{"paper_id":"3c56bc14-10cb-4af0-9190-54912d7bcb9f","body_text":"1\nVol.:(0123456789)Scientific Reports |         (2024) 14:5984  | https://doi.org/10.1038/s41598-024-55993-8\nwww.nature.com/scientificreports\nEpidemiologic and genetic \nassociations of female reproductive \ndisorders with depression \nor dysthymia: a Mendelian \nrandomization study\nShuyi Ling 1,2, Yuqing Dai 1,2, Ruoxin Weng 1, Yuan Li 1, Wenbo Wu 1, Ziqiong Zhou 1, \nZhisheng Zhong 1* & Yuehui Zheng 1*\nObservational studies have previously reported an association between depression and certain \nfemale reproductive disorders. However, the causal relationships between depression and different \ntypes of female reproductive disorders remain unclear in terms of direction and magnitude. We \nconducted a comprehensive investigation using a two-sample bi-directional Mendelian randomization \nanalysis, incorporating publicly available GWAS summary statistics. Our aim was to establish a causal \nrelationship between genetically predicted depression and the risk of various female reproductive \npathological conditions, such as ovarian dysfunction, polycystic ovary syndrome(PCOS), ovarian \ncysts, abnormal uterine and vaginal bleeding(AUB), endometriosis, leiomyoma of the uterus, female \ninfertility, spontaneous abortion, eclampsia, pregnancy hypertension, gestational diabetes, excessive \nvomiting in pregnancy, cervical cancer, and uterine/endometrial cancer. We analyzed a substantial \nsample size, ranging from 111,831 to 210,870 individuals, and employed robust statistical methods, \nincluding inverse variance weighted, MR-Egger, weighted median, and MR-PRESSO, to estimate \ncausal effects. Sensitivity analyses, such as Cochran’s Q test, MR-Egger intercept test, MR-PRESSO, \nleave-one-out analysis, and funnel plots, were also conducted to ensure the validity of our results. \nFurthermore, risk factor analyses were performed to investigate potential mediators associated with \nthese observed relationships. Our results demonstrated that genetic predisposition to depression or \ndysthymia was associated with an increased risk of developing PCOS (OR = 1.43, 95% CI 1.28–1.59; \nP = 6.66 ×  10–11), ovarian cysts (OR = 1.36, 95% CI 1.20–1.55; P = 1.57 ×  10–6), AUB (OR = 1.41, 95% \nCI 1.20–1.66; P = 3.01 ×  10–5), and endometriosis (OR = 1.43, 95% CI 1.27–1.70; P = 2.21 ×  10–7) after \nBonferroni correction, but no evidence for reverse causality. Our study did not find any evidence \nsupporting a causal or reverse causal relationship between depression/dysthymia and other types \nof female reproductive disorders. In summary, our study provides evidence for a causal relationship \nbetween genetically predicted depression and specific types of female reproductive disorders. Our \nfindings emphasize the importance of depression management in the prevention and treatment of \nfemale reproductive disorders, notably including PCOS, ovarian cysts, AUB, and endometriosis.\nKeywords Depression or dysthymia, Female reproductive disorders, Mendelian randomization, Causality, \nG WA S\nAbbreviations\nPCOS  Polycystic ovary syndrome\nWHO  World Health Organization\nRCT   Randomized controlled trial\nMR  Mendelian randomization\nOPEN\n1Reproductive Health Department, Shenzhen Traditional Chinese Medicine Hospital, The Fourth Clinical Medical \nCollege of Guangzhou University of Chinese Medicine, Shenzhen 518000, Guangdong, China.  2These authors \ncontributed equally: Shuyi Ling and Yuqing Dai. *email: zhong_zhisheng@sina.com; yuehuizheng@163.com\n\n2\nVol:.(1234567890)Scientific Reports |         (2024) 14:5984  | https://doi.org/10.1038/s41598-024-55993-8\nwww.nature.com/scientificreports/\nGW AS  Genome-wide association study\nAUB  Abnormal uterine and vaginal bleeding\nSNPs  Single nucleotide polymorphisms\nIVW  Inverse variance weighting\nWM  Weighted median\nBMI  Body mass index\nHPA  Hypothalamic–pituitary–adrenal\nHPO  Hypothalamic–pituitary–ovarian\nCRH  Corticotropin-releasing hormone\nDepression stands as the most prevalent psychiatric disorder worldwide. In 2017, the World Health Organiza -\ntion (WHO) reported that over 300 million individuals, accounting for 4.4% of the global population, suffered \nfrom  depression1. From 1990 to 2017, the global incidence of depression has increased 49.86%2. Moreover, it is \nprojected by WHO that depression will emerge as a principal contributor to the global burden of disease by  20303.\nDepression has been found to have associations with various female reproductive disorders. Its prevalence \nhas been estimated to be approximately 31% in patients with  PCOS4, ranging from 11% 5 to 27% 6 and 31.3% \n7 in females with infertility, 15.6% in those with  AUB8, 18.6% in individuals with spontaneous  abortion9, and \n27% in patients with ovarian  cancer10. Moreover, it is noteworthy that depression presents a substantial risk ele-\nment for the onset of gestational diabetes among expectant mothers, exhibiting a correlated augmented risk of \n29%11. Additionally, patients diagnosed with PCOS exhibit 4 times greater likelihood of developing depression in \ncomparison to women without  PCOS12. Furthermore, it is imperative to acknowledge that women have a higher \nprevalence of depression compared to men, with a risk ratio of approximately 2:1 13. This significant difference \nemphasizes the importance of considering the impact of depression on women’s reproductive health. Previous \nstudies primarily relied on observational studies, including case–control  studies14,15 and cross-sectional  studies7,16 \nand cohort  studies9,17. Although these studies provided an estimate of the relationship between depression and \nreproductive status, the causal relationship remains unclear.\nThe traditional design of observational studies comes with inherent limitations, which often challenge the \ninference of causality. Potential mixed bias and reverse causality may lead to biased correlations and  conclusions18. \nFurthermore, conducting randomized controlled trials (RCTs), recognized as the gold standard for establish-\ning causal inference, is unethical and impractical in this case due to the need for substantial human resources, \ntime-consuming follow-up, and the inability to randomly assign depression to different individual groups. To \novercome these limitations, Mendelian randomization (MR) has been increasingly employed to infer credible \ncausal relationships between risk factors and disease  outcomes19. MR utilizes genetic variation, randomly dis -\ntributed during meiosis, as an instrumental variable associated with environmental exposure. This approach \nenables the evaluation of the causal association between depression/dysthymia and different types of female \nreproductive  disorders20. Two-sample bi-directional MR analysis explores both directions of causality, providing \na comprehensive comprehension of the association between exposure and outcome variables. MR studies have \nbeen conducted to explore the causal relationship between depression and  PCOS21,  endometriosis22, and ovar-\nian  cancer23. However, to date, there is a lack of systematic MR studies that have revealed the causal association \nbetween depression/dysthymia and other female reproductive disorders.