glucose levels were measured before injection and 30, 60, 90, and 120 min after. Blood samples were collected from a tail scratch, and an Accu-Check glucometer (Roche, Basel, Switzerland) was used for quantification. Behavioural testing After the GTT and ITT were completed, the sucrose preference test (SPT) was used to assess the loss of interest in a rewarding stimulus, and anxiety-like and exploratory behaviours were measured using an open field test (OFT) with an automated video-tracking system (VisuTrack; Xinruan, Shanghai, China). SPT Before the SPT was conducted, the rats were placed in separate cages for training sessions. During training, two bottles of 1% sucrose solution were placed in each cage for 24 h, after which one bottle of sucrose solution and one bottle of distilled water were placed in each cage for 24 h. At some point during the second 24-h period, the two bottles were switched. Before testing, the rats were deprived of food and water for 24 h. During testing, the rats were alone in a cage and had free access to both the sucrose solution and distilled water for 1 h; the consumed volume of each was recorded to calculate the sucrose preference as follows:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} &Sucrose preference value(\%)\\&=\left(\frac{sucrose solution intake volume}{sucrose solution intake volume+water intake volume}\right)\times 100\% \end{aligned}$$\end{document} OFT The OFT was conducted in a dark and quiet laboratory to quantify the exploratory behaviour of the rats in an unfamiliar environment. The testing apparatus was a large box (88 cm × 88 cm × 50 cm) with a black floor and four black walls, with a camera mounted above the centre of the box to record the rats’ activities and track their movement. The rats were acclimatised for 30 min prior to entry in the testing room. During testing, each rat was allowed to explore freely in the box for 5 min. After each session, 75% ethanol solution was used to remove urine and faeces from the box. VisuTrack software was used to record and analyse the total distance travelled, the rearing frequency, the number of crossings between regions, and the duration of time spent in the central area. Enzyme-linked immunosorbent assay (ELISA) After the behavioural testing was completed, the rats were anaesthetized using 2% sodium pentobarbital (0.2 mL/100 g, administered intraperitoneally), and aortic blood samples were collected. After centrifugation, serum samples were collected, frozen, and stored at −80 °C. Serum concentrations of LH, follicle-stimulating hormone (FSH), FT, estradiol (E2), and INS were quantified using ELISA kits (ml470613, 96 T, MlBio, Shanghai; ml403115, 96 T, MlBio, Shanghai; ml002865, 96 T, MlBio, Shanghai; ml002871, 96 T, MlBio, Shanghai; and ml302840, 96 T, MlBio, Shanghai, respectively), (The detection limits: 1.5 mIU/mL for LH, 12 IU/L for FSH, 12 ng/mL for FT, 2 nmol/mL for E2 and 1400uIU/L for INS). In the healthy controls and women with PCOS, fasting INS, Kiss1, norepinephrine (NE), dopamine (DA), and 5-hydroxytryptamine/serotonin (5-HT) levels were also measured using ELISA kits (ml064302, MlBio, Shanghai, China; ml060071, MlBio, Shanghai, China; ml024648, MlBio, Shanghai, China; ml077133, MlBio, Shanghai, China; and ml057425, MlBio, Shanghai, China, respectively), (The detection limits: 1400uIU/L for INS, 12 IU/L for Kiss1 400 pg/mL for NE, 120 nmol/L for DA and 2400 pg/mL for 5-HT). Fasting blood glucose levels were measured using a blood glucose test kit (mlsh0199; MlBio, Shanghai, China) based on the glucose oxidase–peroxidase method. The HOMA-IR index was calculated as follows [54]:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} &HOMA-IR index\\&=\left[fasting blood glucose concentration\left(mmol/L\right) \right. \\& \left. \times fasting insulin concentration\left(mU/L\right)\right]/22.5\end{aligned}$$\end{document} Liquid chromatography-tandem mass spectrometry (LC–MS/MS) Immediately after the blood samples were collected, the rats were decapitated, and dissections were performed on ice. Hippocampal tissue was separated from one of the hemispheres of each of the brains, frozen in liquid nitrogen, and stored at −80 °C. The analysis was performed using an AB SCIEX Triple Quad 7500 mass spectrometer (AB Sciex Pte. Ltd., USA) in selective reaction monitor and positive ionization mode. An ACQUITY HSS PFP column (2.1 × 100 mm, 1.7 μm, Waters, USA) was used, equipped with an infinite ExionLC analysis system binary pump. Solvent A was acetonitrile containing 0.1% formic acid and 5 mM amine acetate, and solvent B was water containing 0.1% formic acid and 5 mM amine acetate. The flow rate was 0.2 mL/min. At 0 min, the proportion of solvent B was 98%; this was reduced to 0% at 8 min and returned to 98% after 2 min, with one collection cycle lasting 12 min. The column temperature was 40 ℃. The sample was kept at 8 ℃, and the injection volume was 2 µL. The temperature of the ion source for electrospray ionization was maintained at 500 ℃. The pressure of the air curtain was set to 40 psi. The collision-activated dissociated gas setting was set to 9. The ion spray voltage was 1,500 V. The pressure levels of ion gas 1 and ion gas 2 were set to 40 and 70 psi, respectively. All data were analysed using SCIEX OS 3.1.6. Software. Haematoxylin and eosin (H&E) histological staining The long and short diameters of the ovarian tissues, and the masses of the ovarian and uterine tissues were measured. The ovarian index and volume were calculated according to the following formulae:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Ovarian index\left(\%\right)=\frac{ovarian mass}{body mass}\times 100\%$$\end{document}\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} Ovarian volume\left({cm}^{3}\right)=&\left(\raisebox{1ex}{$\pi $}\!\left/ \!\raisebox{-1ex}{$6$}\right.\right)\times long diameter\left(cm\right)\\&\times {short diameter}^{2}\left({cm}^{2}\right) \end{aligned}$$\end{document} After sample measurements were taken, the ovarian tissues were fixed in 4% paraformaldehyde for 48 h, embedded in paraffin, and sectioned using a microtome at a thickness of 4 µm. After deparaffinization, sections were stained and observed using an Aperio VERSA digital pathology scanner (Leica Biosystems, Shanghai, China). Terminal deoxyribonucleotidyl transferase-mediated deoxyuridine triphosphate nick end-labelling (TUNEL) assays Paraffin-embedded ovarian tissue Sects. (4 µm) were dewaxed with dimethylbenzene, hydrated, washed, permeabilized with Proteinase K (20 μg/mL), and incubated at 37 °C for 30 min. The sections were stained using TUNEL reaction mixture and incubated at 37 °C for 60 min. After washing, 4’,6-diamidino-2-phenylindole was added to the slides for a 10-min incubation at room temperature to label cell nuclei. After washing three more times, the cells were examined using a fluorescence microscope (XD; Sunny Optical Technology (Group) Co., LTD, Zhejiang, China) and imaged using K-Viewer software (version 1.7.1.1; kfbio, China). Immunohistochemistry One of the hemispheres of each of the brains was fixed in 4% paraformaldehyde, embedded, and sectioned (4 µm). After dewaxing and rehydration, the samples were transferred to a citrate buffer and microwaved for 15 min for antigen retrieval. After washing, the