A Mendelian randomization study on causal relationship between metabolic factors and abnormal spermatozoa

Introduction

Male infertility is a global health issue afflicting up to 12% of men worldwide, with a primary or contributing cause in approximately 50% of infertile couples (1,2). An updated systematic review confirms a severe decrease of 51.6% in testicular sperm production among men from various continents between 1973 and 2018. And this decline is even continuing in the 21st century at an accelerated pace globally (3). The dramatically increasing disease burden of abnormal spermatozoa reflects the decline in male fertility worldwide, which imposes psychological and social pressure on patients and weighs on economic burden of health-care systems (4). Research on the causes of abnormal spermatozoa is urgently needed to develop appropriate interventions for potential risk factors, to monitor access to quality fertility care, and eventually to mitigate the crisis in male reproductive health.

Previous observational or epidemiological studies on abnormal spermatozoa have put a spotlight on metabolic factors, including dietary habit, endocrine diseases and obesity epidemics (5,6). Further, researches on transgenerational transmission of epigenetic modifications suggest that state of impaired metabolism, such as obesity or diabetes, might have significant effects on male fertility and even compromise reproductive potential of offspring (7,8). However, since potential residual confounding and reverse causation arise the potential for spurious associations in those conventional studies, whether the associations of certain modifiable factors with the risk of abnormal spermatozoa are definitely causal remains undermined. In addition, there is still limited evidence determining causality between metabolic traits and semen parameters due to lack of rigorous randomized controlled trials (RCTs) designed and conducted with strict measured criteria (9,10).

Mendelian randomization (MR) utilizes specific genetic variants as instruments variables (IVs) in observational settings to strengthen causal inference between risk factors and disease outcomes (11). The random allocation of genetic variants at conception implies that the estimates through MR are less susceptible to confounding bias from environmental factors and reverse causality (11). This proves that MR is a reliable tool in the absence of RCTs to generate robust evidence and to seek risk factors for diseases. At present, there is still a lack of large-scale RCTs to infer the causality of metabolic factors with risk of abnormal spermatozoa. Hence, we conduct this MR study aiming to fill the gap in male fertility protection research and give rise to worldwide concern in this field. We present this article in accordance with the STROBE-MR reporting checklist (available at https://tau.amegroups.com/article/view/10.21037/tau-24-187/rc).

Methods

Study design

A two-sample MR analysis was conducted in this study using publicly available genome-wide association studies (GWASs) summary-level data for 16 modifiable factors and abnormal spermatozoa.

The MR framework and analytic process conformed to the STROBE-MR guidelines (12), and the overview of study design is presented in Figure 1. All the original GWASs summary datasets cited in our study were publicly available and had been approved by their corresponding ethical review committees respectively; therefore, no separate ethical approval was required for this study. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013).

Figure 1 MR study design overview (detailed assumptions can be found in the method of genetic instrument selection); HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; T2MD, type 2 diabetes mellitus; NAFLD, non-alcoholic fatty liver disease; SNP, single-nucleotide polymorphism; MR, Mendelian randomization; IVW, inverse variance weighted; MR-PRESSO, MR-Pleiotropy Residual Sum and Outlier methods.

Genetic instrument selection

The selection of genetic instrument was based on three critical assumptions: (I) Assumption 1 is the relevance assumption that the genetic variants used as IVs should be highly related to the exposure; (II) Assumption 2 is the independence assumption that the genetic variants for the exposure should not be associated with any confounders in the causality between the exposure and outcome; (III) Assumption 3 is the exclusivity assumption that the genetic variants affect the outcome merely through their effects on the exposure, rather than draw a direct connection to the outcome.

The selection of metabolic factors in this MR study was based on their clinical applicability. These factors are closely linked to metabolic syndrome, but the role of them was controversial in recent studies and lack of RCTs to confirm their causality with abnormal spermatozoa. Through evaluating these factors, we aimed to indicate the significance of lipid, glucose, and purine metabolism within the testicular environment and the impact on sperm quality, which might help elucidate the influence of metabolic factors on male reproductive health and contribute valuable insights to the emerging field of interest in reproductive biology. The 16 metabolic factors included body mass index, body fat percent, waist-to-hip ratio, fasting blood glucose, fasting blood insulin, glycated hemoglobin, high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), triglycerides, lipoprotein A, adiponectin, leptin, resistin, uric acid, type 2 diabetes mellitus (T2DM) and non-alcoholic fatty liver disease (NAFLD) (13-20).