\nIn this study, we conducted a two-sample bi-directional MR analysis using publicly available genome-wide \nassociation study (GW AS) summary statistics. Our study represents the first comprehensive report of the causal \nrelationships between depression/dysthymia and 14 common female reproductive disorders, including ovarian \ndysfunction, PCOS, ovarian cysts, AUB, endometriosis, leiomyoma of the uterus, female infertility, spontaneous \nabortion, eclampsia, pregnancy hypertension, gestational diabetes, excessive vomiting in pregnancy, cervical \ncancer and uterine/endometrial cancer, through the application of MR analysis. The findings of this investiga -\ntion hold the potential to yield significant insights into the causal links between depression/dysthymia and \nfemale reproductive disorders, consequently offering constructive recommendations for the implementation of \npreventive intervention strategies.\nMethods\nStudy design\nThis study utilized a two-sample bi-directional MR analysis to examine the causal effect of depression or dysthy-\nmia on female reproductive disorders, leveraging by GW AS summary statistics. We also employed instrumental-\nvariable analysis, which emulates a RCT by simulating the random allocation of single nucleotide polymorphisms \n(SNPs) in offsprings.\nTo ensure the robustness of our MR design, we adhered to the guidelines outlined in STROBE-MR24 and care-\nfully evaluated three crucial assumptions. First, the genetic instrument used should strongly predict the exposure \nof interest, as determined by meeting the genome-wide significance threshold (P < 5 ×  10–8) for the instrumental \n variants25. Second, the genetic instruments must be independent of any confounding factors that might influence \nboth the exposure and the outcome of  interest26. At last, it is crucial to establish that the genetic instruments solely \nimpact the outcome through their association with the exposure, rather than through alternative  pathways27.\nIn the reverse MR analysis, we employed a relaxed P threshold (P < 5 ×  10–6) for the instrument-exposure \nassociation in order to include more SNPs for traits with limited SNPs (≤ 3) after linkage disequilibrium (LD) \npruning. This approach has been used in many previous MR  studies28–30. However, it may increase the risk of \nviolating the first assumption of MR design.\n\n3\nVol.:(0123456789)Scientific Reports |         (2024) 14:5984  | https://doi.org/10.1038/s41598-024-55993-8\nwww.nature.com/scientificreports/\nData sources: exposure and outcome variables in GWAS\nThe FinnGen consortium (https:// www. finng en. fi/ fi, accessed on July 10, 2023) provided GW AS data for expo-\nsure (depression or dysthymia: ICD-10 code F3[2, 3]/F341, 48,847 cases & 225,483 controls) and outcomes: \novarian dysfunction (ICD-10 code E28, 2,010 cases & 200,581 controls), PCOS (ICD-10 code E282, 13,142 \ncases & 107,564 controls), ovarian cysts (ICD-10 code N83[0–2], 20,750 cases & 107,564 controls); uterine \nconditions: AUB (ICD-10 code N93, 10,319 cases & 107,564 controls), endometriosis(ICD-10 code N80, 15,088 \ncases & 10,7564 controls), leiomyoma of uterus(ICD-10 code D25, 31,661 cases & 179,209 controls); fertility or \npregnancy-related diseases: female infertility (ICD-10 code N97, 13,142 cases & 107,564 controls), spontaneous \nabortion (ICD-10 code O03, 16,906 cases & 149,622 controls), eclampsia (ICD-10 code O15, 452 cases & 194,266 \ncontrols), pregnancy hypertension (ICD-10 code O10|O11|O13|O14|O15|O16, 14,727 cases & 196,143 controls), \ngestational diabetes (ICD-10 code O244, 13,039 cases & 197,831 controls), excessive vomiting in pregnancy \n(ICD-10 code O21, 2,361 cases & 179,899 controls). The GW AS data from the UK Biobank study (http:// www. \nneale lab. is/ uk- bioba nk/) provided additional outcomes, including cervical cancer (1450 cases & 192,703 controls) \nand uterine/endometrial cancer (906 cases & 193,247 controls). Detailed information about the characteristics \nof the studies and consortia used can be found in Additional file 1: Table S5.\nAs per the International Statistical Classification of Diseases and Related Health Problems 10th Revision, \ndepression or dysthymia is a multifaceted mental health disorder encompassing various conditions such as \ndepressive episode, recurrent depressive disorder, and dysthymia. Depressive episode is characterized by symp-\ntoms such as low mood, reduced energy, decreased activity, loss of interest, and difficulty concentrating. The \nseverity of the symptoms can range from mild to moderate or severe, depending on their number and intensity. \nRecurrent depressive disorder involves repeated episodes of depression without any history of mania, and the \nseverity and duration can vary. Dysthymia, on the other hand, is a chronic form of depression that persists for \nseveral years but does not meet the criteria for recurrent depressive disorder.\nMR analysis\nTo identify the causal relationship between depression/dysthymia and female reproductive disorders, three differ-\nent MR methods, namely random effect inverse variance weighting (IVW), MR-Egger, weighted median (WM), \nand MR-PRESSO were utilized to address heterogeneity of variation and pleiotropic effects. Using multiple esti-\nmators in MR analysis improves the robustness and consistency of our findings by accounting for potential biases \nand uncertainties. Each estimator has unique strengths and limitations and makes different assumptions about \ngenetic instrument validity and pleiotropy, which could affect the accuracy of estimates. By utilizing multiple \nestimators, we can evaluate the sensitivity of our results to different assumptions and increase confidence in the \nvalidity of our findings while mitigating concerns related to underlying assumptions. SNPs and abnormal values \nassociated with female reproductive status, as identified by MR-PRESSO, were  excluded31. IVW served as the \nprimary outcome, while MR-Egger and weighted median were employed to improve the estimation of IVW , as \nthey offer more reliable estimates in a broader range of scenarios, albeit with lower efficiency (wider confidence \nintervals). MR-Egger, although allowing for pleiotropic effects in all genetic variations, assumes that such effects \nare independent of the association between variation and  exposure32. The weighted median method permits the \ninclusion of invalid instruments under the assumption that at least half of the instruments used in MR analysis \nare  valid33. In IVW analysis, the weighted regression slope of the SNP result, showing effect on the SNP exposure \nwith the intercept constrained to zero, represents the estimated outcome. For significant estimates, the MR-Egger \nintercept test and leave-one-out analysis were employed to further assess horizontal pleiotropy. Cochran’s Q test \nwas also used to identify heterogeneity. A funnel plot was utilized to evaluate possible directional pleiotropy, \nakin to assessing publication bias in meta-analysis.\nFurthermore, prior to MR analysis, stringent filtering steps were implemented to ensure SNP quality. Firstly, \nlinkage disequilibrium (LD,  R2 ≥ 0.001 within 10 MB) was aggregated. Secondly, SNPs were required to reach the \ngenome-wide significance threshold of P < 5 ×  10–8 in relation to the relevant exposure. Thirdly, we assessed the \nstrength of our instrument variables using two parameters: the proportion of variance explained  (R2) and the F \nstatistic. The  R2 was calculated as  R2 = β2 × 2 × MAF × (1 − MAF), where β represents the estimated effect and MAF \nindicates the minor allele  frequency34. The F statistic was calculated using the formula F  = [(N – k  −  1)/k] ×  R2/\n(1 −   R2), where N represents the sample size, k represents the number of included SNPs, and  R2 represents the \nproportion of variance explained by the genetic  variants34. The obtained F statistic values ranged from 215 to \n400, as outlined in Additional file 1: Table S1, strongly indicating that the selected genetic variants effectively \nserve as suitable proxies for the investigated  exposure35,36.