sections were treated with 3% hydrogen peroxide. After blocking non-specific binding, the sections were incubated overnight at 4 °C with primary antibodies against Kiss1 (DF7133, Affinity Biosciences, China; 1:100), Kiss1 receptor (Kiss1R) (DF7123, Affinity Biosciences, China; 1:100), and GnRH (DF8553, Affinity Biosciences, China; 1:100). For negative controls, non-immune goat serum was used instead of primary antibodies. A goat anti-rabbit biotinylated immunoglobulin G secondary antibody (ab6721, Abcam, UK, 1:500) was added for a 40-min incubation at 37 °C. 3,3′-diaminobenzidine solution was added for chromogenic staining, and haematoxylin was used for counterstaining. The intensity of Kiss1, Kiss1R, and GnRH labelling was quantified using ImageJ software (version 1.53r; Wayne Raband, USA). Sections were examined using a digital pathology scanner (Aperio VERSA; Leica Biosystems, Shanghai, China). Western blotting Total proteins in each sample were extracted using radioimmunoprecipitation assay buffer, and the protein concentration was determined by the bicinchoninic acid method. Proteins were isolated by sodium dodecyl sulphate–polyacrylamide gel electrophoresis and transferred onto polyvinylidene fluoride membranes. After blocking in 5% skim milk at room temperature for 1 h, the samples were incubated overnight in a solution containing primary antibodies against insulin-like growth factor 1 (IGF-1) (DF6096, Affinity Biosciences, China; 1:1,000), IGF binding protein 1 (IGFBP-1) (DF7130, Affinity Biosciences, China; 1:1,000), and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (YM3029, ImmunoWay Biotechnology, China; 1:20,000) at 4 °C. The samples were subsequently incubated in a horseradish peroxidase-conjugated goat anti-rabbit immunoglobulin G (Beijing Dingguo Changsheng Biotechnology Co. Ltd., China; 1: 5,000) secondary antibody solution for 2 h, after which the density of each immunoreactive band was quantified using ImageJ software (version 1.53r; Wayne Raband, USA). Statistical analysis Statistical Package for the Social Sciences software (version 26.0; IBM, USA) was used for all statistical analyses. Measurement data are described as the mean ± standard error of the mean. Student’s t-tests and one-way analysis of variance were used for comparisons between two or more groups, respectively. The Kolmogorov–Smirnov test was used to assess the normality of the data distributions, and Levene’s test was used to assess the homogeneity of variance. If the variances were equal, post hoc comparisons between pairs of groups were performed using the least significant difference test. If the data were not normally distributed or the variance was unequal, the non-parametric Kruskal–Wallis’s test was used. In all analyses, p < 0.05 was considered statistically significant. Results Comparison of demographic and clinical variables between women with PCOS and healthy controls The clinical and demographic data of the enrolled patients were collected and analysed. There was no significant difference in age between the women with PCOS and the healthy controls (p = 0.143); however, the former group had a significantly higher BMI than the latter (p = 0.031). There was no significant difference in fasting blood glucose levels between the two groups (p = 0.367); however, the serum concentrations of Kiss1 and fasting INS were significantly higher in the women with PCOS, as was the HOMA-IR index (p = 0.013, p = 0.002, and p = 0.001, respectively). Compared with that of the healthy controls, the serum DA and 5-HT levels were significantly lower in the PCOS group (p = 0.049 and p = 0.041, respectively). The NE levels did not significantly differ between the two groups (p = 0.075), despite an obvious downward trend in the PCOS group. These comparisons are summarised in Table 1 and Fig. 1.Table 1Comparison of demographic and clinical variables between women with PCOS and healthy controlsParametersControl (n = 11)PCOS (n = 27)p-valuen (%)n (%)Age in years, mean (SEM)28.182 (1.034)30.444 (0.865)0.143Weight in kg, mean (SEM)60.773 (2.565)66.389 (2.589)0.209Weight classification0.205 Underweight0 (0.000)1 (3.704) Normal weight9 (81.818)13 (48.148) Overweight2 (18.182)7 (25.926) Obese0 (0.000)6 (22.222)Height in cm, mean (SEM)166.364 (1.479)164.185 (0.839)0.186BMI in kg/m2, mean (SEM)21.898 (0.733)a24.646 (0.977)a0.031Laboratory indicators, mean (SEM) Kiss1, ng/mL12.851 (1.227)a16.095 (0.621)a0.013 NE, pg/mL306.909 (15.531)247.609 (19.515)0.075 DA, nmol/L117.809 (7.029)a94.327 (6.738)a0.049 5-HT, ng/mL269.155 (13.848)227.622 (11.139)0.041 INS, mU/L6.464 (1.655)a13.186 (1.057)a0.002 GLU, mmol/L3.951 (0.086)4.086 (0.087)0.367 HOMA-IR index, mean (SEM)1.132 (0.288)a2.406 (0.203)a0.001Abbreviations: PCOS Polycystic ovary syndrome, SEM Standard error of the mean, BMI Body mass index, Kiss1 Kisspeptin, NE Norepinephrine, DA Dopamine, 5-HT 5-hydroxytryptamine/serotonin, INS Insulin, GLU blood glucose, HOMA-IR Homeostatic model assessment of insulin resistanceaValues significantly differ between groups at p < 0.05Fig. 1Comparisons of laboratory parameters between women with polycystic ovary syndrome (PCOS) and healthy controls. A Serum kisspeptin (Kiss1) levels. B Serum norepinephrine (NE) levels. C Serum dopamine (DA) levels. D Serum 5-hydroxytryptamine/serotonin (5-HT) levels. E Fasting blood glucose levels. F Fasting serum insulin levels. G Homeostatic model assessment of insulin resistance (HOMA-IR) index values, calculated as the fasting blood glucose level (mmol/L) × fasting insulin level (mU/L)/22.5. *: p < 0.05 compared to the control group; **: p < 0.01 compared to the control group Table 1Comparison of demographic and clinical variables between women with PCOS and healthy controlsParametersControl (n = 11)PCOS (n = 27)p-valuen (%)n (%)Age in years, mean (SEM)28.182 (1.034)30.444 (0.865)0.143Weight in kg, mean (SEM)60.773 (2.565)66.389 (2.589)0.209Weight classification0.205 Underweight0 (0.000)1 (3.704) Normal weight9 (81.818)13 (48.148) Overweight2 (18.182)7 (25.926) Obese0 (0.000)6 (22.222)Height in cm, mean (SEM)166.364 (1.479)164.185 (0.839)0.186BMI in kg/m2, mean (SEM)21.898 (0.733)a24.646 (0.977)a0.031Laboratory indicators, mean (SEM) Kiss1, ng/mL12.851 (1.227)a16.095 (0.621)a0.013 NE, pg/mL306.909 (15.531)247.609 (19.515)0.075 DA, nmol/L117.809 (7.029)a94.327 (6.738)a0.049 5-HT, ng/mL269.155 (13.848)227.622 (11.139)0.041 INS, mU/L6.464 (1.655)a13.186 (1.057)a0.002 GLU, mmol/L3.951 (0.086)4.086 (0.087)0.367 HOMA-IR index, mean (SEM)1.132 (0.288)a2.406 (0.203)a0.001Abbreviations: PCOS Polycystic ovary syndrome, SEM Standard error of the mean, BMI Body mass index, Kiss1 Kisspeptin, NE Norepinephrine, DA Dopamine, 5-HT 5-hydroxytryptamine/serotonin, INS Insulin, GLU blood glucose, HOMA-IR Homeostatic model assessment of insulin resistanceaValues significantly differ between groups at p < 0.05 Comparison of demographic and clinical variables between women with PCOS and healthy controls Abbreviations: PCOS Polycystic ovary syndrome, SEM Standard error of the mean, BMI Body mass index, Kiss1 Kisspeptin, NE Norepinephrine, DA Dopamine, 5-HT 5-hydroxytryptamine/serotonin, INS Insulin, GLU blood glucose, HOMA-IR Homeostatic model assessment of insulin resistance aValues significantly differ between groups at p < 0.05 Comparisons of