Following the prerequisite mentioned above, single-nucleotide polymorphisms (SNPs) strongly associated at the genome-wide significance level (P≤5×10^–8) with the metabolic factors above were extracted from relevant GWASs (Table 1). Then, we excluded those SNPs in high linkage disequilibrium and remained the independent SNPs (r²<0.001) as IVs, which was estimated based on the 1000 Genomes European reference panel (21). Moreover, all these selected genetic instruments mentioned above should have strong potential to predict abnormal spermatozoa with F statistic greater than 10 (22). After filtering with these strict criteria, the remaining SNPs were regarded as eligible IVs. Detailed information on data sources of exposures including number of participants and IVs is shown in Table 1.

Data source of outcome

We obtained the GWAS summary data for abnormal spermatozoa from FinnGen Consortium R7 which was publicly released in 2022. The diagnostic criteria for disease were based on laboratory findings, including sperm count, vitality and morphology. Totally, the cohort of abnormal spermatozoa included 1,913 cases and 293,878 controls. The exposure and the outcome cohorts were both from European ancestry individuals to avoid violation caused by population differences (22). We retrieved summary data from the FinnGen, and extracted SNPs associated with metabolic status (including the effects of each of the SNPs on abnormal spermatozoa, beta coefficients and standard errors).

Statistical analysis

After harmonization to omit palindromic and incompatible SNPs across the GWASs of exposure and outcome data, we employed several MR approaches to run MR estimates of metabolic factor for abnormal spermatozoa, namely the inverse variance weighted (IVW), weighted median and MR-Egger. Multiple approaches stated above were on the basis of their different underlying assumptions for horizontal pleiotropy (violation of the exclusion restriction assumption that SNPs affect abnormal spermatozoa not merely through the exposure). As the primary statistical analysis, IVW is a method of weighting averages of random variables, where instruments can affect the outcome only through the exposure of interest and not by any alternative pathway (23). Exceptionally, when unbalanced horizontal pleiotropy exists, the selected SNPs might be invalid IVs in the IVW meta-analysis. Therefore, the MR-Egger and weighted median methods were further used to supplement IVW estimates as they could provide more robust estimates to the results in a broader set of scenarios but with relatively less efficiency. The MR-Egger can generate corrected MR estimates after adjustment for pleiotropic effects (24). The weighted median method is capable of consistent estimates if more than 50% of the genetic instruments are valid (25).

Sensitivity analysis has been essential in MR studies to detect potential horizontal pleiotropy and the heterogeneity. Firstly, Cochrane Q value was used as a marker of their pleiotropic effects to assess heterogeneity among SNP estimates (23). Secondly, we obtained the intercept test from the MR-Egger regression as an indicator of directional pleiotropy (P<0.05 was considered as the existence of directional pleiotropy (24). Moreover, we performed MR-Pleiotropy Residual Sum and Outlier methods (MR-PRESSO) to identify the outlier variants in MR estimates, remove them and correct horizontal pleiotropy (26). Eventually, leave-one-out analysis was also employed to evaluate whether the MR evaluation might be biased by some single SNP.

The above results were reported as the odds ratios (ORs) and 95% confidence intervals (CIs). We also adopted two-sided P values, and regarded P values less than 0.05 as suggestive significance. All the MR analyses were conducted using the R software (version 4.3.1) with the R package “TwoSampleMR”, “MRPRESSO”, and “forestplot”.

Results

The detailed information of data sources for the genetic instruments and the numbers of valid SNPs for each of 16 exposures in this present study are presented in Table 1. All F statistics for the overall instruments were more than 10, indicating the qualified power of the available genetic instruments.

In terms of three metabolic traits, the primary analysis IVW indicated that genetically predicted one standard deviation (SD) increase in body fat percentage might be causally associated with a higher risk of abnormal spermatozoa (OR =1.49, 95% CI: 1.01–2.20, P=0.046), while no causal effect was observed for body mass index (BMI) (OR =1.91, 95% CI: 0.73–5.01, P=0.19), and waist-to-hip ratio (OR =0.09, 95% CI: 0.00–697.20, P=0.60) (Figure 2).

