Dissecting Causal Relationships Between Gut Microbiota, Blood Metabolites, and Postpartum Depression : A Mendelian Randomization Study

doi:10.21203/rs.3.rs-4911853/v1

Download PDF

Article

Dissecting Causal Relationships Between Gut Microbiota, Blood Metabolites, and Postpartum Depression : A Mendelian Randomization Study

https://doi.org/10.21203/rs.3.rs-4911853/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background: There is substantial evidence indicating a significant link between the gut microbiota (GM), blood metabolites and postpartum depression (PPD) . However, the causal relationships underlying these associations have yet to be established.

Methods: This study elucidated the causal relationship among 196 GM taxa, 224 blood metabolites, and PPD from a genetic perspective, employing two-sample Mendelian randomization (MR) and mediation analyses. To validate the relevant findings, we further selected data (GM and blood metabolites) from the the IEU Open GWAS and GWAS Catalog for analysis. Our primary analysis utilized the inverse variance weighted method. To enhance the robustness of our results, we also applied MR-Egger method, weighted median method, Cochran’s Q test, MR-Egger regression, and MR-PRESSO.

Results: MR analysis results revealed a nominal association (p < 0.05) between 13 GM taxa, 6 blood metabolites, and PPD. After Bonferroni correction, only Bifidobacteriales (OR: 0.844, 95% CI: 0.760-0.938; p = 0.0017) showed a significant causal relationship with a lower risk of PPD. In our validation results, the higher level of Alphaproteobacteria (OR: 1.057, 95% CI: 1.024-1.091; Bonferroni-corrected p = 0.0006) retained a strong causal relationship with a higher risk of PPD. Finally, mediation analysis revealed that the impact of Odoribacteron PPD was mediated indirectly through Hyodeoxycholate, with a mediation proportion of 16.8%.

Conclusion: Our findings underscore the importance of elucidating the underlying mechanisms between the GM, blood metabolites, and PPD. These results provide novel insights into microbiome-based therapies and metabolite-targeted interventions for PPD.

Biological sciences/Computational biology and bioinformatics

Biological sciences/Microbiology

Biological sciences/Molecular biology

Biological sciences/Neuroscience

postpartum depression

gut microbiota

blood metabolites

mendelian randomization

mediation analyses

Postpartum depression (PPD) is a prevalent psychological disorder affecting women after childbirth, characterized by symptoms such as insomnia, depression, anxiety, irritability, and hallucinations(Payne and Maguire, 2019; Stewart and Vigod, 2019). The rapid pace of modern social life and increasing family pressures have contributed to a gradual rise in the prevalence of PPD, which ranges between 5.0% and 26.32%(Liu et al., 2022). Notably, the prevalence in developing countries (15.0%) surpasses that in developed countries (12.0%)(Liu et al., 2022). PPD not only impacts the quality of life for mothers but also increases the risk of impaired emotional, social, and cognitive development in infants(Zhao and Zhang, 2020). In rare cases, it may even escalate the risk of maternal infanticide or suicide, making it a focal point in clinical and public health research(Zhao and Zhang, 2020).

In recent years, research on the gut microbiota (GM) and blood metabolites in mood disorders has garnered widespread attention. The GM, a pivotal regulator within the host, plays a crucial role in modulating neurotransmitter synthesis (such as serotonin) and metabolite production (such as short-chain fatty acids). By the modulation of the microbiota-gut-brain axis, it influences the central nervous system, thereby impacting emotions and behavior(Li et al., 2022; Zhang et al., 2023). A clinical study revealed a close correlation between the onset of PPD and a reduction in the abundance of Faecalibacterium, Phascolarctacterium, and Butyricicoccus(Zhou et al., 2020). Animal studies have also identified that multiple key GM (sp.2.1.33B and sp. CAG:755) are associated with PPD(Tian et al., 2021). Additionally, research has demonstrated that antidepressant drugs (e.g., fluoxetine) exert anti-postpartum depression effects through the regulation of the GM(Ramsteijn et al., 2020; Zheng et al., 2023). Furthermore, the regulation of blood metabolites is closely intertwined with various physiological and pathological processes, including energy metabolism, inflammatory responses, and neural regulation. One study disclosed that five metabolites (glycerol, threonine, 2-hydroxybutanoic acid, erythritol, and phenylalanine) are closely linked to PPD, suggesting their potential utility in the diagnosis and physiological pathological examination of PPD(Brann et al., 2021).