\nRisk factors\nIn order to investigate the genetic mechanisms that link depression/dysthymia with female reproductive dis-\norders, we conducted MR analyses using depression/dysthymia as exposure and several potential mediators \nas outcomes. These potential mediators included drinking, smoking, coffee intake, body mass index (BMI), \ncirculating leptin levels, obesity, fasting insulin, insulin secretion rate, and  diabetes37–39. GW AS summary data \nfor these potential mediators were obtained from the IEU OpenGW AS database (https:// gwas. mrcieu. ac. uk/, \naccessed on August 2, 2023)40. Detailed information regarding each data source can be found in Table 1. Depres-\nsion/ dysthymia were considered as exposures, while the aforementioned potential risk factors were treated as \noutcomes for Mendelian randomization analysis. The primary results were evaluated based on estimates derived \nfrom IVW . Statistical significance was defined as P < 0.05.\n\n4\nVol:.(1234567890)Scientific Reports |         (2024) 14:5984  | https://doi.org/10.1038/s41598-024-55993-8\nwww.nature.com/scientificreports/\nStatistical analysis\nThe statistical analyses were conducted using the TwoSampleMR package (version 0.5.7) and MRPRESSO pack-\nage (version 1.0) within the R environment (version 4.3.0).To address the issue of multiple testing, a Bonferroni \ncorrection was applied by setting the significance threshold at 0.05 divided by the number of MR estimates (14), \nresulting in a Bonferroni-corrected P-value of 3.57 ×  10−3 . Additionally, associations with a P-value less than 0.05 \nbut not yet meeting the Bonferroni-corrected threshold were considered nominally  significant41.\nEthics approval and consent to participate\nThe data utilized in this study were obtained from publicly available, de-identified sources and were originally \ncollected from participant studies that had already received approval from an ethics committee regarding human \nexperimentation. As a result, no additional ethical approval was necessary for this particular study.\nResults\nMR analysis\nWithin the spectrum of gynecological conditions encompassing ovarian or uterine disorders, the IVW analysis \nrevealed a significant positive association between depression/dysthymia and several conditions, including PCOS \n(OR = 1.43, 95% CI 1.28–1.59; P = 6.66 ×  10–11), ovarian cysts (OR = 1.36, 95% CI 1.20–1.55; P = 1.57 ×  10–6), AUB \n(OR = 1.41, 95% CI 1.20–1.66; P = 3.01 ×  10–5), and endometriosis (OR = 1.47, 95% CI 1.27–1.71; P = 2.21 ×  10–7). \nThese findings were consistent with other MR method results. Additionally, the MR-Egger and WM analyses \nsuggested a nominal correlation between depression/dysthymia and leiomyoma of the uterus. The IVW and \nMR-PRESSO analyses also showed consistent directions, but without statistical significance. However, there \nwas no observed causal relationship between depression/dysthymia and ovarian dysfunction (OR = 1.38, 95% \nCI 0.98–1.94; P = 0.063) (Fig. 1).\nAmong fertility and pregnancy-related diseases, evidence suggested a nominal correlation between depres -\nsion/dysthymia and the risk of gestational diabetes through IVW analyses (OR = 1.22, 95% CI 1.06–1.40; \nP = 0.007). This association has been consistently observed in other MR analyses, except for the MR-EGGER \nanalysis. Furthermore, MR-PRESSO analysis indicated a nominally significant correlation between depression/\ndysthymia and female infertility (OR = 1.15, 95% CI 1.04–1.27; P = 0.016). However, this association was not \nfound to be statistically significant in other MR analysis methods, which showed inconsistent results. In addition, \nmultiple analyses showed that there was no statistically significant association between depression/dysthymia and \nother pregnancy-related conditions such as spontaneous abortion, eclampsia, pregnancy-induced hypertension, \nand hyperemesis gravidarum(Fig. 2).\nIn the context of common reproductive-related cancers, IVW and MR-PRESSO analyses revealed a nominally \nsignificant correlation between depression/dysthymia and cervical cancer, while MR-Egger analysis showed \nthe opposite direction without statistical significance. Furthermore, there was no observed causal relationship \nbetween depression/dysthymia and uterine/endometrial cancer(Fig.  3).\nOur analysis of reverse causality, specifically focusing on depression as the outcome and female reproduc -\ntive status as the exposure, yielded no evidence in support of reverse causality. Among all the factors examined, \nPCOS showed nominal statistical significance in both the IVW analysis and MR PRESSO analysis(see Additional \nfile 1: Table S3). However, it is important to interpret these findings cautiously as they do not provide definitive \nevidence for a causal relationship.\nSensitivity analysis\nTo evaluate the robustness of the aforementioned findings, a series of sensitivity analyses were conducted, includ-\ning Cochran’s Q test, MR-Egger intercept test, and MR-PRESSO global test (Table 2). Heterogeneity was observed \nin the Q test analysis between depression/dysthymia and pregnancy hypertension (Q  = 27.97, P = 0.045), while \nother outcomes did not exhibit heterogeneity. The use of random-effects IVW as the main estimation method \nadequately accounted for acceptable  heterogeneity42. Additionally, excepting P value of leiomyoma of uterus (MR-\nEgger Intercept = -0.04, P = 0.031), P values of the MR-Egger intercept tests from other outcomes were above 0.05, \nindicating the absence of pleiotropic bias in the examined female reproductive disorders, except for leiomyoma \nTable 1.  Data source for risk factors related to female reproductive disorders.\nTraits Category Consortium Sample size Ancestry GW AS ID\nDrinking Binary UK Biobank 360,726 European ukb-d-20117_2\nSmoking Binary UK Biobank 91,353 European ukb-d-22506_111\nCoffee intake Categorical ordered UK Biobank 428,860 European ukb-b-5237\nBMI Continuous UK Biobank 461,460 European ukb-b-19953\nCirculating leptin levels Continuous EBI 21,758 European ebi-a-GCST90012076\nObesity Binary UK Biobank 463,010 European ukb-b-15541\nFasting insulin Continuous EBI 16,386 Hispanic or Latin American ebi-a-GCST90002239\nInsulin secretion rate Continuous EBI 527 European ebi-a-GCST004488\nDiabetes Binary UK Biobank 461,578 East Asian ukb-b-10753\n\n5\nVol.:(0123456789)Scientific Reports |         (2024) 14:5984  | https://doi.org/10.1038/s41598-024-55993-8\nwww.nature.com/scientificreports/\nFigure 1.  Causal effects for depression or dysthymia on ovarian and uterine-related disorders. Summary of \nthe Mendelian randomization (MR) estimates derived from the inverse-variance weighted (IVW), MR-Egger, \nweighted median (WM) and MR-PRESSO methods.\nFigure 2.  Causal effects for depression or dysthymia on fertility and pregnancy-related disorders.\n\n6\nVol:.