Figure 2 The effect estimates of metabolic exposures on abnormal spermatozoa. BMI, body mass index; HbA1c, glycated hemoglobin; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; T2DM, type 2 diabetes mellitus; NAFLD, non-alcoholic fatty liver disease; OR, odds ratio; CI, confidence interval.

For the 11 serum metabolites including three glucometabolic related parameters, four lipid parameters, and three kinds of adipokines and uric acid, we observed a causal relationship between resistin and higher risk of abnormal spermatozoa (OR =1.55, 95% CI: 1.11–2.19, P=0.01). However, there was no significant potential association between fasting blood glucose, fasting blood insulin, glycated hemoglobin, HDL-C, LDL-C, triglycerides, lipoprotein A, adiponectin, leptin and uric acid and abnormal spermatozoa (Figure 2).

For T2DM (OR 0.99, 95% CI: 0.89–1.10, P=0.88) as well as NAFLD (OR =0.97, 95% CI: 0.79–1.19, P=0.75), no causal relationship was found (Figure 2). Meanwhile, the weighted median analysis and the sensitivity analyses including the test of heterogeneity and pleiotropy supported the above causation between metabolic factors and abnormal spermatozoa, which are presented in Table 2.

Discussion

In this population-based MR study, we investigated the causal effects of 16 metabolic factors on abnormal spermatozoa. Our findings suggested that genetically determined increased body fat percent and resistin were correlated with a higher risk of abnormal spermatozoa. In terms of other metabolic factors including BMI, waist-to-hip ratio, fasting blood glucose, fasting blood insulin, glycated hemoglobin, HDL-C, LDL-C, triglycerides, lipoprotein A, adiponectin, leptin, uric acid, T2DM and NAFLD, no causal effects were observed.

The association between obesity and infertility has been extensively followed with interest as the global epidemic of obesity rises sharply. However, the impact of overweight and obesity on semen parameters is still inconclusive. Although the first systematic review and meta-analysis concluded that the observed effects of obesity on sperm concentration were not significant (27), some updated systematic reviews have leaned toward the viewpoint that overweight and/or obesity categories might be associated with lower sperm quality including semen volume, sperm count and concentration, sperm vitality, total motility and normal morphology (28,29). Obesity might affect sperm directly or indirectly through several probable mechanisms including alterations in male sexual hormone profile (30), increased production of reactive oxygen species (ROS) and inflammatory mediators (31,32), and interaction of adipose tissue on testicular temperature inducing adverse living environment for sperm (33), and epigenetic changes including DNA methylation reprogramming and modification of noncoding RNAs in sperm (34,35). In our MR analysis, we employed three measured markers related to obesity, one of which namely body fat percent showed a harmful effect on sperm, but with weak statistical significance. Our finding supports the suggestion that obesity affects sperm quality, but the causal mechanism between obesity and spermatogenesis and sperm maturation is still unclear, and further research should be carried on to confirm this relationship in the future.

With regard to lipid metabolism, which plays a crucial role in spermatogenesis and obesity, its fluctuations in serum lipid metabolites might be a sign of abnormal spermatozoa. In light of the current knowledge, no studies have utilized MR to reveal a causal relationship between level of lipid metabolites and semen parameters. Therefore, the positive result of resistin brings a new insight that circulating levels of some serum adipokines that are less commonly used in clinical practice, might be independent risk factors for abnormal spermatozoa. Resistin is a 12.5 kDa pro-inflammatory adipokine that circulates in human blood as a dimeric protein (36). In the reproductive system, it is mainly expressed in rat Leydig cells, Sertoli cells and macrophages where it might regulate testicular functions under physiological condition, but this has not been demonstrated in humans (37). Recent researches reported that there are two putative binding sites for resistin: adenylyl cyclase-associated protein 1 (CAP1) and toll-like receptor 4 (TLR-4), which have been found in human sperm, and it is acknowledged that resistin activates various signaling pathways in different tissues such as Akt, MAPK, STAT3 and peroxisome proliferator-activated receptor gamma (PPARγ) (38,39). Nevertheless, results of previous studies upon the relationship between resistin and semen are heterogeneous. Moretti et al. found that concentration of resistin in seminal plasma was negatively correlated with sperm motility, and this adipokine is associated with markers of inflammation in seminal plasma such as elastase, interleukin-6 (IL-6), and tumor necrosis factor-α (TNF-α) (40). When inflammation occurs, the levels of these cytokines and ROS increase, which accounts for the impaired process of spermatogenetic present in human testis and induces a decrease in spermatic concentration, motility, and sperm count (40,41). However, two other researches did not show significant correlation between resistin concentrations in seminal plasma and sperm parameters (42,43). Given the small sample sizes and the low number of studies available, it is difficult to draw a definite conclusion through these researches. Indeed, based on MR, our study removed confounding factors to ensure that circulating resistin is independent of any body parameter and provided strong support for conclusion that the presence of this adipokine would be related to an alteration of sperm parameters. Furthermore, it is still necessary to elucidate specific mechanism of action of resistin and clarify the intertwined relationship between resistin and spermatozoa.