Evidence from various perspectives has hinted at potential links between GM, blood metabolites, and the risk of PPD. If these associations prove causal, GM and blood metabolites could emerge as novel targets for the screening, prevention, and treatment of PPD. Unfortunately, most existing observational studies on PPD have predominantly focused on associations with depressive-related GM and blood metabolites. This focus hampers the exploration of PPD pathogenesis and drug discovery. Moreover, causal relationships inferred from clinical trials and animal studies are prone to biases due to confounding factors (such as lifestyle, socioeconomic status, and etc.) and reverse causation. Overcoming these influencing factors and clarifying the causal relationship between GM, blood metabolites, and PPD remains a key scientific question in this field.

Similar to randomized controlled trials, Mendelian randomization (MR) studies are a novel research approach aimed at elucidating causal relationships between exposures and outcomes(Bowden and Holmes, 2019). This method provides more robust evidence for causal inference than observational epidemiology. In MR studies, single nucleotide polymorphisms (SNPs) are used as instrumental variables (IVs) to estimate causal relationships between exposures and outcomes. SNPs conform to the principle of random assignment of genetic variants at meiosis, which avoids the effect of confounding factors and the potential impact of reverse causation since genetic variants precede the onset of disease(Lawlor et al., 2008).

Our study will utilize the largest available datasets of GM, blood metabolites, and PPD, which have the highest sample sizes and SNP counts. Through MR studies, we aim to comprehensively explore the causal relationships between GM, blood metabolites and PPD, and elucidate the micro-mechanisms by which GM mediates the involvement of blood metabolites in PPD pathogenesis. Our study will address the key scientific question mentioned earlier, providing new novel insights into the pathogenesis, diagnosis, prevention, and treatment of PPD. Additionally, the methodology of this study will serve as a reference for scientific researchers in other medical fields, aiding in addressing key scientific questions within their domains.

Study design

Summary statistics on the GM, blood metabolites, and PPD were obtained from the respective consortia for the study design. Figure 1 illustrates the study design diagram. The STROBE-MR guidelines were used to design the study(Skrivankova et al., 2021).

Data sources

This study defined the abundance of GM and the concentration of blood metabolites as the exposure factors, with PPD as the outcome (Supplementary Table 1). The genetic association results of the exposure factors were derived from two datasets: the GM abundance GWAS dataset (MiBioGen consortium, https://mibiogen.gcc.rug.nl/) and the blood metabolites GWAS dataset (IEU Open GWAS database, https://gwas.mrcieu.ac.uk/). Summary data of GM in the MiBioGen consortium included 18,340 participants of multiple ancestries from 24 cohorts, of which 78% were Europeans(Kurilshikov et al., 2021). The GM data covered 211 GM taxa with a mean abundance of more than 1%, encompassing 131 genera, 35 families, 20 orders, 16 classes, and 9 phyla. After excluding 15 GM taxa belonging to unknown groups (12 genera and 3 families), 196 bacterial taxa were ultimately included in the MR analysis(Li et al., 2023). The blood metabolome GWAS dataset investigated the relationship between 452 blood metabolites and germline variants in 7824 European subjects(Shin et al., 2014). After eliminating 201 metabolites with unidentified chemical properties (prefixed by "X-"), the MR analysis ultimately included 224 blood metabolites. The genetic association results of the outcome were obtained from the Finngen public database (https://r8.finngen.fi/), the GWAS data included 13,657 patients with PPD and 236,178 control cases of European ancestry. The International Classification of Diseases, 10th Revision (ICD-10) codes F32, F33, and F53.0 were used to define PPD (Kurki et al., 2023). Information on the GWAS datasets used to verify the significant and potential causalities between 196 GM, 224 blood metabolites, and PPD is shown in Supplementary Table 1.

IVs selection

MR studies must satisfy three key assumptions: (1) IVs are strongly associated with exposure, (2) IVs are independent of confounding factors, and (3) IVs are solely associated with outcomes through exposure(Emdin et al., 2017). To meet the assumptions, the following selection criteria were employed to choose the IVs: (1) SNPs associated with each genus at the locus-wide significance threshold ( P < 1× 10^− 5 ) were selected as potential IVs; (2) 1000 Genomes Project European sample data were used as the reference panel to calculate the linkage disequilibrium (LD) between the SNPs, and among those SNPs, those with r² < 0.001 were selected (clumping window size = 10,000 kb); (3) the strength of IVs was assessed by the F-statistic, with an F-statistic > 10 considered as having no significant weak instrumental bias [F-statistic = beta²/se²](Cao et al., 2023; Xie et al., 2023). To eliminate the influence of confounding factors, we further searched the LDtrait (https://ldlink.nih.gov/?tab=ldtrait)(Lin et al., 2020; Rodriguez-Martin et al., 2023) to check whether the selected SNPs of significant MR estimates in this study were associated with other PPD risk factors, which mainly concerning major depression, age at first birth, and history of depression(Silverman et al., 2017; Zuo et al., 2023).