(1234567890)Scientific Reports |         (2024) 14:5984  | https://doi.org/10.1038/s41598-024-55993-8\nwww.nature.com/scientificreports/\nof uterus(Fig. 4). Furthermore, leave-one-out analysis revealed that no SNP significantly influenced the results, \nand funnel plots displayed symmetrical distributions (Fig.  5; Additional file 2: Figs. S2 and S3), signifying the \nabsence of estimation violations. No heterogeneity was detected in the other analyses. The sensitivity analysis \nresults of the reverse causality analysis are presented in the Additional file 1: Table S4.\nRisk factors analysis\nTo investigate the potential factors that mediate the association between depression/dysthymia and an increased \nrisk of female reproductive pathological conditions, we utilized MR methods to evaluate its impact on several \ncommon risk factors associated with these conditions. The results presented in Table  3 demonstrate that the \ncausal effect of depression or dysthymia on female reproductive disorders remained unaffected by the potential \nrisk factors examined, except for drinking.\nDiscussion\nBased on large-scale GW AS data from the Finngen and UK Biobank, this study employed a variety of MR \napproaches to comprehensively examine the potential causal or reverse association between depression/dys-\nthymia and female reproductive disorders. Our research provides compelling evidence that individuals with \ndepression /dysthymia have a significantly higher risk of developing the following conditions: PCOS (42.9% \nincreased risk), ovarian cysts (36.4% increased risk), AUB (41.2% increased risk), and endometriosis (47.3% \nincreased risk). These results call for more attentions on depression/ dysthymia management and treatment in \nterm of reducing female reproductive diseases such as endometriosis, PCOS and AUB. For instance, integrating \nscreening for depressive symptoms during routine gynecological exams, implementing non-pharmacological \ninterventions such as cognitive-behavioral therapy and exercise, ensuring careful administration of pharmaco-\nlogical treatments under professional guidance, and strengthening social support systems are all crucial com-\nponents in addressing this issue.\nThe utilization of MR in our investigations provides a decreased susceptibility to biases stemming from \nconfounding factors or reverse causation, as compared to observational epidemiological studies. In general, the \nstatistical power of the IVW approach is significantly higher compared to other MR approaches, particularly \nMR-Egger43. Confidence intervals were derived from the same statistical equations used to calculate P val -\nues. Consequently, it is expected that the MR-Egger results, with lower statistical power, would yielded wider \nFigure 3.  Causal effects for depression or dysthymia on reproductive-related cancers.\nTable 2.  Sensitivity analysis of the causal association between depression/dysthymia and the risk of female \nreproductive disorders.\nOutcome Cochran Q value Q test P MR-Egger intercept P MR-PRESSO P value\nOvarian dysfunction 22.437 0.263 0.036 0.495 0.290\nPCOS 29.854 0.054 −0.028 0.095 0.091\nOvarian cysts 27.109 0.102 −0.037 0.134 0.134\nAUB 22.141 0.277 −0.019 0.463 0.292\nEndometriosis 27.213 0.100 0.003 0.133 0.142\nLeiomyoma of uterus 27.331 0.053 −0.037 0.031 0.059\nFemale infertility 11.865 0.891 0.010 0.624 0.894\nSpontaneous abortion 14.608 0.747 0.028 0.119 0.744\nEclampsia 23.613 0.211 0.085 0.451 0.214\nPregnancy hypertension 27.966 0.045 0.017 0.488 0.044\nGestational diabetes 23.247 0.227 −0.009 0.688 0.232\nCervical cancer 18.231 0.441 0.001 0.395 0.461\nUterine/endometrial cancer 16.717 0.543 0.001 0.118 0.549\n\n7\nVol.:(0123456789)Scientific Reports |         (2024) 14:5984  | https://doi.org/10.1038/s41598-024-55993-8\nwww.nature.com/scientificreports/\nconfidence intervals and non-significant P values when compared to IVW in the present study. Thus, IVW was \nprimarily employed as the main method for identifying potentially significant findings. Sensitivity analyses \nand other MR methods were utilized to ensure the robustness of the IVW estimates. If there is any evidence of \nhorizontal pleiotropy, IVW estimates could be biased. In such cases, MR-Egger estimates should be considered \nbecause this method accommodates the unbalanced or directional effects of horizontal pleiotropy across all \n SNPs44. Most MR analyses also require consistent beta directions across all MR  approaches37,45, as is the case in \nour study, which means that the beta coefficients of all MR approaches should consistently be positive or nega-\ntive to obtain a robust conclusion. While it is essential to exercise caution when interpreting causal associations \nderived from MR due to the presence of untestable assumptions inherent to this method, the convergence of our \nestimates across various methodologies and analytical approaches strongly supports the causal involvement of \ndepression or dysthymia in the etiology of female reproductive disorders.\nOur study provides initial evidence indicating that genetically predicted depression/dysthymia may be a \ncausal factor, rather than a consequence, of various female reproductive diseases. Specifically, our results demon-\nstrate significant associations between genetically predicted depression/dysthymia and the following conditions: \nEndometriosis (OR = 1.47, 95% CI 1.27–1.71), PCOS (OR  = 1.43, 95% CI 1.28–1.59), AUB (OR  = 1.41, 95% CI \n1.20–1.66), and Ovarian cysts (OR  = 1.36, 95% CI 1.20–1.55). These findings align with other MR studies that \nhave also suggested depression as a risk factor for  PCOS21 and endometriosis 46.\nIt is important to emphasize that our study did not establish a causal relationship between depression and \nsome other conditions, including ovarian dysfunction, leiomyoma of uterus, female infertility, spontaneous \nabortion, eclampsia, pregnancy hypertension, excessive vomiting in pregnancy, cervical cancer, or uterine/\nendometrial cancer. Additionally, our analysis of reverse causality found no evidence supporting such a reverse \nlink. Previous MR studies have also assessed the causal relationship between depression and ovarian cancer, \nrevealing no significant  association23. Moreover, a cohort study supports our findings by indicating that depres-\nsion is unlikely to be the cause of excessive vomiting in  pregnancy47. However, it should be noted that several \nobservational studies suggest a higher likelihood of depression among patients with ovarian  dysfunction 48, \nfemale  infertility49, leiomyoma of  uterus50,  abortion51, pregnancy  hypertension52,  eclampsia53, cervical  cancer54 \nand endometrial  cancer55. These studies propose that depression may either result from these conditions or con-\ntribute to their development. The discrepancies observed between the results of MR studies and observational \nstudies, as well as the controversies within the latter, can be attributed to confounding factors and biases inherent \nin real-world epidemiological studies. Notably, MR, which functions as an analogous approach to randomized \nFigure 4.  Scatter plots depicting the impact of genetically predicted depression/dysthymia on the risk of female \nreproductive disorders.\n\n8\nVol:.(1234567890)Scientific Reports |         (2024) 14:5984  | https://doi.org/10.1038/s41598-024-55993-8\nwww.nature.com/scientificreports/\ncontrolled trials, emerges as a more effective tool for drawing causal inferences due to its reduced susceptibility \nto confounding  influences23.