Although we found that the resistin measurement in serum among lipid biomarkers appeared to be robust risk factor for abnormal spermatozoa, there was lack of a causal association between other lipid parameters or adipokines and sperm quality. What account for this contradictory result might be the complex etiology between lipids and sperm production, which also leads to the conflict in the current epidemiology researches or observational cohort studies on the relationship between them (44). The proportion of lipids, such as cholesterol, present in sperm membranes has close relationship with the sperm morphology and fertility potential (45). The lipid in the sperm membrane, as one of the raw materials to maintain the stability of the cell membrane or as chemical messengers between cells, is essential to a certain extent with its physiological variation of content throughout the process of sperm differentiation, sperm maturation, capacitation, and acrosome reaction (45,46). However, it seems that the lipids in seminal plasma might be derived from epithelial cells in the male reproductive tract rather than blood (47). Taking the reverse into consideration, most previous researches suggested that in the case of increased lipid profile, decreased testosterone with leptin resistance, excess oestrogen can cause disturbances in spermatogenesis, apoptosis (abortive), and sperm damage (30,48,49). In addition, excessive blood lipid also involves an increase in ROS and breaks the balance of antioxidant system, so that the decline of count, morphological deterioration and DNA fragmentation of sperm would eventually occur (31). While our results suggested that such associations between lipid profile and abnormal spermatozoa might not be directly causal, and more comprehensive mechanism researches are necessary to further confirm their relationship.

Another important factor that might affect male fertility is purine metabolism. Uric acid is an important product related to purine nucleotide metabolism, which is the backbone of DNA molecules in sperm, and serum uric acid were highly correlated with testosterone levels (50). Some studies noted that a specific concentration of uric acid in semen can effectively sustain and enhance sperm motility and morphology, while also protecting functional integrity of sperm through the neutralization of oxidation processes, including endogenous free radicals and exogenous toxins (51). In contrast, various researches have consistently demonstrated that elevated levels of uric acid can exert a detrimental impact on sperm function such as the fertilization rate, to some extent, by diminishing the activity of crucial enzymes within sperm (52,53). However, our study did not find a direct causal relationship between serum uric acid and abnormal spermatozoa, therefore the determination of uric acid in the clinic needs further discussion in the management of infertility.

T2DM and NAFLD are metabolic diseases related to abnormal spermatozoa. Current cross-sectional studies investigating semen parameters and men with diabetes mellitus are heterogeneous. Over the decades, several cohort studies have identified associations between T2DM and increased risk of reduced sperm count and motility and increased morphological abnormalities (50,54). Characteristics of T2DM include abnormal glucose and lipid metabolism, which might progress to hyperinsulinemia and insulin resistance. These changes can not only induce disturbances in endocrine control and dysregulated spermatogenesis but also affect the sperm maturation process by increasing the substantial implications in the sperm DNA/chromatin levels of diabetes patients (55,56). On the contrary, there are meta-analyses supporting a negative effect of diabetes on sperm morphology but no effect on sperm count, with contradictory results concerning other semen parameters (57,58), which might be consistent with our conclusion that no significant causal effects were observed between T2MD (including several glycemic traits) glycemic traits and semen analysis. As for NAFLD, it is well known that NAFLD is closely linked to rise in the prevalence of diabetes, obesity, and male infertility. In contrast, the causal relationship between NAFLD and abnormal spermatozoa is still unclear, with a few studies being reported among relatively small cohorts (59,60). An early case-control study showed that NAFLD could significantly affect sperm concentration, count and total motility instead of semen volume and morphology (60), and NAFLD impairs reproductive function in male rats by decreasing the synthesis of testicular testosterone (61). In this MR study, we found no strong evidence to support associations between T2DM, NAFLD and abnormal spermatozoa, which is not in line with observational studies (55,60), these null findings should be cautiously interpreted given high heterogeneity in these analyses as well as a few genetic instruments for the metabolic diseases. Thus, the effects of T2DM and NAFLD on metabolic profiles need to be further explored.