laboratory parameters between women with polycystic ovary syndrome (PCOS) and healthy controls. A Serum kisspeptin (Kiss1) levels. B Serum norepinephrine (NE) levels. C Serum dopamine (DA) levels. D Serum 5-hydroxytryptamine/serotonin (5-HT) levels. E Fasting blood glucose levels. F Fasting serum insulin levels. G Homeostatic model assessment of insulin resistance (HOMA-IR) index values, calculated as the fasting blood glucose level (mmol/L) × fasting insulin level (mU/L)/22.5. *: p < 0.05 compared to the control group; **: p < 0.01 compared to the control group Correlation analysis between parameters in women with PCOS Correlations between the levels of various monoamine neurotransmitters and the glucose metabolism index in women with PCOS were analysed (see Table 2). NE and DA were both negatively correlated with INS levels and the HOMA-IR index (for NE and INS: p = 0.007, r = −0.429; for NE and the HOMA-IR index: p = 0.004, r = −0.456; for DA and INS: p = 0.010, r = −0.412; and for DA and the Homa-IR index: p = 0.007, r = −0.431). NE levels were also negatively correlated with blood glucose concentrations (p = 0.048, r = −0.323).Table 2Correlation analysis of the relationships between various parameters in women with PCOSParametersSpearman’s rp-valueKiss1 levels and other parameters BMI0.1470.377 Serum NE levels−0.4120.010a Serum DA levels−0.3530.030a Serum 5-HT levels−0.4230.008a Fasting insulin levels0.3540.029a Fasting blood glucose levels0.2410.145 HOMA-IR index0.3710.022aNE levels and other parameters BMI−0.4110.010a Serum Kiss1 levels−0.4120.010a Serum DA levels0.8750.000a Serum 5-HT levels0.4900.002a Fasting insulin levels−0.4290.007a Fasting blood glucose levels−0.3230.048a HOMA-IR index−0.4560.004aDA levels and other parameters BMI−0.5120.001a Serum Kiss1 levels−0.3530.030a Serum NE levels0.8750.000a Serum 5-HT levels0.6820.000a Fasting insulin levels−0.4120.010a Fasting blood glucose levels−0.2710.100a HOMA-IR index−0.4310.007a5-HT levels and other parameters BMI−0.4270.008a Serum Kiss1 levels−0.4230.008a Serum NE levels0.4900.002a Serum DA levels0.6820.000a Fasting insulin levels−0.1960.238 Fasting blood glucose levels−0.1920.249 HOMA-IR index−0.2150.196Abbreviations: PCOS, polycystic ovary syndrome; Kiss1, kisspeptin; BMI, body mass index; NE, norepinephrine; DA, dopamine; 5-HT, 5-hydroxytryptamine/serotonin; HOMA-IR, homeostatic model assessment of insulin resistanceaA statistically significant relationship Table 2Correlation analysis of the relationships between various parameters in women with PCOSParametersSpearman’s rp-valueKiss1 levels and other parameters BMI0.1470.377 Serum NE levels−0.4120.010a Serum DA levels−0.3530.030a Serum 5-HT levels−0.4230.008a Fasting insulin levels0.3540.029a Fasting blood glucose levels0.2410.145 HOMA-IR index0.3710.022aNE levels and other parameters BMI−0.4110.010a Serum Kiss1 levels−0.4120.010a Serum DA levels0.8750.000a Serum 5-HT levels0.4900.002a Fasting insulin levels−0.4290.007a Fasting blood glucose levels−0.3230.048a HOMA-IR index−0.4560.004aDA levels and other parameters BMI−0.5120.001a Serum Kiss1 levels−0.3530.030a Serum NE levels0.8750.000a Serum 5-HT levels0.6820.000a Fasting insulin levels−0.4120.010a Fasting blood glucose levels−0.2710.100a HOMA-IR index−0.4310.007a5-HT levels and other parameters BMI−0.4270.008a Serum Kiss1 levels−0.4230.008a Serum NE levels0.4900.002a Serum DA levels0.6820.000a Fasting insulin levels−0.1960.238 Fasting blood glucose levels−0.1920.249 HOMA-IR index−0.2150.196Abbreviations: PCOS, polycystic ovary syndrome; Kiss1, kisspeptin; BMI, body mass index; NE, norepinephrine; DA, dopamine; 5-HT, 5-hydroxytryptamine/serotonin; HOMA-IR, homeostatic model assessment of insulin resistanceaA statistically significant relationship Correlation analysis of the relationships between various parameters in women with PCOS Abbreviations: PCOS, polycystic ovary syndrome; Kiss1, kisspeptin; BMI, body mass index; NE, norepinephrine; DA, dopamine; 5-HT, 5-hydroxytryptamine/serotonin; HOMA-IR, homeostatic model assessment of insulin resistance aA statistically significant relationship The correlations between Kiss1 levels and other parameters were also analysed. Kiss1 levels were negatively correlated with the concentrations of NE, DA, and 5-HT (for Kiss1 and NE: p = 0.010, r = −0.412; for Kiss1 and DA: p = 0.030, r = −0.353; for Kiss1 and 5-HT: p = 0.008, r = −0.423), and they were positively correlated with INS levels and the HOMA-IR index (for Kiss1 and INS: p = 0.029, r = 0.354; for Kiss1 and the HOMA-IR index: p = 0.022, r = 0.371). Oestrous cycle changes in the rat models In the animal experiments, vaginal smears from the rats were continually tested. The control group rats had normal oestrous cycles, with each lasting 4–5 days, whereas the PCOS and PCOS + CUMS group rats were always in the interoestrous period, in which greater numbers of white blood cells were observed, with few keratinocytes and, occasionally, some nucleated epithelial cells. Changes in the oestrous cycle were only observed in one rat from the PCOS + CUMS group and in one rat from the PCOS group. After one week of intervention, the oestrous cycle of the CUMS rats began to slow down. The oestrous cycle changes in each group are summarised in Fig. 2.Fig. 2Changes in the oestrous cycle of rats in each of the four groups. Abbreviations: CUMS, chronic unpredictable mild stress; PCOS, polycystic ovary syndrome; P, proestrus stage; E, oestrus stage; MD, metoestrus or dioestrus stage Changes in the oestrous cycle of rats in each of the four groups. Abbreviations: CUMS, chronic unpredictable mild stress; PCOS, polycystic ovary syndrome; P, proestrus stage; E, oestrus stage; MD, metoestrus or dioestrus stage Body, ovarian, and uterine mass changes in rats The body mass of each rat was measured at the start of the experiment and once a week thereafter. At baseline, there was no significant difference in body mass between groups (p = 0.898). After one week, the body mass of the rats in the PCOS and PCOS + CUMS groups was significantly higher than that of the rats in the CUMS group (p = 0.001). After two weeks, the body mass of the rats in the CUMS was significantly lower than that of the controls (p = 0.006), whereas the body mass of the rats in the PCOS + CUMS and PCOS groups was significantly higher (p = 0.042, p = 0.01). At the end of the experiment, the bilateral ovarian mass, ovarian volume, and uterine mass of the rats in each group were measured. The ovarian volume significantly differed between the four groups on both sides (left: p = 0.000; right: p = 0.021), with greater volumes in the model groups than in the control group. Compared with that of the control group, the ovarian mass was significantly higher in the PCOS and PCOS + CUMS groups on both sides (PCOS left: p = 0.000; PCOS right: p = 0.002; PCOS + CUMS left: p = 0.000; PCOS + CUMS right: p = 0.000). The ovarian mass was also higher in the CUMS group than in the control group, but only on the right side (p = 0.044). Based on the body and ovarian masses, the ovarian index was calculated in each group. The ovarian index of the CUMS and PCOS + CUMS groups was significantly higher than that of the control group on both sides (CUMS left: p = 0.008; CUMS right: p = 0.004; PCOS + CUMS left: p = 0.019; PCOS + CUMS right: p = 0.026). Although there was also an increasing trend in the PCOS group, the ovarian index did not significantly differ from that of the control group on either side (left: p = 0.102; right: p = 0.157). The uterine masses of the PCOS and PCOS + CUMS groups were lower than those of the control and CUMS groups (p = 0.000 for all comparisons), and the uterine mass of the CUMS group was also significantly lower than that of the control group (p = 0.000). Intergroup comparisons of the ovarian mass, volume, and index, body mass changes over time, and uterine mass are shown in Fig. 3, along with representative images of ovarian and uterine tissues.Fig. 3Changes in the body, ovarian, and uterine mass in the four groups of rats. A–C Bilateral intergroup comparisons of the ovarian mass, ovarian volume, and ovarian index, respectively. D Intergroup comparison of the changes in body mass over time. E Intergroup comparison of the uterine mass. F Representative images of the ovaries (top) and uterus (bottom) of rats in each group. *: p < 0.05 vs. control group; **: p < 0.01 vs. control group. #: p < 0.05 vs. CUMS group; ##: p < 0.01 vs. CUMS group. CUMS, chronic unpredictable mild stress; PCOS, polycystic ovary syndrome Changes in the body, ovarian, and uterine mass in the four groups of rats. A–C Bilateral intergroup comparisons of the ovarian mass, ovarian volume, and ovarian index, respectively. D Intergroup comparison of the changes in body mass over time. E Intergroup comparison of the uterine mass. F Representative images of the ovaries (top) and uterus (bottom) of rats in each group. *: p < 0.05 vs. control group; **: p < 0.01 vs. control group. #: p < 0.05 vs. CUMS group; ##: p < 0.01 vs. CUMS group. CUMS, chronic unpredictable mild stress; PCOS, polycystic ovary syndrome Pathological findings in ovarian tissue in rats The H&E staining revealed that the ovarian tissue of the control group rats had a normal morphological structure. More specifically, follicles and the corpus luteum were observed at various stages of development in the cortex. The oocyte, cumulus, and radial crown structures were visible in the follicles, and multiple layers of morphologically complete granular cells were densely arranged around the follicular cavity. In the PCOS rats, however, the ovarian membranes were thickened, and multiple dilated cystic follicles were observed. Oocytes, cumulus, and radial crown structures were absent, and the number of granule cell layers was significantly lower than that of the controls; these changes were accompanied by an increased number of immature small follicles, and the corpus luteum was difficult to observe. Similar changes were observed in the CUMS group, including the presence of several cystic dilated follicles and the absence of normal structures. In the PCOS + CUMS group, the ovarian membrane was significantly thickened, the number of cystic dilated follicles was further increased, normal structures were absent, and the number of granule cell layers was further reduced. Representative images are shown in Fig. 4(A).Fig. 4Intergroup comparison of the changes in reproductive function in rats. A Representative images of the haematoxylin and eosin staining of the pathophysiology of ovarian tissues in each of the four groups. Magnification of images in the top row = 40 ×; scale bars = 500 μm. Magnification of the images in the bottom row = 200 ×; scale bars = 100 μm. B–F Intergroup comparisons of the changes in serum levels of sex hormones in rats. B FSH levels. C LH levels. D LH/FSH ratios. E FT levels. F E2 levels. *: p < 0.05 vs. control group; **: p < 0.01 vs. control group. #: p < 0.05 vs. CUMS group; ##: p < 0.01 vs. CUMS group. CUMS, chronic unpredictable mild stress; PCOS, polycystic ovary syndrome; FSH, follicle-stimulating hormone; LH, luteinizing hormone; FT, free testosterone; E2, estradiol Intergroup comparison of the changes in reproductive function in rats. A Representative images of the haematoxylin and eosin staining of the pathophysiology of ovarian tissues in each of the four groups. Magnification of images in the top row = 40 ×; scale bars = 500 μm. Magnification of the images in the bottom row = 200 ×; scale bars = 100 μm. B–F Intergroup comparisons of the changes in serum levels of sex hormones in rats. B FSH levels. C LH levels. D LH/FSH ratios. E FT levels. F E2 levels. *: p < 0.05 vs. control group; **: p < 0.01 vs. control group. #: p < 0.05 vs. CUMS group; ##: p < 0.01 vs. CUMS group. CUMS, chronic unpredictable mild stress; PCOS, polycystic ovary syndrome; FSH, follicle-stimulating hormone; LH, luteinizing hormone; FT, free testosterone; E2, estradiol Changes in serum levels of sex hormones Compared with the levels in the control group, the concentrations of FSH, LH, and FT significantly differed in the PCOS and PCOS + CUMS groups (PCOS: p = 0.023, 0.008, and 0.016, respectively; PCOS + CUMS: p = 0.019, 0.003, and 0.006, respectively); however, E2 levels in those two groups did not differ from those of the control group (p = 0.555 and p = 0.845, respectively). More specifically, compared with the concentrations in the control group, the FSH levels in the PCOS and PCOS + CUMS groups were lower, whereas the LH and FT levels were higher; the LH/FSH ratio was also significantly increased (p = 0.001 and p = 0.000, respectively). The levels of FSH, LH, and FT, and the LH/FSH ratio in the CUMS group showed the same trend as seen in the PCOS group, although none reached the level of statistical significance (p = 0.428, 0.331, 0.263, and 0.235, respectively). These comparisons are shown in Fig. 4(B–F). Depressive-like behaviours in the model rats To assess depressive-like behaviours, the SPT and OFT were performed at baseline and six weeks after the start of the experiment. The OFT was conducted over a 5-min session to assess the total distance travelled, the rearing frequency, the number of times the rats crossed into different regions during exploration, and the duration of time spent in the central area; there was no significant intergroup difference for any of the four parameters at baseline (p = 0.333, p = 0.129, p = 0.665, and p = 0.728, respectively). After 6 weeks of intervention, compared with that of the control group, the rearing and crossing frequencies and the duration of time spent in the central area were all significantly lower in the CUMS and PCOS + CUMS groups (CUMS: p = 0.024, p = 0.015, and p = 0.019, respectively; PCOS + CUMS: p = 0.031, p = 0.019, and p = 0.009, respectively); however, in the PCOS group, only the number of crossings and the duration of time spent in the central area were significantly reduced (p = 0.044 and p = 0.009, respectively). The OFT results are shown in Fig. 5(A–E).Fig. 5Assessment of depressive-like behaviours. A–D Results of the open field testing at baseline and after intervention for the 5-min test session. A The total distance travelled. B The duration of time spent in the central area. C The rearing frequency. D The total number of crossings. E Representative tracings of movement recorded in the open field test in each group. *: p < 0.05 vs. control group; **: p < 0.01 vs. control group. #: p < 0.05 vs. CUMS group; ##: p < 0.01 vs. CUMS group. CUMS, chronic unpredictable mild stress; PCOS, polycystic ovary syndrome