\nAs indicated by the risk factor analyses, drinking behaviors may play a role in the susceptibility of female \nreproductive disorders linked to depression or dysthymia. Research has demonstrated a positive association \nbetween alcohol dependence and depression, indicating the potential involvement of interconnected neuro-\nbiological  mechanisms56,57. It is widely recognized that alcohol negatively affects female  reproduction38. How-\never, whether depression influences women’s reproductive health through similar neural mechanisms remains \nuncertain.\nNumerous mechanisms have been proposed to elucidate the impact of depression on female reproduc-\ntive status. It is widely believed that depression exerts its influence on female reproduction through the \nFigure 5.  Funnel plots depicting the impact of genetically predicted depression or dysthymia on the risk \nof female reproductive disorders. The funnel plots show the Inverse variance weighted and MR-Egger MR \nestimate of each depression/dysthymia single-nucleotide polymorphism with different female reproductive \ndisorders versus 1/standard error (1/SEIV). (a) PCOS; (b) ovarian cysts; (c) AUB; (d) endometriosis; (e) \nleiomyoma of the uterus; (f) gestational diabetes.\nTable 3.  Risk factors analysis.\nOutcomes\nIVW Causal effect (95% \nCI) P value Heterogeneity Q value P value MR-Egger Intercept P value\nDrinking 1.016(1.007–1.025) 0.001 41.833 0.001 −0.001 0.672\nSmoking 0.999 (0.991–1.008) 0.885 22.509 0.210 −0.002 0.358\nCoffee intake 1.024 (0.992–1.057) 0.141 57.301 2.91 ×  10−6 0.009 0.202\nBMI 0.923 (0.848–1.004) 0.061 214.664 2.19 ×  10−37 0.003 0.872\nCirculating leptin levels 0.927 (0.811–1.061) 0.272 19.599 0.188 −0.033 0.197\nObesity 0.999 (0.996–1.003) 0.687 36.406 0.003 4.56 ×  10−6 0.995\nFasting insulin 0.99 (0.922–1.062) 0.771 18.934 0.396 −0.003 0.820\nInsulin secretion rate 0.67 (0.433–1.038) 0.073 13.667 0.691 −0.027 0.776\nDiabetes 0.994 (0.983–1.005) 0.273 52.285 1.20 ×  10−6 −0.001 0.629\n\n9\nVol.:(0123456789)Scientific Reports |         (2024) 14:5984  | https://doi.org/10.1038/s41598-024-55993-8\nwww.nature.com/scientificreports/\nhypothalamic–pituitary–adrenal (HPA) axis and the hypothalamic-pituitary-ovarian (HPO)  axis58–61.Cortico-\ntropin-releasing hormone (CRH), originating from the hypothalamus, is implicated in various reproductive \nprocesses, including follicular development, ovulation, and luteolysis in the ovarian  CRH62,63. Furthermore, \nrecent findings indicate that CRH-R1 is expressed in reproductive tissues such as the ovary, endometrium, and \nmyometrium, and plays a pivotal role in regulating reproductive  functions62–64.In addition, abnormal lactic acid \nmetabolism and glycolysis may serve as a link between depression and reproductive diseases. Studies conducted \non mice have demonstrated that the modulation of lactic acid homeostasis can influence neuronal excitability and \ndepression-like  behavior65. Associations have been found between lactic acid and uterine  remodeling66, abnormal \nglycolysis and ovarian  cancer67, as well as oocyte quality of PCOS  patients68.Moreover, depression often coex-\nists with an imbalance in intestinal  flora69–71. Disruptions in gastrointestinal ecology actively contribute to the \ndevelopment and metastasis of gynecological tumors, such as cervical cancer, endometrial cancer, and ovarian \n cancer72. Notably, several studies have revealed that the use of probiotics can ameliorate depressive symptoms \nand regulate sex hormone levels, offering potential therapeutic benefits for women with PCOS and gestational \n diabetes73–75.Besides, chronic inflammation and oxidative stress are prominent features of  depression76,77 and \npathological conditions pertaining to female reproductive health, including PCOS, ovarian dysfunction, endo -\nmetriosis, gestational diabetes, and leiomyoma of  uterus78–82. Inflammatory processes are intertwined with the \nonset of depression, which can further exacerbate the inflammatory response and detrimentally impact the \nreproductive  system83–85.\nHowever, it is essential to acknowledge that our study possesses several inherent limitations that necessitate \ncautious interpretation. Firstly, the generalizability of our findings to diverse ethnic groups with distinct life-\nstyles and cultural backgrounds may be limited, as our study exclusively focused on individuals of European \nancestry. Secondly, it is essential to recognize the inherent challenges of MR analyses, which rely on the random \nallocation of genetic variants, in fully disentangling mediation from pleiotropy. Plausible scenarios exist wherein \ngenetic variants within our genome may exert simultaneous influences on multiple phenotypes. Furthermore, \nthe absence of significant findings in our study can be attributed to statistical power constraints and inadequate \nrepresentation of SNPs. The persisting issue of \"missing heritability\" in various polygenic diseases and traits, \nparticularly psychiatric disorders, may be addressed through ongoing research utilizing SRS and LRS technolo-\ngies to explore rare  variants86. Consequently, our ability to draw definitive conclusions regarding true causal \nrelationships is impeded. Although we attempted to enhance sensitivity by relaxing the exposure threshold in \nour reverse causality MR analysis, the limited number of strongly associated SNPs analyzed may result in reduced \nstatistical power to detect significant associations or limit the generalizability of the findings. Given the inher -\nent limitations of the Finngen and UK Biobank datasets, such as our inability to access participants’ individual \ndata, it is imperative that future studies are conducted to validate causal relationships and explore underlying \nmechanisms. Such investigations are crucial for generating meaningful clinical recommendations that accurately \ninform medical practice.\nIn conclusion, utilizing extensive genetic summary data, our study provides strengthened evidence support-\ning a causal link between depression/dysthymia and the risk of specific female reproductive disorders, including \nendometriosis, PCOS, AUB, ovarian cysts and gestational diabetes. However, the reverse causal relationship \nbetween these conditions and depression remains undetermined. These findings highlight the significance of \nmental health in both the prevention and treatment of female reproductive disorders. While our results align \nwith previous observational studies to some extent, further validation through larger prospective studies and \nin-depth mechanistic investigations is necessary to comprehensively elucidate the causal relationship between \ndepression and various types of reproductive conditions.\nData availability\nAll data are publicly available. The data sources for this study include the FinnGen consortium (https:// www.  \nfinng en. fi/ fi), the UK Biobank (http:// www. neale lab. is/ uk- bioba nk/), and the IEU OpenGW AS database (https:// \ngwas. mrcieu. ac. uk/) .\nReceived: 1 September 2023; Accepted: 29 February 2024\nReferences\n 1. World Health Organization. Depression and Other Common Mental Disorders: Global Health Estimates (WHO, 2017).\n 2. Liu, Q. et al. Changes in the global burden of depression from 1990 to 2017: Findings from the global burden of disease study. J. \nPsychiatr. Res. 126, 134–140. https:// doi. org/ 10. 1016/j. jpsyc hires. 2019. 08. 002 (2020).\n 3. World Health Organization. The Global Burden of Disease: 2004 Update. (WHO, 2008).