The major strength of our present study lies in the first utilization of genetic instruments as proxies for various metabolic factors to perform two-sample MR analysis to infer their causal effects on risk of abnormal spermatozoa. After rigorous IVs selection and sensitivity testing, MR paradigm minimizes environmental confounding, diminishes reverse causality and derives robust evidence for causal effects of these risk factors, rather than just an association provided by most current cross-sectional studies, which might also be modified by the development and progression of the disease. In addition, we confined our analysis to the large-scale population of European ancestry, which effectively minimize the bias caused by the population structure.

Limitations of this MR investigation are also taken into account. Firstly, there is a lack of more detailed description upon the severity of sperm in patients with abnormal spermatozoa (regarding substantial parameters such as count, motility, malformation rate and DNA fragmentation index, etc.) in Finngen GWAS database or other available GWAS data of consistent race; thus, hierarchical analyses of different types of abnormal spermatozoa were unable to be conducted. A broader study containing subgroups of abnormal spermatozoa can be considered in the future to confirm the effects of metabolism on various types of abnormal sperm parameters more precisely. Secondly, the small sample size of GWASs might contribute to imprecision in the selection of genetic instruments, which could result in insufficient power to detect small or moderate associations. To solve these faults, larger GWASs and more well-designed clinical trials are necessary to better reveal the cause-and-effect correlation and to evaluate interventions targeting body fat or resistin. It is also indispensable to conduct the investigation of specific mechanisms by which resistin affects sperm through in-vitro and in-vivo experiments. Another limitation is that our study population of consistent ancestry might obstruct the generalizability of our findings to other populations, which is also urged to be solved in the larger-scale MR study in the future.

Conclusions

In conclusion, this is the first wide angled MR analysis to explore the causality from metabolic factors on abnormal spermatozoa. Our MR analysis provides suggestive evidence that body fat percentage and resistin are risk factors for abnormal spermatozoa, but does not support the causal impact of other metabolic factors including BMI, glucose parameters, lipid profile, uric acid, T2DM and NAFLD. Further MR study with more genetic instruments for metabolic risk factors and diseases, followed by the investigation in larger-scale GWASs on subgroups of abnormal spermatozoa in the future, are necessary to confirm our findings.

Acknowledgments

Authors thank all investigators for sharing GWASs summary-level datasets.

Funding: This work was supported by the Key Science and Technology Program of Shunde Hospital, Southern Medical University (The First People’s Hospital of Shunde) (No. SRSP2022021).

Footnote

Reporting Checklist: The authors have completed the STROBE-MR reporting checklist. Available at https://tau.amegroups.com/article/view/10.21037/tau-24-187/rc

Peer Review File: Available at https://tau.amegroups.com/article/view/10.21037/tau-24-187/prf

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tau.amegroups.com/article/view/10.21037/tau-24-187/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013).