Assessment of depressive-like behaviours. A–D Results of the open field testing at baseline and after intervention for the 5-min test session. A The total distance travelled. B The duration of time spent in the central area. C The rearing frequency. D The total number of crossings. E Representative tracings of movement recorded in the open field test in each group. *: p < 0.05 vs. control group; **: p < 0.01 vs. control group. #: p < 0.05 vs. CUMS group; ##: p < 0.01 vs. CUMS group. CUMS, chronic unpredictable mild stress; PCOS, polycystic ovary syndrome As shown in Fig. 6(A), there was no significant difference in the sucrose preference values between the groups at baseline (p = 0.137). However, after six weeks of intervention, the sucrose preference values in the CUMS and PCOS + CUMS groups were significantly lower than that of the control group (p = 0.003 and p = 0.013, respectively), suggesting that the rats were less motivated to seek a rewarding stimulus.Fig. 6The results of the sucrose preference testing and comparison of hippocampal monoamine neurotransmitter levels in the model rats. A Results of the sucrose preference testing at baseline and after the intervention. B–C Monoamine neurotransmitter levels in hippocampal tissue in each group, assessed using liquid chromatography-tandem mass spectrometry. B Levels of dopamine (DA). C Levels of norepinephrine (NE). D Levels of 5-hydroxytryptamine/serotonin (5-HT). *: p < 0.05 vs. control group; **: p < 0.01 vs. control group. #: p < 0.05 vs. CUMS group. CUMS, chronic unpredictable mild stress; PCOS, polycystic ovary syndrome The results of the sucrose preference testing and comparison of hippocampal monoamine neurotransmitter levels in the model rats. A Results of the sucrose preference testing at baseline and after the intervention. B–C Monoamine neurotransmitter levels in hippocampal tissue in each group, assessed using liquid chromatography-tandem mass spectrometry. B Levels of dopamine (DA). C Levels of norepinephrine (NE). D Levels of 5-hydroxytryptamine/serotonin (5-HT). *: p < 0.05 vs. control group; **: p < 0.01 vs. control group. #: p < 0.05 vs. CUMS group. CUMS, chronic unpredictable mild stress; PCOS, polycystic ovary syndrome Downregulation of monoamine neurotransmitter levels in the hippocampus of PCOS, CUMS, and PCOS + CUMS model rats Hippocampal tissues collected from each group of rats were subjected to LC–MS/MS to quantify the levels of various neurotransmitters, including DA (Fig. 6(B)), NE (Fig. 6(C)), and 5-HT (Fig. 6(D)). Compared with the levels in the control group, the DA, 5-HT, and NE levels in the CUMS and PCOS + CUMS groups were significantly reduced (CUMS: p = 0.032, p = 0.002, and p = 0.002, respectively; PCOS + CUMS: p = 0.009, p = 0.001, and p = 0.002, respectively); in the PCOS group, the DA and 5-HT levels were also significantly diminished (p = 0.035 and p = 0.002, respectively), whereas the NE levels were unchanged (p = 0.056). Abnormal glucose metabolism in the PCOS, CUMS, and PCOS + CUMS model rats To investigate whether the CUMS paradigm or letrozole administration affected glucose metabolism in the model rats, a GTT and an ITT were performed. For the GTT, blood glucose levels were measured after short-term fasting and at 30, 60, 90, and 120 min after glucose administration. The rats in the CUMS and PCOS + CUMS groups exhibited higher fasting glucose levels than those in the control group; however, the differences were not statistically significant (p = 0.248 and p = 0.165, respectively) (Fig. 7(A)). Compared to that of the control group, the area under the curve of the CUMS, PCOS, and PCOS + CUMS groups increased, but not significantly (p = 0.156, p = 0.064, and p = 0.074, respectively) (Fig. 7(B)).Fig. 7Confirmation of glucose metabolism abnormalities in the model rats. A Changes in blood glucose levels in each group during the oral glucose tolerance test (GTT). B The area under the curve values for each group in the oral GTT. C Changes in blood glucose levels in each group during the insulin tolerance test (ITT). D The area under the curve values for each group in the ITT. E Fasting blood glucose levels in each group. F Fasting insulin levels in each group. G The homeostatic model assessment of insulin resistance (HOMA-IR) index in each group, calculated as the fasting blood glucose level (mmol/L) × fasting insulin level (mU/L)/22.5. *: p < 0.05 vs. control group; **: p < 0.01 vs. control group. #: p < 0.05 vs. CUMS group. CUMS, chronic unpredictable mild stress; PCOS, polycystic ovary syndrome Confirmation of glucose metabolism abnormalities in the model rats. A Changes in blood glucose levels in each group during the oral glucose tolerance test (GTT). B The area under the curve values for each group in the oral GTT. C Changes in blood glucose levels in each group during the insulin tolerance test (ITT). D The area under the curve values for each group in the ITT. E Fasting blood glucose levels in each group. F Fasting insulin levels in each group. G The homeostatic model assessment of insulin resistance (HOMA-IR) index in each group, calculated as the fasting blood glucose level (mmol/L) × fasting insulin level (mU/L)/22.5. *: p < 0.05 vs. control group; **: p < 0.01 vs. control group. #: p < 0.05 vs. CUMS group. CUMS, chronic unpredictable mild stress; PCOS, polycystic ovary syndrome In the fifth week of the intervention, an ITT was conducted, and the blood glucose levels of the rats were measured after short-term fasting and at 30, 60, 90, and 120 min after INS injection. The fasting blood glucose levels in the CUMS and PCOS + CUMS groups were significantly higher than those in the control group (p = 0.014 and p = 0.005, respectively) (Fig. 7(C)). Compared to that of the control group, the area under the curve values for the CUMS, PCOS, and PCOS + CUMS groups were significantly increased (p = 0.000, p = 0.007, and p = 0.001, respectively) (Fig. 7(D)). At the end of the experiment, INS and blood glucose levels were measured after food and water fasting for a period of more than 8 h (Fig. 7(E, F)). The INS levels in the PCOS and PCOS + CUMS groups were significantly higher than those in the control group (p = 0.001 and p = 0.001, respectively); a similar increasing trend was observed in the CUMS group, although the difference was not significant (p = 0.107). The fasting blood glucose levels of the CUMS, PCOS, and PCOS + CUMS groups were significantly higher than those of the control group (p = 0.011, p = 0.003, and p = 0.002, respectively). Finally, the HOMA-IR index of the PCOS and PCOS + CUMS groups increased significantly (p = 0.001 and p = 0.001, respectively); a similar but non-significant trend was observed in the CUMS group (p = 0.080) (Fig. 7(G)). Assessment of granulosa cell apoptosis TUNEL staining was performed to assess the level of apoptosis occurring in ovarian granulosa cells in each group, as shown in Fig. 8(A, B). In the ovarian tissue, the total area and the TUNEL-positive fluorescence intensity were quantified using ImageJ software (Version 1.53r, Wayne Raband, USA), and the TUNEL fluorescence ratio was expressed as the TUNEL-positive fluorescence intensity/total area. The TUNEL fluorescence ratio in the ovarian tissues of the PCOS and PCOS + CUMS groups was higher than that in the control