\n 4. Dybciak, P . et al. Depression in polycystic ovary syndrome: A systematic review and meta-analysis. J. Clin. Med. 12(20), 1–16. \nhttps:// doi. org/ 10. 3390/ jcm12 206446 (2023).\n 5. Volgsten, H., Skoog Svanberg, A., Ekselius, L., Lundkvist, O. & Sundstrom, P . I. Prevalence of psychiatric disorders in infertile \nwomen and men undergoing in vitro fertilization treatment. Hum. Reprod. 23(9), 2056–2063. https://  doi. org/ 10. 1093/ humrep/ \nden154 (2008).\n 6. Chen, T. H., Chang, S. P ., Tsai, C. F . & Juang, K. D. Prevalence of depressive and anxiety disorders in an assisted reproductive \ntechnique clinic. Hum. Reprod. 19(10), 2313–2318. https://  doi. org/ 10. 1093/ humrep/ deh414 (2004).\n 7. Wang, L., Tang, Y . & Wang, Y . Predictors and incidence of depression and anxiety in women undergoing infertility treatment: A \ncross-sectional study. PLoS One 18(4), e0284414. https:// doi. org/ 10. 1371/ journ al. pone. 02844 14 (2023).\n 8. Kayhan, F ., Alptekin, H. & Kayhan, A. Mood and anxiety disorders in patients with abnormal uterine bleeding. Eur. J. Obstet. \nGynecol. Reprod. Biol. 199, 192–197. https:// doi. org/ 10. 1016/j. ejogrb. 2016. 02. 033 (2016).\n 9. Kulathilaka, S., Hanwella, R. & de Silva, V . A. Depressive disorder and grief following spontaneous abortion. BMC Psychiatry 16, \n100. https:// doi. org/ 10. 1186/ s12888- 016- 0812-y (2016).\n\n10\nVol:.(1234567890)Scientific Reports |         (2024) 14:5984  | https://doi.org/10.1038/s41598-024-55993-8\nwww.nature.com/scientificreports/\n 10. Ghamari, D. et al. The prevalence of depression and anxiety in women with ovarian cancer: An updated systematic review and \nmeta-analysis of cross-sectional studies. Asian Pac. J. Cancer Prev.  24(10), 3315–3325. https://  doi. org/ 10. 31557/ APJCP . 2023. 24. \n10. 3315 (2023).\n 11. Zhang, C., Jing, L. & Wang, J. Does depression increase the risk of gestational diabetes mellitus? A systematic review and meta-\nanalysis. Pak. J. Med. Sci. 39(1), 285–292. https:// doi. org/ 10. 12669/ pjms. 39.1. 6845 (2023).\n 12. Zhuang, J., Wang, X., Xu, L., Wu, T. & Kang, D. Antidepressants for polycystic ovary syndrome. Cochrane Database Syst Rev.  \n2013(5), CD008575. https:// doi. org/ 10. 1002/ 14651 858. CD008 575. pub2 (2013).\n 13. Kessler, R. C. Epidemiology of women and depression. J. Affect. Disord. 74(1), 5–13. https:// doi. org/ 10. 1016/ s0165- 0327(02) 00426-3 \n(2003).\n 14. Basirat, Z. et al. Stress, depression, sexual function, and alexithymia in infertile females with and without polycystic ovary syn-\ndrome: A case–control study. Int. J. Fertil. Steril. 13(3), 203–208. https:// doi. org/ 10. 22074/ ijfs. 2019. 5703 (2019).\n 15. Ahmadi, M. et al. Mental and personality disorders in infertile women with polycystic ovary: A case–control study. Afr. Health \nSci. 20(3), 1241–1249. https:// doi. org/ 10. 4314/ ahs. v20i3. 28 (2020).\n 16. Asdaq, S. M. B. et al. Impact of polycystic ovary syndrome on eating behavior, depression and health related quality of life: A \ncross-sectional study in Riyadh. Saudi J. Biol. Sci. 27(12), 3342–3347. https://  doi. org/ 10. 1016/j. sjbs. 2020. 08. 039 (2020).\n 17. Jacob, L., Polly, I., Kalder, M. & Kostev, K. Prevalence of depression, anxiety, and adjustment disorders in women with spontaneous \nabortion in Germany—A retrospective cohort study. Psychiatry Res.  258, 382–386. https:// doi. org/ 10. 1016/j. psych res. 2017. 08. 064 \n(2017).\n 18. Sekula, P ., Del Greco, M. F ., Pattaro, C. & Kottgen, A. Mendelian randomization as an approach to assess causality using observa-\ntional data. J. Am. Soc. Nephrol. 27(11), 3253–3265. https://  doi. org/ 10. 1681/ ASN. 20160 10098 (2016).\n 19. Richmond, R. C. & Davey, S. G. Mendelian randomization: Concepts and scope. Cold Spring Harb. Perspect. Med. 12(1), a040501. \nhttps:// doi. org/ 10. 1101/ cshpe rspect. a0405 01 (2022).\n 20. Burgess, S. & Thompson, S. G. Multivariable Mendelian randomization: The use of pleiotropic genetic variants to estimate causal \neffects. Am. J. Epidemiol. 181(4), 251–260. https://  doi. org/ 10. 1093/ aje/ kwu283 (2015).\n 21. Zhu, T. & Goodarzi, M. O. Causes and consequences of polycystic ovary syndrome: Insights from Mendelian randomization. J. \nClin. Endocrinol. Metab. 107(3), e899–e911. https://  doi. org/ 10. 1210/ clinem/ dgab7 57 (2022).\n 22. McGrath, I. M., Montgomery, G. W . & Mortlock, S. Insights from Mendelian randomization and genetic correlation analyses into \nthe relationship between endometriosis and its comorbidities. Hum. Reprod. Update  29(5), 655–674. https:// doi. org/ 10. 1093/ \nhumupd/ dmad0 09 (2023).\n 23. Zhu, G. L. et al. Causal relationship between genetically predicted depression and cancer risk: A two-sample bi-directional Men-\ndelian randomization. BMC Cancer 22(1), 353. https:// doi. org/ 10. 1186/ s12885- 022- 09457-9 (2022).\n 24. Skrivankova, V . W . et al. Strengthening the reporting of observational studies in epidemiology using Mendelian randomization: \nThe Strobe-Mr statement. JAMA. 326(16), 1614–1621. https://  doi. org/ 10. 1001/ jama. 2021. 18236 (2021).\n 25. Harrison, R., Munafo, M. R., Davey Smith, G. & Wootton, R. E. Examining the effect of smoking on suicidal ideation and attempts: \nTriangulation of epidemiological approaches. Br. J. Psychiatry 217(6), 701–707. https://  doi. org/ 10. 1192/ bjp. 2020. 68 (2020).\n 26. Pasman, J. A. et al. Gwas of lifetime cannabis use reveals new risk loci, genetic overlap with psychiatric traits, and a causal influence \nof schizophrenia. Nat. Neurosci. 21(9), 1161–1170. https:// doi. org/ 10. 1038/ s41593- 018- 0206-1 (2018).\n 27. Boef, A. G., Dekkers, O. M. & le Cessie, S. Mendelian randomization studies: A review of the approaches used and the quality of \nreporting. Int. J. Epidemiol. 44(2), 496–511. https:// doi. org/ 10. 1093/ ije/ dyv071 (2015).\n 28. Mavromatis, L. A. et al. Association between brain structure and alcohol use behaviors in adults: A Mendelian randomization and \nmultiomics study. JAMA Psychiatry 79(9), 869–878. https:// doi. org/ 10. 1001/ jamap sychi atry. 2022. 2196 (2022).\n 29. Chen, X. et al. Depression and prostate cancer risk: A Mendelian randomization study. Cancer Med. 9(23), 9160–9167. https://  \ndoi. org/ 10. 1002/ cam4. 3493 (2020).\n 30. Vaucher, J. et al. Cannabis use and risk of schizophrenia: A Mendelian randomization study. Mol. Psychiatry  23(5), 1287–1292. \nhttps:// doi. org/ 10. 1038/ mp. 2016. 252 (2018).\n 31. Verbanck, M., Chen, C. Y ., Neale, B. & Do, R. Detection of widespread horizontal pleiotropy in causal relationships inferred from \nMendelian randomization between complex traits and diseases. Nat. Genet. 50(5), 693–698. https:// doi. org/ 10. 1038/ s41588- 018- \n0099-7 (2018).\n 32. Bowden, J., Davey Smith, G. & Burgess, S. Mendelian randomization with invalid instruments: Effect estimation and bias detection \nthrough Egger regression. Int. J. Epidemiol. 44(2), 512–525. https:// doi. org/ 10. 1093/ ije/ dyv080 (2015).\n 33. Bowden, J., Davey Smith, G., Haycock, P . C. & Burgess, S. Consistent estimation in mendelian randomization with some invalid \ninstruments using a weighted median estimator. Genet. Epidemiol. 40(4), 304–314. https:// doi. org/ 10. 1002/ gepi. 21965 (2016).\n 34. Xia, J. W . et al. Both indirect maternal and direct fetal genetic effects reflect the observational relationship between higher birth \nweight and lower adult bone mass. BMC Med. 20(1), 361. https:// doi. org/ 10. 