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Cox CM, Thoma ME, Tchangalova N, et al. Infertility prevalence and the methods of estimation from 1990 to 2021: a systematic review and meta-analysis. Hum Reprod Open 2022;2022:hoac051. [Crossref] [PubMed]
2. Agarwal A, Mulgund A, Hamada A, et al. A unique view on male infertility around the globe. Reprod Biol Endocrinol 2015;13:37. [Crossref] [PubMed]
3. Levine H, Jørgensen N, Martino-Andrade A, et al. Temporal trends in sperm count: a systematic review and meta-regression analysis of samples collected globally in the 20th and 21st centuries. Hum Reprod Update 2023;29:157-76. [Crossref] [PubMed]
4. Barratt CLR, De Jonge CJ, Anderson RA, et al. A global approach to addressing the policy, research and social challenges of male reproductive health. Hum Reprod Open 2021;2021:hoab009. [Crossref] [PubMed]
5. Boeri L, Capogrosso P, Ventimiglia E, et al. Undiagnosed prediabetes is highly prevalent in primary infertile men - results from a cross-sectional study. BJU Int 2019;123:1070-7. [Crossref] [PubMed]
6. Martins AD, Majzoub A, Agawal A. Metabolic Syndrome and Male Fertility. World J Mens Health 2019;37:113-27. [Crossref] [PubMed]
7. Craig JR, Jenkins TG, Carrell DT, et al. Obesity, male infertility, and the sperm epigenome. Fertil Steril 2017;107:848-59. [Crossref] [PubMed]
8. Kasman AM, Del Giudice F, Eisenberg ML. New insights to guide patient care: the bidirectional relationship between male infertility and male health. Fertil Steril 2020;113:469-77. [Crossref] [PubMed]
9. Rimmer MP, Howie RA, Anderson RA, et al. Protocol for developing a core outcome set for male infertility research: an international consensus development study. Hum Reprod Open 2022;2022:hoac014. [Crossref] [PubMed]
10. Rimmer MP, Howie RA, Subramanian V, et al. Outcome reporting across randomized controlled trials evaluating potential treatments for male infertility: a systematic review. Hum Reprod Open 2022;2022:hoac010. [Crossref] [PubMed]
11. Davey Smith G, Hemani G. Mendelian randomization: genetic anchors for causal inference in epidemiological studies. Hum Mol Genet 2014;23:R89-98. [Crossref] [PubMed]
12. Skrivankova VW, Richmond RC, Woolf BAR, et al. Strengthening the Reporting of Observational Studies in Epidemiology Using Mendelian Randomization: The STROBE-MR Statement. JAMA 2021;326:1614-21. [Crossref] [PubMed]
13. Shungin D, Winkler TW, Croteau-Chonka DC, et al. New genetic loci link adipose and insulin biology to body fat distribution. Nature 2015;518:187-96. [Crossref] [PubMed]
14. Chen J, Spracklen CN, Marenne G, et al. The trans-ancestral genomic architecture of glycemic traits. Nat Genet 2021;53:840-60. [Crossref] [PubMed]
15. Dastani Z, Hivert MF, Timpson N, et al. Novel loci for adiponectin levels and their influence on type 2 diabetes and metabolic traits: a multi-ethnic meta-analysis of 45,891 individuals. PLoS Genet 2012;8:e1002607. [Crossref] [PubMed]
16. Yaghootkar H, Zhang Y, Spracklen CN, et al. Genetic Studies of Leptin Concentrations Implicate Leptin in the Regulation of Early Adiposity. Diabetes 2020;69:2806-18. [Crossref] [PubMed]
17. Folkersen L, Gustafsson S, Wang Q, et al. Genomic and drug target evaluation of 90 cardiovascular proteins in 30,931 individuals. Nat Metab 2020;2:1135-48. [Crossref] [PubMed]
18. Sakaue S, Kanai M, Tanigawa Y, et al. A cross-population atlas of genetic associations for 220 human phenotypes. Nat Genet 2021;53:1415-24. [Crossref] [PubMed]
19. Xue A, Wu Y, Zhu Z, et al. Genome-wide association analyses identify 143 risk variants and putative regulatory mechanisms for type 2 diabetes. Nat Commun 2018;9:2941. [Crossref] [PubMed]
20. Ghodsian N, Abner E, Emdin CA, et al. Electronic health record-based genome-wide meta-analysis provides insights on the genetic architecture of non-alcoholic fatty liver disease. Cell Rep Med 2021;2:100437. [Crossref] [PubMed]