group (p = 0.005 and p = 0.001, respectively); a similar trend was observed in the CUMS group, although the difference was not significant (p = 0.129).Fig. 8Assessment of granulosa cell apoptosis and changes in IGF-1/IGFBP-1 expression in ovarian tissues of rats. A Representative images of TUNEL staining in each group used to visualize the degree of apoptosis occurring in granulosa cells. Green fluorescence indicates positive (apoptotic) cells, and blue fluorescence indicates DAPI staining of all cell nuclei. B The apoptosis rate in each group, calculated as the fraction of the total number of cells with positive immunolabeling. C and E Relative IGF-1 expression levels in ovarian tissues quantified by western blotting using GAPDH as an internal control. D and F Relative IGFBP-1 expression levels in ovarian tissues quantified by western blotting using GAPDH as an internal control. *: p < 0.05 vs. control group; **: p < 0.01 vs. control group. #: p < 0.05 vs. CUMS group; ##: p < 0.01 vs. CUMS group. CUMS, chronic unpredictable mild stress; PCOS, polycystic ovary syndrome; TUNEL, terminal deoxynucleotidyl transferase dUTP nick-end labelling; DAPI, 4’,6-diamidino-2-phenylindole; IGF-1, insulin-like growth factor 1; IGFBP-1, insulin-like growth factor binding protein 1; GAPDH, glyceraldehyde 3-phosphate dehydrogenase Assessment of granulosa cell apoptosis and changes in IGF-1/IGFBP-1 expression in ovarian tissues of rats. A Representative images of TUNEL staining in each group used to visualize the degree of apoptosis occurring in granulosa cells. Green fluorescence indicates positive (apoptotic) cells, and blue fluorescence indicates DAPI staining of all cell nuclei. B The apoptosis rate in each group, calculated as the fraction of the total number of cells with positive immunolabeling. C and E Relative IGF-1 expression levels in ovarian tissues quantified by western blotting using GAPDH as an internal control. D and F Relative IGFBP-1 expression levels in ovarian tissues quantified by western blotting using GAPDH as an internal control. *: p < 0.05 vs. control group; **: p < 0.01 vs. control group. #: p < 0.05 vs. CUMS group; ##: p < 0.01 vs. CUMS group. CUMS, chronic unpredictable mild stress; PCOS, polycystic ovary syndrome; TUNEL, terminal deoxynucleotidyl transferase dUTP nick-end labelling; DAPI, 4’,6-diamidino-2-phenylindole; IGF-1, insulin-like growth factor 1; IGFBP-1, insulin-like growth factor binding protein 1; GAPDH, glyceraldehyde 3-phosphate dehydrogenase Changes in IGF-1/IGFBP-1 levels in ovarian tissues The expression levels of IGF-1 and IGFBP-1 have been shown to be correlated with the degree of IR [2],Guto et al., 2023). Therefore, to investigate whether letrozole administration and the CUMS paradigm could cause local IR in ovarian tissue, the levels of IGF-1 and IGFBP-1 were quantified using western blotting (Fig. 8(C–F)). IGF-1 levels were downregulated in the PCOS + CUMS groups compared to those in the control group (p = 0.022), whereas no significant difference was observed in the PCOS and CUMS group (p = 0.079 and p = 0.976, respectively). Compared to that of the control group, there was no statistically significant difference in IGFBP-1 expression in the CUMS, PCOS, or PCOS + CUMS groups (p = 0.117). Changes in Kiss1, Kiss1R, and GnRH expression in the hypothalamus of model rats The expression levels of Kiss1, Kiss1R, and GnRH in the hypothalamus of rats were quantified by immunohistochemistry (Fig. 9). Hypothalamic expression of Kiss1 was significantly higher in the CUMS, PCOS, and PCOS + CUMS groups than in the control group (p = 0.005, p = 0.001, and p = 0.001, respectively). Hypothalamic Kiss1R expression was also significantly increased in the CUMS and PCOS + CUMS groups (p = 0.027 and p = 0.004, respectively), although no significant change was observed in the PCOS group (p = 0.191). Similarly, the expression of GnRH in the PCOS and PCOS + CUMS groups was significantly elevated compared to that of the control group (p = 0.042 and p = 0.013, respectively), whereas there was no significant change in the CUMS group (p = 0.535).Fig. 9Changes in Kiss1, Kiss1R, and GnRH expression in the hypothalamus of rats. A Immunohistochemical labelling of Kiss1 (top row), Kiss1R (middle row), and GnRH (bottom row) in the hypothalamus of rats in each group, with brown particles indicating positive antibody labelling. Magnification of the images = 200 ×; scale bars = 100 μm. B–D IOD values of Kiss1, Kiss1R, and GnRH labelling, respectively, in each group. *: p < 0.05 vs. control group; **: p < 0.01 vs. control group. #: p < 0.05 vs. CUMS group; ##: p < 0.01 vs. CUMS group. CUMS, chronic unpredictable mild stress; PCOS, polycystic ovary syndrome; Kiss1, kisspeptin; Kiss1R, kisspeptin receptor; GnRH, gonadotropin-releasing hormone; IOD, integrated optical density Changes in Kiss1, Kiss1R, and GnRH expression in the hypothalamus of rats. A Immunohistochemical labelling of Kiss1 (top row), Kiss1R (middle row), and GnRH (bottom row) in the hypothalamus of rats in each group, with brown particles indicating positive antibody labelling. Magnification of the images = 200 ×; scale bars = 100 μm. B–D IOD values of Kiss1, Kiss1R, and GnRH labelling, respectively, in each group. *: p < 0.05 vs. control group; **: p < 0.01 vs. control group. #: p < 0.05 vs. CUMS group; ##: p < 0.01 vs. CUMS group. CUMS, chronic unpredictable mild stress; PCOS, polycystic ovary syndrome; Kiss1, kisspeptin; Kiss1R, kisspeptin receptor; GnRH, gonadotropin-releasing hormone; IOD, integrated optical density Discussion PCOS is the most common cause of ovulatory infertility and is characterised by anovulation, clinical or biochemical manifestations of hyperandrogenism, and polycystic ovarian changes. The clinical manifestations of PCOS are heterogeneous, and its aetiology is complex, making it difficult to determine the precise underlying pathophysiological mechanisms that promote its development. However, there is an accumulating body of evidence suggesting that psychological factors are involved in its pathogenesis. In a previous study, the present authors reported that the administration of letrozole combined with exposure to CUMS could lead to the establishment of a model in rats that reflects the ovarian pathological and endocrinological changes observed clinically in patients with PCOS, as well as similar behavioural changes and alterations in related neurotransmitter levels [48]. However, the effects of this model on energy metabolism and the link with reproductive dysfunction were previously unexplored. Thus, in the present study, three model groups were established and compared in rats, including the PCOS, CUMS, and PCOS + CUMS groups. The results confirmed that the expression levels of LH and FT were elevated in the PCOS group relative to those in the control group, as was the LH/FSH ratio, consistent with the characteristics of the endocrine disorders observed clinically in patients with PCOS. There was no change in E2 levels, which is consistent with the findings of Wu et al. [70]. The FT level was also elevated in the CUMS group, and the endocrine characteristics of the rats in the PCOS + CUMS group were consistent with but more severe than those of the PCOS rats. Therefore, it is speculated that CUMS may aggravate