1186/ s12916- 022- 02531-w (2022).\n 35. Wei, Y ., Zhan, Y ., Lofvenborg, J. E., Tuomi, T. & Carlsson, S. Birthweight, BMI in adulthood and latent autoimmune diabetes in \nadults: A Mendelian randomisation study. Diabetologia. 65(9), 1510–1518. https:// doi. org/ 10. 1007/ s00125- 022- 05725-2 (2022).\n 36. Luo, J., Xu, Z., Noordam, R., van Heemst, D. & Li-Gao, R. Depression and inflammatory bowel disease: A bidirectional two-sample \nMendelian randomization study. J. Crohns Colitis 16(4), 633–642. https:// doi. org/ 10. 1093/ ecco- jcc/ jjab1 91 (2022).\n 37. Venkatesh, S. S. et al. Obesity and risk of female reproductive conditions: A Mendelian randomisation study. PLoS Med. 19(2), \ne1003679. https:// doi. org/ 10. 1371/ journ al. pmed. 10036 79 (2022).\n 38. de Angelis, C. et al. Smoke, alcohol and drug addiction and female fertility. Reprod. Biol. Endocrinol. 18(1), 21. https:// doi. org/ 10. \n1186/ s12958- 020- 0567-7 (2020).\n 39. Soylu, L. I. et al. Coffee, tea and caffeine consumption and risk of primary infertility in women: A Danish cohort study. Acta Obstet. \nGynecol. Scand. 97(5), 570–576. https:// doi. org/ 10. 1111/ aogs. 13307 (2018).\n 40. Elsworth, B., Lyon, M., Alexander, T., Liu, Y ., Matthews, P ., Hallett, J., Bates, P ., Palmer, T., Haberland, V ., Smith, G.D., Zheng, J., \nHaycock, P ., Gaunt, T.R. & Hemani, G. The Mrc Ieu Opengwas data infrastructure. Biorxiv 2020.08.10.244293v1 https:// doi. org/ \n10. 1101/ 2020. 08. 10. 244293 (2020).\n 41. Vuckovic, D. et al. The polygenic and monogenic basis of blood traits and diseases. Cell 182(5), 1214-1231e1211. https:// doi. org/ \n10. 1016/j. cell. 2020. 08. 008 (2020).\n 42. Burgess, S. et al. Guidelines for performing Mendelian randomization investigations. Wellcome Open Res. 4, 186. https:// doi. org/ \n10. 12688/ wellc omeop enres. 15555.2 (2019).\n 43. Lin, Z., Deng, Y . & Pan, W . Combining the strengths of inverse-variance weighting and Egger regression in Mendelian randomiza-\ntion using a mixture of regressions model. PLoS Genet. 17(11), e1009922. https:// doi. org/ 10. 1371/ journ al. pgen. 10099 22 (2021).\n 44. Bowden, J. et al. Assessing the suitability of summary data for two-sample Mendelian randomization analyses using Mr-Egger \nregression: The role of the I2 statistic. Int. J. Epidemiol. 45(6), 1961–1974. https:// doi. org/ 10. 1093/ ije/ dyw220 (2016).\n 45. Chen, X. et al. Causal relationship between physical activity, leisure sedentary behaviors and Covid-19 risk: A Mendelian rand-\nomization study. J. Transl. Med. 20(1), 216. https:// doi. org/ 10. 1186/ s12967- 022- 03407-6 (2022).\n 46. Koller, D. et al. Epidemiologic and genetic associations of endometriosis with depression, anxiety, and eating disorders. JAMA \nNetw. Open 6(1), e2251214. https:// doi. org/ 10. 1001/ jaman etwor kopen. 2022. 51214 (2023).\n\n11\nVol.:(0123456789)Scientific Reports |         (2024) 14:5984  | https://doi.org/10.1038/s41598-024-55993-8\nwww.nature.com/scientificreports/\n 47. Kjeldgaard, H. K., Eberhard-Gran, M., Benth, J. S., Nordeng, H. & Vikanes, A. V . History of depression and risk of hyperemesis \ngravidarum: A population-based cohort study. Arch. Womens Ment. Health 20(3), 397–404. https:// doi. org/ 10. 1007/ s00737- 016- \n0713-6 (2017).\n 48. Ates, S., Aydin, S., Ozcan, P ., Bakar, R. Z. & Cetin, C. Sleep, depression, anxiety and fatigue in women with premature ovarian \ninsufficiency. J. Psychosom. Obstet. Gynaecol. 43(4), 482–487. https:// doi. org/ 10. 1080/ 01674 82X. 2022. 20690 08 (2022).\n 49. Nik Hazlina, N. H., Norhayati, M. N., Shaiful Bahari, I. & Nik Muhammad Arif, N. A. Worldwide prevalence, risk factors and \npsychological impact of infertility among women: A systematic review and meta-analysis. BMJ Open. 12(3), e057132. https:// doi. \norg/ 10. 1136/ bmjop en- 2021- 057132 (2022).\n 50. Shen, T. C., Y ang, C. Y ., Huang, Y . J., Lin, C. L. & Sung, F . C. Risk of depression in patients with uterine leiomyoma: A nationwide \npopulation-based cohort study. J. Affect. Disord. 213, 126–130. https:// doi. org/ 10. 1016/j. jad. 2017. 02. 020 (2017).\n 51. Farren, J. et al. Posttraumatic stress, anxiety and depression following miscarriage and ectopic pregnancy: A multicenter, prospec-\ntive, cohort study. Am. J. Obstet. Gynecol. 222(4), 367e361–367e322. https:// doi. org/ 10. 1016/j. ajog. 2019. 10. 102 (2020).\n 52. Packer, C. H., Pilliod, R. A., Chatroux, L. R., Caughey, A. B. & Valent, A. M. Increased rates of adverse perinatal outcomes in women \nwith gestational diabetes and depression. J. Matern. Fetal Neonatal Med. 34(23), 3862–3866. https:// doi. org/ 10. 1080/ 14767 058. \n2019. 17016 47 (2021).\n 53. Postma, I. R., Bouma, A., Ankersmit, I. F . & Zeeman, G. G. Neurocognitive functioning following preeclampsia and eclampsia: A \nlong-term follow-up study. Am. J. Obstet. Gynecol. 211(1), 37e31-39. https:// doi. org/ 10. 1016/j. ajog. 2014. 01. 042 (2014).\n 54. Y ang, C. M. et al. Anxiety and depression risk in Taiwan women with breast cancer and cervical cancer. Front. Oncol. 12, 946029. \nhttps:// doi. org/ 10. 3389/ fonc. 2022. 946029 (2022).\n 55. Chen, C. Y . et al. Risk of depression following uterine cancer: A nationwide population-based study. Psychooncology  26(11), \n1770–1776. https:// doi. org/ 10. 1002/ pon. 4360 (2017).\n 56. Wei, X. et al. The neuropeptide Y single-nucleotide polymorphism Rs16147:T>C moderates the effect of alcohol dependence on \ndepression in male Chinese Han population. Front. Psychiatry 13, 1012850. https:// doi. org/ 10. 3389/ fpsyt. 2022. 10128 50 (2022).\n 57. Li, J. et al. Inhibition of Ampa receptor and Camkii activity in the lateral habenula reduces depressive-like behavior and alcohol \nintake in rats. Neuropharmacology 126, 108–120. https:// doi. org/ 10. 1016/j. neuro pharm. 2017. 08. 035 (2017).\n 58. Zhai, Q. Y ., Wang, J. J., Tian, Y ., Liu, X. & Song, Z. Review of psychological stress on oocyte and early embryonic development in \nfemale mice. Reprod. Biol. Endocrinol. 18(1), 101. https:// doi. org/ 10. 1186/ s12958- 020- 00657-1 (2020).\n 59. Bangasser, D. A. & Valentino, R. J. Sex differences in stress-related psychiatric disorders: neurobiological perspectives. Front. \nNeuroendocrinol. 35(3), 303–319. https:// doi. org/ 10. 1016/j. yfrne. 2014. 03. 008 (2014).\n 60. Pariante, C. M. & Lightman, S. L. The Hpa axis in major depression: Classical theories and new developments. Trends Neurosci. \n31(9), 464–468. https:// doi. org/ 10. 1016/j. tins. 2008. 06. 006 (2008).\n 61. Wagner, M. et al. Hair cortisol concentration in healthy children and adolescents is related to puberty, age, gender, and body mass \nindex. Horm. Res. Paediatr. 92(4), 237–244. https:// doi. org/ 10. 1159/ 00050 4914 (2019).\n 62. Hillhouse, E. W . & Grammatopoulos, D. K. The molecular mechanisms underlying the regulation of the biological activity of \ncorticotropin-releasing hormone receptors: Implications for physiology and pathophysiology. Endocr. Rev. 27(3), 260–286. https:// \ndoi. org/ 10. 1210/ er. 2005- 0034 (2006).\n 63. Petraglia, F ., Imperatore, A. & Challis, J. R. Neuroendocrine mechanisms in pregnancy and parturition. Endocr. Rev. 31(6), 783–816. \nhttps:// doi. org/ 10. 1210/ er. 2009- 0019 (2010).\n 64. Whirledge, S. & Cidlowski, J. A. Glucocorticoids and reproduction: Traffic control on the road to reproduction. Trends Endocrinol. \nMetab. 28(6), 399–415. https:// doi. org/ 10. 