21. Clarke L, Zheng-Bradley X, Smith R, et al. The 1000 Genomes Project: data management and community access. Nat Methods 2012;9:459-62. [Crossref] [PubMed]
22. Burgess S, Davies NM, Thompson SG. Bias due to participant overlap in two-sample Mendelian randomization. Genet Epidemiol 2016;40:597-608. [Crossref] [PubMed]
23. Burgess S, Butterworth A, Thompson SG. Mendelian randomization analysis with multiple genetic variants using summarized data. Genet Epidemiol 2013;37:658-65. [Crossref] [PubMed]
24. Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. Int J Epidemiol 2015;44:512-25. [Crossref] [PubMed]
25. Bowden J, Davey Smith G, Haycock PC, et al. Consistent Estimation in Mendelian Randomization with Some Invalid Instruments Using a Weighted Median Estimator. Genet Epidemiol 2016;40:304-14. [Crossref] [PubMed]
26. Verbanck M, Chen CY, Neale B, et al. Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nat Genet 2018;50:693-8. [Crossref] [PubMed]
27. MacDonald AA, Herbison GP, Showell M, et al. The impact of body mass index on semen parameters and reproductive hormones in human males: a systematic review with meta-analysis. Hum Reprod Update 2010;16:293-311. [Crossref] [PubMed]
28. Sermondade N, Faure C, Fezeu L, et al. BMI in relation to sperm count: an updated systematic review and collaborative meta-analysis. Hum Reprod Update 2013;19:221-31. [Crossref] [PubMed]
29. Salas-Huetos A, Maghsoumi-Norouzabad L, James ER, et al. Male adiposity, sperm parameters and reproductive hormones: An updated systematic review and collaborative meta-analysis. Obes Rev 2021;22:e13082. [Crossref] [PubMed]
30. Mah PM, Wittert GA. Obesity and testicular function. Mol Cell Endocrinol 2010;316:180-6. [Crossref] [PubMed]
31. Nätt D, Kugelberg U, Casas E, et al. Human sperm displays rapid responses to diet. PLoS Biol 2019;17:e3000559. [Crossref] [PubMed]
32. Fan W, Xu Y, Liu Y, et al. Obesity or Overweight, a Chronic Inflammatory Status in Male Reproductive System, Leads to Mice and Human Subfertility. Front Physiol 2018;8:1117. [Crossref] [PubMed]
33. Palmer NO, Bakos HW, Fullston T, et al. Impact of obesity on male fertility, sperm function and molecular composition. Spermatogenesis 2012;2:253-63. [Crossref] [PubMed]
34. Keyhan S, Burke E, Schrott R, et al. Male obesity impacts DNA methylation reprogramming in sperm. Clin Epigenetics 2021;13:17. [Crossref] [PubMed]
35. Bodden C, Hannan AJ, Reichelt AC. Diet-Induced Modification of the Sperm Epigenome Programs Metabolism and Behavior. Trends Endocrinol Metab 2020;31:131-49. [Crossref] [PubMed]
36. Elfassy Y, Bastard JP, McAvoy C, et al. Adipokines in Semen: Physiopathology and Effects on Spermatozoas. Int J Endocrinol 2018;2018:3906490. [Crossref] [PubMed]
37. Roumaud P, Martin LJ. Roles of leptin, adiponectin and resistin in the transcriptional regulation of steroidogenic genes contributing to decreased Leydig cells function in obesity. Horm Mol Biol Clin Investig 2015;24:25-45. [Crossref] [PubMed]
38. Lee S, Lee HC, Kwon YW, et al. Adenylyl cyclase-associated protein 1 is a receptor for human resistin and mediates inflammatory actions of human monocytes. Cell Metab 2014;19:484-97. [Crossref] [PubMed]
39. Benomar Y, Gertler A, De Lacy P, et al. Central resistin overexposure induces insulin resistance through Toll-like receptor 4. Diabetes 2013;62:102-14. [Crossref] [PubMed]
40. Moretti E, Collodel G, Mazzi L, et al. Resistin, interleukin-6, tumor necrosis factor-alpha, and human semen parameters in the presence of leukocytospermia, smoking habit, and varicocele. Fertil Steril 2014;102:354-60. [Crossref] [PubMed]
41. Kurowska P, Mlyczyńska E, Dawid M, et al. Endocrine disruptor chemicals, adipokines and reproductive functions. Endocrine 2022;78:205-18. [Crossref] [PubMed]