Mendelian randomization and mediation analysis

Univariate MR analysis was performed to identify the causal relationship. The inverse variance weighted (IVW) approach assumes the validity of all genetic variants, complying with the three fundamental assumptions, which is considered the most robust MR method(Hartwig et al., 2017). Thus the IVW fixed-effect (IVW-fe) method or the IVW random effect (IVW-re) method was used as the primary MR analysis. The IVW-re method was utilized in cases of heterogeneity (P < 0.05), while the IVW-fe method was applied in its absence (P > 0.05) (Xu et al., 2024). The IVW results were validated using other methods, including MR-Egger and weighted median. If the IVW method yielded a significant result (p < 0.05), it was considered a positive result even if the other methods did not show significance, as long as the effect of the other methods were in the same direction(Li et al., 2024). For the primary MR results, multiple-testing significance was determined at each feature level using Bonferroni correction (p < 0.05/n, where n are the number of bacterial taxa included in each feature level and blood metabolites)(Dai et al., 2023; Long et al., 2023). Hence, the multiple-testing significance thresholds were 5.56 × 10^− 3(0.05/9), 3.13× 10^− 3(0.05/16), 2.5 × 10^− 3(0.05/20), 1.56 × 10^− 3(0.05/32), 4.20 × 10^− 4(0.05/119), and 2.23×10 − 4(0.05/224), respectively, for phylum, class, order, family, and genus. MR results with p values less than Bonferroni-corrected threshold were considered significant. Meanwhile, we considered MR estimates with p < 0.05 as nominal significant(Luo et al., 2023).

Mediation analysis aims to investigate whether a variable mediates the relationship between two other variables. Through mediation MR analysis, we can construct a pathway from exposure factors to outcomes through a mediating factor, helping elucidate the potential mechanism of exposure factors affecting outcomes. For example, if the abundance of a taxon and the concentration of a blood metabolite are both causally associated (i.e., positively associated) with PPD, and the abundance of the taxon is also causally associated (i.e., positively associated) with the concentration of the blood metabolite, a triangular relationship is formed(Wu et al., 2023). We used two-step Mendelian randomization for mediation analysis. The proportion of the total effect mediated by blood metabolites was estimated by dividing the indirect effect by the total effect (β1 × β2/β3), which β1 representing the effect of GM on the blood metabolites, β2 representing the effect of blood metabolites on PPD, and β3 representing the effect of GM on PPD(Sanderson, 2021).

Cochran’s Q test (IVW and MR-Egger) was employed to identify potential heterogeneous IVs. Additionally, MR-Egger regression and MR-PRESSO were conducted to assess whether directional horizontal pleiotropy might be influencing the results of the MR analyses(Rees et al., 2017; Verbanck et al., 2018).

All MR analyses were performed in R (version 4.3.1) using the following R packages: TwoSampleMR, ggplot2, forestplot, forestploter, and MRPRESSO(Hemani et al., 2018).

Causal effects of GM on PPD

The F-statistics of the IVs were all largely > 10, indicating no evidence of weak instrument bias ( Fig. 2). When evaluating the causal effects of GM on PPD, after a rigorous Bonferroni correction, a marginal significant causal relationship between Bifidobacteriales and PPD risk was identified using the IVW method. Additionally, we detected potential causalities between twelve GM taxa and PPD (p < 0.05, see Fig. 2 ).