1016/j. tem. 2017. 02. 005 (2017).\n 65. Y ao, S. et al. Astrocytic lactate dehydrogenase a regulates neuronal excitability and depressive-like behaviors through lactate \nhomeostasis in mice. Nat. Commun. 14(1), 729. https:// doi. org/ 10. 1038/ s41467- 023- 36209-5 (2023).\n 66. Wu, H., Huang, H. & Zhao, Y . Interplay between metabolic reprogramming and post-translational modifications: From glycolysis \nto lactylation. Front. Immunol. 14, 1211221. https:// doi. org/ 10. 3389/ fimmu. 2023. 12112 21 (2023).\n 67. Wu, Z. et al. Paris saponins VII inhibits glycolysis of ovarian cancer via the Rorc/Ack1 signaling pathway. Biochem. Pharmacol.  \n213, 115597. https:// doi. org/ 10. 1016/j. bcp. 2023. 115597 (2023).\n 68. Zhao, Y . K. et al. Correlation between abnormal energy metabolism of ovarian granulosa cells and in vitro fertilization-embryo \ntransfer outcomes in patients with polycystic ovary syndrome and obesity. J. Ovarian Res. 16(1), 145. https:// doi. org/ 10. 1186/ \ns13048- 023- 01204-3 (2023).\n 69. Yu, T., Ding, Y ., Qian, D., Lin, L. & Tang, Y . Characteristics of fecal microbiota in different constipation subtypes and association \nwith colon physiology, lifestyle factors, and psychological status. Ther. Adv. Gastroenterol. 16, 17562848231154100. https:// doi. \norg/ 10. 1177/ 17562 84823 11541 01 (2023).\n 70. Zhu, F ., Tu, H. & Chen, T. The microbiota–gut–brain axis in depression: The potential pathophysiological mechanisms and micro-\nbiota combined antidepression effect. Nutrients 14(10), 1–15. https:// doi. org/ 10. 3390/ nu141 02081 (2022).\n 71. Yu, T., Ding, Y ., Qian, D., Lin, L. & Tang, Y . Characteristics of fecal microbiota in different constipation subtypes and association \nwith colon physiology, lifestyle factors, and psychological status. Ther. Adv. Gastroenterol. 16, 1–16. https:// doi. org/ 10. 1177/ 17562 \n84823 11541 01 (2023).\n 72. Korczynska, L. et al. The role of microbiota in the pathophysiology of uterine fibroids—A systematic review. Front. Cell Infect. \nMicrobiol. 13, 1177366. https:// doi. org/ 10. 3389/ fcimb. 2023. 11773 66 (2023).\n 73. Huang, R., Wang, K. & Hu, J. Effect of probiotics on depression: A systematic review and meta-analysis of randomized controlled \ntrials. Nutrients 8(8), 483. https:// doi. org/ 10. 3390/ nu808 0483 (2016).\n 74. Zhang, J. et al. Probiotic Bifidobacterium lactis V9 regulates the secretion of sex hormones in polycystic ovary syndrome patients \nthrough the gut–brain axis. mSystems 4(2), 19. https:// doi. org/ 10. 1128/ mSyst ems. 00017- 19 (2019).\n 75. Hasain, Z. et al. Gut microbiota and gestational diabetes mellitus: A review of host–gut microbiota interactions and their therapeutic \npotential. Front. Cell Infect. Microbiol. 10, 188. https:// doi. org/ 10. 3389/ fcimb. 2020. 00188 (2020).\n 76. Vavakova, M., Durackova, Z. & Trebaticka, J. Markers of oxidative stress and neuroprogression in depression disorder. Oxid. Med. \nCell Longev. 2015, 898393. https:// doi. org/ 10. 1155/ 2015/ 898393 (2015).\n 77. Lindqvist, D. et al. Oxidative stress, inflammation and treatment response in major depression. Psychoneuroendocrinology  76, \n197–205. https:// doi. org/ 10. 1016/j. psyne uen. 2016. 11. 031 (2017).\n 78. Siddiqui, S., Mateen, S., Ahmad, R. & Moin, S. A brief insight into the etiology, genetics, and immunology of polycystic ovarian \nsyndrome (PCOS). J. Assist. Reprod. Genet. 39(11), 2439–2473. https://  doi. org/ 10. 1007/ s10815- 022- 02625-7 (2022).\n 79. Sanverdi, I., Kilicci, C., Cogendez, E., Abide Y ayla, C. & Ozkaya, E. Utility of complete blood count parameters to detect premature \novarian insufficiency in cases with oligomenorrhea/amenorrhea. J. Clin. Lab. Anal. 32(5), e22372. https:// doi. org/ 10. 1002/ jcla. \n22372 (2018).\n 80. Jiang, L., Y an, Y ., Liu, Z. & Wang, Y . Inflammation and endometriosis. Front. Biosci. (Landmark Ed). 21(5), 941–948. https://  doi. \norg/ 10. 2741/ 4431 (2016).\n 81. Pinto, Y . et al. Gestational diabetes is driven by microbiota-induced inflammation months before diagnosis. Gut. 72(5), 918–928. \nhttps:// doi. org/ 10. 1136/ gutjnl- 2022- 328406 (2023).\n 82. Duan, Y . et al. Correlation between platelet–lymphocyte ratio and neutrophil–lymphocyte ratio in patients with uterine leiomyoma: \nA cross-sectional study. J. Oncol. 2022, 3257887. https:// doi. org/ 10. 1155/ 2022/ 32578 87 (2022).\n\n12\nVol:.(1234567890)Scientific Reports |         (2024) 14:5984  | https://doi.org/10.1038/s41598-024-55993-8\nwww.nature.com/scientificreports/\n 83. Dantzer, R., O’Connor, J. C., Freund, G. G., Johnson, R. W . & Kelley, K. W . From inflammation to sickness and depression: When \nthe immune system subjugates the brain. Nat. Rev. Neurosci. 9(1), 46–56. https:// doi. org/ 10. 1038/ nrn22 97 (2008).\n 84. Rudnicka, E. et al. Chronic low grade inflammation in pathogenesis of PCOS. Int. J. Mol. Sci. 22(7), 1–12. https:// doi. org/ 10. 3390/ \nijms2 20737 89 (2021).\n 85. Y ang, Z. et al. Controlling chronic low-grade inflammation to improve follicle development and survival. Am. J. Reprod. Immunol. \n84(2), e13265. https:// doi. org/ 10. 1111/ aji. 13265 (2020).\n 86. Owen, M. J. & Williams, N. M. Explaining the missing heritability of psychiatric disorders. World Psychiatry  20(2), 294–295. \nhttps:// doi. org/ 10. 1002/ wps. 20870 (2021).\nAcknowledgements\nWe wish to acknowledge the participants and investigators of the UK Biobank, the FinnGen consortium and \nIEU-OpenGW AS project.\nAuthor contributions\nThe study was designed by Y .H.Z., S.Y .L., and Y .Q.D. S.Y .L. and Y .Q.D. were responsible for the initial draft of \nthe manuscript and verification of the underlying data. Statistical analyses were conducted by S.Y .L., Y .Q.D., \nand R.X.W . Y .L., W .B.W ., and Z.Q.Z. contributed to data acquisition and analysis. Z.S.Z. and Y .H.Z. participated \nin the interpretation of the data. All authors have reviewed and approved the final version of the manuscript.\nFunding\nThis work was supported by Basic Research Scheme of Shenzhen Science and Technology Innovation Commi\nssion(JCYJ20230807094815031, JCYJ20220531092208018);the National Nature Science Foundation of China \n(No. 81671455).\nCompeting interests \nThe authors declare no competing interests.\nAdditional information\nSupplementary Information The online version contains supplementary material available at https:// doi. org/ \n10. 1038/ s41598- 024- 55993-8.\nCorrespondence and requests for materials should be addressed to Z.Z. or Y .Z.\nReprints and permissions information is available at www.nature.com/reprints.\nPublisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and \ninstitutional affiliations.\nOpen Access  This article is licensed under a Creative Commons Attribution 4.0 International \nLicense, which permits use, sharing, adaptation, distribution and reproduction in any medium or \nformat, as long as you give appropriate credit to the original author(s) and the source, provide a link to the \nCreative Commons licence, and indicate if changes were made. The images or other third party material in this \narticle are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the \nmaterial. If material is not included in the article’s Creative Commons licence and your intended use is not \npermitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from \nthe copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.\n© The Author(s) 2024","source_license":"CC0","license_restricted":false}