42. Thomas S, Kratzsch D, Schaab M, et al. Seminal plasma adipokine levels are correlated with functional characteristics of spermatozoa. Fertil Steril 2013;99:1256-1263.e3. [Crossref] [PubMed]
43. Kratzsch J, Paasch U, Grunewald S, et al. Resistin correlates with elastase and interleukin-6 in human seminal plasma. Reprod Biomed Online 2008;16:283-8. [Crossref] [PubMed]
44. Pakpahan C, Rezano A, Margiana R, et al. The Association Between Lipid Serum and Semen Parameters: a Systematic Review. Reprod Sci 2023;30:761-71. [Crossref] [PubMed]
45. Keber R, Rozman D, Horvat S. Sterols in spermatogenesis and sperm maturation. J Lipid Res 2013;54:20-33. [Crossref] [PubMed]
46. Chen S, Wang M, Li L, et al. High-coverage targeted lipidomics revealed dramatic lipid compositional changes in asthenozoospermic spermatozoa and inverse correlation of ganglioside GM3 with sperm motility. Reprod Biol Endocrinol 2021;19:105. [Crossref] [PubMed]
47. Liu CY, Chou YC, Lin SH, et al. Serum lipid profiles are associated with semen quality. Asian J Androl 2017;19:633-8. [Crossref] [PubMed]
48. Kwon O, Kim KW, Kim MS. Leptin signalling pathways in hypothalamic neurons. Cell Mol Life Sci 2016;73:1457-77. [Crossref] [PubMed]
49. Andrade G, Iori I, Hsieh MK, et al. Serum lipid profile levels and semen quality: new insights and clinical perspectives for male infertility and men's health. Int Urol Nephrol 2023;55:2397-404. [Crossref] [PubMed]
50. Kurniawan AL, Hsu CY, Chao JC, et al. Association of Testosterone-Related Dietary Pattern with Testicular Function among Adult Men: A Cross-Sectional Health Screening Study in Taiwan. Nutrients 2021;13:259. [Crossref] [PubMed]
51. Banihani SA. Role of Uric Acid in Semen. Biomolecules 2018;8:65. [Crossref] [PubMed]
52. Allahkarami S, Atabakhsh M, Moradi M, et al. Correlation of uric acid, urea, ammonia and creatinine of seminal plasma with semen parameters and fertilization rate of infertile couples. Avicenna J Med Biochem 2017;5:76-80. [Crossref]
53. Ji X, Yu L, Han C, et al. Investigating the effects of rare ginsenosides on hyperuricemia and associated sperm damage via nontargeted metabolomics and gut microbiota. J Ethnopharmacol 2024;332:118362. [Crossref] [PubMed]
54. Zhong O, Ji L, Wang J, et al. Association of diabetes and obesity with sperm parameters and testosterone levels: a meta-analysis. Diabetol Metab Syndr 2021;13:109. [Crossref] [PubMed]
55. Imani M, Talebi AR, Fesahat F, et al. Sperm parameters, DNA integrity, and protamine expression in patients with type II diabetes mellitus. J Obstet Gynaecol 2021;41:439-46. [Crossref] [PubMed]
56. Shi GJ, Zheng J, Wu J, et al. Beneficial effects of Lycium barbarum polysaccharide on spermatogenesis by improving antioxidant activity and inhibiting apoptosis in streptozotocin-induced diabetic male mice. Food Funct 2017;8:1215-26. [Crossref] [PubMed]
57. Pergialiotis V, Prodromidou A, Frountzas M, et al. Diabetes mellitus and functional sperm characteristics: A meta-analysis of observational studies. J Diabetes Complications 2016;30:1167-76. [Crossref] [PubMed]
58. Lotti F, Maggi M. Effects of diabetes mellitus on sperm quality and fertility outcomes: Clinical evidence. Andrology 2023;11:399-416. [Crossref] [PubMed]
59. Li Y, Liu L, Wang B, et al. Nonalcoholic fatty liver disease and alteration in semen quality and reproductive hormones. Eur J Gastroenterol Hepatol 2015;27:1069-73. [Crossref] [PubMed]
60. Ommati MM, Li H, Jamshidzadeh A, et al. The crucial role of oxidative stress in non-alcoholic fatty liver disease-induced male reproductive toxicity: the ameliorative effects of Iranian indigenous probiotics. Naunyn Schmiedebergs Arch Pharmacol 2022;395:247-65. [Crossref] [PubMed]
61. Li Y, Liu L, Wang B, et al. Impairment of reproductive function in a male rat model of non-alcoholic fatty liver disease and beneficial effect of N-3 fatty acid supplementation. Toxicol Lett 2013;222:224-32. [Crossref] [PubMed]