endocrine disorders in patients with PCOS. In terms of ovarian changes, the volume and mass of the bilateral ovaries in the PCOS group were significantly higher than those of the control group, and the histological assessment of the local ovarian pathology revealed the presence of multiple dilated cystic follicles, with similar pathological manifestations observed in the rats in the CUMS group. A significant increase in the volume and mass of both ovaries was also evident in the PCOS + CUMS group, with the changes being more severe than those observed in the PCOS group. The uterine mass of the rats was significantly reduced by the administration of letrozole, an aromatase inhibitor, as well as by exposure to the CUMS paradigm. These findings suggest that CUMS induced morphological changes in the ovarian and uterine tissues, although greater abnormalities were observed in the PCOS rats. The results of the behavioural testing revealed a reduced sucrose preference, as well as lower activity levels and exploratory behaviours in the OFT. Although there was no significant change in sucrose preference in the PCOS rats, they did exhibit decreased activity in the OFT as well. The results of the LC–MS/MS analysis of monoamine neurotransmitter levels revealed significantly lower levels of NE, DA, and 5-HT in the hippocampus following exposure to the CUMS paradigm, with similar changes observed in the PCOS and PCOS + CUMS groups. Collectively, these findings are indicative of an imbalance in monoamine neurotransmitter levels in the PCOS model rats, which is consistent with the conclusion of the aforementioned study [48]. IR in women with PCOS is a key pathophysiological factor [9]. In the central nervous system, INS can induce LH release either by increasing the expression of the gene that encodes for GnRH or by directly stimulating the pituitary gland to secrete LH and promoting androgen synthesis in the ovaries and adrenal glands [68, 77]. The HOMA-IR index has been widely used in scientific research to assess the degree of INS sensitivity [65]. In the present study, the fasting blood glucose and INS levels, as well as the HOMA-IR index, were significantly elevated in the PCOS and PCOS + CUMS groups, despite the absence of a high-fat diet. Notably, glucose metabolism was also dysregulated in the CUMS group, with a similar trend to that observed in the PCOS group. In the ITT, the area under the curve of the CUMS group was significantly higher than that of the control group, and the fasting blood glucose and INS levels, as well as the HOMA-IR index, were also increased. Despite these observations, exposure to the CUMS paradigm caused a reduction in body mass and slowed the weight gain of PCOS rats, which is consistent with other experimental results [78]. In a clinical population, the present authors have also observed changes in neurotransmitter levels and altered glucose metabolism in women with PCOS, with a higher HOMA-IR index and fasting INS levels, a reduction in DA and 5-HT levels, and a downward trend in NE levels. To explore the mechanisms, a small-scale clinical study and corresponding animal experiments were conducted. Kiss1, previously known as metastin, is a suppressor of metastasis, and its precursor protein can be hydrolysed to produce kisspeptin-54/14/13/10 [36]. It is primarily distributed in the arcuate and anterior ventral periventricular nuclei of the hypothalamus [21]. Kiss1-expressing neurons act as upstream regulators of GnRH and play key roles in the modulation of the HPO axis, thereby regulating reproductive processes [71]. Kiss1 is also closely related to the function of the HPA axis, which is an important component of the stress response system. A previous study found that Kiss1 neurons in the arcuate nucleus of the hypothalamus also express corticotropin-releasing hormone and glucocorticoid receptors, which are other important components of the HPA axis [45]. As a bridge between the HPO and HPA axes, the presence of Kiss1-expressing neurons may be one of the reasons why exposure to stressors leads to the development of reproductive disorders [41]. Other studies have reported the presence of Kiss1 signals in limbic regions of the brain that are known to be associated with the modulation of emotional responses, linking such behaviours with the regulation of the reproductive axis [10]. The clinical component of this study revealed elevated levels of Kiss1 in patients with PCOS compared to those in the healthy controls, consistent with previous findings [26, 33, 46, 50, 53, 59, 66]. Moreover, Kiss1 expression was negatively correlated with the levels of NE, DA, and 5-HT, all of which are monoamine neurotransmitters. In recent years, Kiss1 expression has been shown to be closely associated with energy metabolism [36]. As a reproductive disorder closely related to metabolic dysfunction, PCOS is also associated with Kiss1 expression, and its underlying mechanism of action has attracted much attention. The present clinical results suggested that Kiss1 levels were positively correlated with INS concentrations and the HOMA-IR index; this finding is consistent with some studies. For example, one study showed that Kiss1 levels in women with PCOS and comorbid obesity negatively correlated with blood glucose concentration, whereas an opposite correlation was observed in non-obese women [14]. However, another study failed to show a correlation between Kiss1 and INS levels [24]. Some studies suggest that Kiss1 has an insulinotropic effect, but this requires sufficient glucose concentrations and depends on the specific Kiss1 peptides and their concentrations [34]. The effect of Kiss1 on insulin secretion may also be mediated through its regulation of the HPO axis. For example, androgens can act via androgen receptors on pancreatic β-cells to stimulate insulin secretion [72]. Some researchers have proposed that Kiss1 interacts with the limbic system in the brain, linking it to emotion-related eating disorders such as binge eating, which may exacerbate insulin resistance in PCOS [28]. However, further basic research is required to elucidate the role of Kiss1 in these processes. In the present study, Kiss1 expression in the hypothalamus of rats was detected by immunohistochemistry, which revealed an upregulated expression in all three model groups, consistent with the results reported by Esparza et al. [20], suggesting that Kiss1-expressing neurons in the hypothalamus are activated and the levels become elevated after the administration of letrozole. Although Kiss1R expression was not significantly elevated in the PCOS group in the present study, both the PCOS and PCOS + CUMS groups exhibited a significant increase in GnRH levels. Kiss1 neurons are important targets in the feedback mechanisms that regulate the activity of the HPO axis, and they play a major role in the regulation of GnRH/LH pulses [71]. The present results are consistent with the neuroendocrine characteristics associated with PCOS. Previous studies have shown that kisspeptin-10 exposure can reduce 5-HT and DA levels and increase 3,4-dihydroxyphenylacetic acid/DA and 5-hydroxyindoleacetic acid/5-HT ratios [44], and DA can suppress the GnRH pulse frequency by inhibiting Kiss1 release [25]. The results of the present study confirmed that chronic