In detail, the genetically predicted lower abundance of Bifidobacteriales was causally associated with a higher risk of PPD (IVW: OR = 0.844, 95% confidence interval [CI] 0.760–0.938, P = 0.0017, see Fig. 2). Verrucomicrobia (IVW: OR = 1.135, 95% CI 1.009–1.278, P = 0.0355), Alphaproteobacteria (IVW: OR = 1.183, 95% CI 1.046–1.338, P = 0.0076), Odoribacter (IVW: OR = 1.215, 95% CI 1.041–1.419, P = 0.0136), and Paraprevotella (IVW: OR = 1.126, 95% CI 1.004–1.263, P = 0.0428) were identified to have suggestive positive causal effects on the risk of PPD (Fig. 2), whereas Clostridia (IVW: OR = 0.850, 95% CI 0.743–0.972, P = 0.0178), Bifidobacteriaceae (IVW: OR = 0.844, 95% CI 0.760–0.938, P = 0.0017), Bifidobacterium (IVW: OR = 0.889, 95% CI 0.799–0.989, P = 0.0300), Eggerthella (IVW: OR = 0.888, 95% CI 0.808–0.976, P = 0.0139), Oxalobacter (IVW: OR = 0.921, 95% CI 0.856–0.992, P = 0.0291), Prevotella7 (IVW: OR 0.925, 95% CI 0.863–0.991, P = 0.0269), Romboutsia (IVW: OR 0.874, 95% CI 0.775–0.987, P = 0.0297), and RuminococcaceaeUCG011 (IVW: OR = 0.914, 95% CI 0.844–0.991, P = 0.0287) had a tendency to causally decrease the risk of PPD (Fig. 2). Except for Romboutsia, Odoribacter, and Paraprevotella, the results of other MR analyses (MR-Egger and Weighted median) for the remaining 10 taxa were consistent with their respective IVW results. For Romboutsia, Odoribacter, and Paraprevotella, the effects estimated by MR-Egger were in the opposite direction to the results of the other two MR analyses, although it was not significant.

The Cochran's Q test revealed heterogeneity of IVs in Paraprevotella (P < 0.05 in IVW and MR-Egger methods, see Supplementary Table 2), so we chose the IVW-re method as the primary MR analysis. The MR-Egger intercept and the MR-PRESSO global test confirmed that there is no significant directional horizontal pleiotropy (P > 0.05, see Supplementary Table 2).

Causal effects of blood metabolites on PPD

The F statistics of all retained SNPs were over 10, indicating sufficient correlation strength between IVs and blood metabolites (Fig. 3). Among the 224 tested blood metabolites phenotypes, after Bonferroni correction, no significant causal relationship between blood metabolites and PPD risk was identified by IVW method. But potential causalities between 6 blood metabolites and PPD risk was identified by IVW method (Fig. 3). In detail, Choline (IVW: OR = 2.297, 95% CI 1.010–5.225, P = 0.0473), Guanosine (IVW: OR = 1.407, 95% CI 1.166–1.697, P = 0.0004), Hyodeoxycholate (IVW: OR = 1.164, 95% CI 1.004–1.349, P = 0.0437), Malate (IVW: OR = 1.993, 95% CI 1.206–3.295, P = 0.0072), and Tryptophan (IVW: OR = 1.901, 95% CI 1.045–3.458, P = 0.0353) were identified to have suggestive positive causal effects on the risk of PPD (Fig. 3). 2-hydroxybutyrate (AHB) (IVW: OR = 0.515, 95% CI 0.338–0.787, P = 0.0021) had a tendency to causally decrease the risk of PPD (Fig. 3). Except for Malate, the results of other MR analyses (MR-Egger and Weighted median) for the remaining 5 blood metabolites were consistent with their respective IVW results (Fig. 3). For Malate, the effect estimated by MR-Egger was in a reversed direction to the results of the other two MR analyses, although it was not significant.

The Cochran's Q test revealed heterogeneity of IVs in Tryptophan (Supplementary Table 3), so we chose the IVW-re method as the primary MR analysis. The MR-Egger intercept and the MR-PRESSO global test confirmed that there are no significant directional horizontal pleiotropy (P > 0.05, Supplementary Table 3).

Mediation analysis results

We found 13 GM taxa and 6 blood metabolites that were causally associated with PPD in previous studies. To investigate the potential mechanism involving GM and blood metabolites in the development of PPD, we explored the causal relationship between 16 GM taxa (exposure) and 6 blood metabolites (outcome, mediator).

All F-statistics were over 10, suggesting no weak instrumental variables were employed (Supplementary Table 4). After Bonferroni correction, no significant causal relationship between GM and blood metabolites was identified by IVW method. Two potential causal relationships were identified (Supplementary Table 4). Hyodeoxycholate (IVW: OR = 1.242, 95% CI 1.041–1.481, P = 0.0159) and Tryptophan (IVW: OR = 1.308, 95% CI 1.005–1.072, P = 0.0228) were identified to have suggestive positive causal effects with Odoribacter (Supplementary Table 4). Heterogeneity and horizontal pleiotropy were not found(Supplementary Table 5).