exposure to psychological stressors resulted in decreased monoamine neurotransmitter levels, including a reduction in DA, weakening the inhibitory effect on Kiss1, thereby accelerating the GnRH pulse frequency, which promoted the reproductive endocrine dysfunction of PCOS. Kiss1 is also known to be involved in the regulation of glucose homeostasis, and elevated circulating Kiss1 levels have been demonstrated in some metabolic disorders such as obesity and type 2 diabetes [36]. Within a certain range, elevated levels of Kiss1 may bind to receptors other than Kiss1R and stimulate INS secretion in pancreatic beta cells [13, 36, 61, 62]. In PCOS, this may be one of the mechanisms that aggravate abnormal glucose metabolism and may lead to a vicious cycle of hyperinsulinemia and hyperandrogenaemia. INS can increase the release of LH and stimulate androgen synthesis by increasing the expression of the GnRH gene or by directly stimulating the pituitary gland [77]. Excessive androgen levels can induce apoptosis in oocytes and granulosa cells and cause follicular dysfunction [57, 67, 73] Ovarian granule cells are particularly important for follicular development. In women with PCOS, the ovaries exhibit a loose arrangement of granule cells, accompanied by follicular cystic dilation. The abnormal apoptosis of granulosa cells may be the main factor responsible for the arrest of follicular maturation [47, 55]. In recent years, there has been an increasing focus on the roles and functions of IGF-related systems. IGF proteins are expressed at multiple sites and regulate diverse cellular processes, including those related to survival, proliferation, differentiation, and glucose metabolism [51, 52, 58]. Some studies have suggested that ovarian IGF-1 may play an important role in the regulation of granulosa cell apoptosis [52, 69]. IGFs act through many signalling pathways, one of the most important of which is the phosphoinositide 3-kinase (PI3K)/protein kinase B (Akt) pathway. IGFs bind to both IGF and INS receptors, leading to the activation of INS receptor substrates, which, in turn, results in the phosphorylation of P13K and the recruitment and activation of Akt, the latter of which halts cell apoptosis by inhibiting a pro-apoptotic protein or by disrupting the pro-apoptotic signalling generated by forkhead box protein O1 [18, 64]. The IGF-1/PI3K/Akt signalling pathway is also capable of promoting DNA repair through homologous recombination and protects granulosa cells against DNA damage and apoptosis [38]. This signalling pathway is also involved in glucose metabolism, and the complex formed by Akt and its downstream factor protein kinase B can traffic glucose transporters to cell membranes, thereby increasing glucose uptake [79]. IGFBPs regulate the biological activity of IGFs,in most cases, they inhibit the action of IGFs by preventing their binding to their respective receptors, although they may also play enhanced or more specific roles [3]. For example, IGFBP-1 is primarily involved in the acute regulation of IGF activity and is itself modulated by metabolic processes. Some findings that have shown decreased IGFBP-1 levels are inversely associated with an increase in fasting INS levels [27, 31, 32]. Therefore, to investigate the specific mechanism driving the stress-induced arrest of follicular maturation and the manifestation of metabolic disorders in PCOS, the rate of apoptosis in granulosa cells and the protein levels of IGF-1/IGFBP-1 in ovarian tissues were detected using TUNEL staining and western blotting, respectively. The TUNEL staining revealed a significant increase in the apoptosis of ovarian granulosa cells in the PCOS and PCOS + CUMS groups, consistent with previous findings. Critically, a similar increase in granulosa cell apoptosis was observed in the CUMS group in the absence of letrozole administration. The western blotting results demonstrated that the IGF-1 levels in ovarian tissue were downregulated significantly in the PCOS + CUMS group. Another study that collected granulosa cells from patients with PCOS reported a downregulation in the expression levels of IGF-1 protein and the gene encoding it in patients with PCOS with IR, whereas the expression of IGF-1R was upregulated [5]. Therefore, the mechanism of PCOS-related psychological stress may be related to the regulation of IGF-1, weakening DNA repair mechanisms, promoting apoptosis of granulosa cells, and reducing INS sensitivity. In conclusion, considering the important correlation between the Kiss1 signalling system and reproductive and metabolic processes, this study explored the mechanism through which CUMS regulates reproductive function, glucose metabolism, and affective behaviours in a rat model of letrozole-induced PCOS. Based on the findings, it is hypothesised that CUMS may reduce monoamine neurotransmitter levels, weakening their inhibitory effect on Kiss1, leading to an upregulation of the Kiss1/Kiss1R/GnRH signalling pathway, thereby promoting INS secretion and further aggravating the development of reproductive endocrine disorders and abnormal glucose metabolism in PCOS. Moreover, regulating the expression of ovarian IGF-1 may, in turn, modulate its downstream signalling pathway, weakening DNA repair mechanisms, promoting granulosa cell apoptosis, and aggravating follicle maturation disorder in PCOS, or its downstream INS signalling pathway may be regulated to reduce INS sensitivity, as shown in Fig. 10. These changes may represent a potential mechanism for the pathogenesis of PCOS-related psychological stress.Fig. 10The mechanism by which chronic psychological stress modulates kisspeptin, thereby altering reproductive function and glucose metabolism in patients with PCOS. Akt, protein kinase B; BAD, B-cell lymphoma 2-associated death promoter; FOXO1, forkhead box protein O1; GnRH, gonadotropin-releasing hormone; IGF, insulin-like growth factor; IGFBP-1, insulin-like growth factor binding protein 1; Kiss1R, kisspeptin receptor; LH, luteinizing hormone; PCOS, polycystic ovary syndrome; PI3K, phosphoinositide 3-kinase The mechanism by which chronic psychological stress modulates kisspeptin, thereby altering reproductive function and glucose metabolism in patients with PCOS. Akt, protein kinase B; BAD, B-cell lymphoma 2-associated death promoter; FOXO1, forkhead box protein O1; GnRH, gonadotropin-releasing hormone; IGF, insulin-like growth factor; IGFBP-1, insulin-like growth factor binding protein 1; Kiss1R, kisspeptin receptor; LH, luteinizing hormone; PCOS, polycystic ovary syndrome; PI3K, phosphoinositide 3-kinase However, this study has some limitations. Firstly, it was not possible to confirm whether there were specific changes in pathways downstream of IGF-1 or whether CUMS affected reproduction, metabolism, and psychological stress in PCOS through other pathways. Secondly, the sample sizes in both the clinical and animal experiments of this study were relatively small, which may limit the generalizability of the results. Future studies will focus on the pathological mechanisms identified herein, expand the sample size, and investigate downstream signaling pathways, with the aim of identifying new therapeutic targets and improving clinical outcomes. Supplementary Information Supplementary Material 1 Supplementary Material 1