Through two-step Mendelian randomization, we observed an indirect effect of Odoribacter on PPD through Hyodeoxycholate, with a mediated proportion of 16.8% of the total effect (Fig. 4).

Validation set

We validated the causal relationships of 13 GM taxa and 6 blood metabolites associated with PPD in the data from the IEU Open GWAS and GWAS Catalog. All F-statistics were over 10, suggesting no weak instrumental variables were employed (Fig. 5). Through verification, we found that one taxa still have a potential causality with PPD (Fig. 5). After a rigorous Bonferroni correction (p < 0.05/10), a significant causal relationship between Alphaproteobacteria and PPD risk was identified (IVW: OR = 1.057, 95% CI 1.024–1.091, P = 0.0006). The results of other MR analyses (MR-Egger and Weighted median) were consistent with their respective IVW results (Fig. 5). Heterogeneity and horizontal pleiotropy were not found (Supplementary Table 8).

RuminococcaceaeUCG011, Prevotella7, 2-hydroxybutyrate, Guanosine, Hyodeoxycholate could not be validated as suitable GWAS data were not found. MR analysis results about other GM taxa and blood metabolites could be found in the supplemental materials(Supplementary Table 6–9). Supplementary Table 8 showed that there was no significant or potential causal relationship between Bifidobacteriales and PPD, possibly due to a smaller sample size (previous sample size: 14,306; current sample size: 7,738). However, the P-value of the results was 0.08 < 0.1, and the direction of the effect (beta) was consistent with previous study. Therefore, the result still reliably supported the conclusion that there was a causal relationship between Bifidobacteriales and PPD.

Causal effects of GM on PPD

This study represents the inaugural comprehensive, large-scale MR analysis examining the genetic basis for a causal relationship between GM and PPD. Prior research on this association predominantly utilized clinical trials and animal studies, which involved collecting and analyzing fecal samples from patients, PPD mouse, and PPD rats with frailty using 16S rRNA gene sequencing(Ramsteijn et al., 2020; Xu et al., 2024; Zhou et al., 2020). However, these approaches often faced challenges due to confounding factors such as major depression, and age at first birth, complicating the establishment of a direct causal link. In contrast, our MR study provides robust evidence supporting a causal connection between specific GM compositions and an increased risk of PPD. This pioneering work paves the way for identifying new biomarkers in future frailty research, marking a significant advancement in the field.

Bifidobacteriales, Bifidobacteriaceae, and Bifidobacterium belong to the Actinobacteria phylum. Research indicates that these entities share a hierarchical taxonomic relationship and are negatively correlated with PPD. Existing evidence confirms a negative association between these entities and depression(Chao and Zhang, 2020). A clinical trial revealed a decrease in the abundance of Bifidobacterium in PPD patients compared to the healthy control group(Zhang et al., 2023). Animal studies also found similar results(Zhao et al., 2022; Zhou et al., 2023). Additionally, another clinical trial found that oral administration of Bifidobacterium breve BB077 can improve depressive symptoms in PPD patients(Vicariotto et al., 2023), providing further evidence for the findings of our study. The above evidence suggests that the three taxa are protective against PPD, indicating a potential direction for therapeutic drug research for PPD.

Our study found a negative causal relationship between Clostridia and an increased risk of PPD. Chen(Chen et al., 2021) discovered that the abundance of Clostridia in patients with ulcerative colitis (UC) and depression significantly decreased compared to healthy individuals and UC patients without depression. Liśkiewicz(Liskiewicz et al., 2021) also found a negative correlation between Clostridia and HAM-D scores. These findings provide robust evidence for our results. Conversely, our study found a positive causal relationship between Verrucomicrobia, Paraprevotella, Alphaproteobacteria, and an increased risk of postpartum depression. The conclusions of Lukić(Lukic et al., 2019) and Zhang(Zhang et al., 2023) support this causal relationship. However, existing research findings on the causal relationship between Odoribacter(Thomann et al., 2022), Oxalobacter(Zou et al., 2022), Romboutsia(Qu et al., 2023; Xie et al., 2022) and depression contradict our results. These discrepancies may arise from differences in disease types, models, confounding factors, and reverse causality. Further research is required to explore possible explanatory mechanisms. Finally, we found that Prevotella7, RuminococcaceaeUCG011, and Eggerthella have been less studied in mood disorders. Their relationship research has great value and may be a potential research direction.

Causal effects of blood metabolites on PPD

Over the past decade, metabolomics research has continuously unveiled blood metabolic biomarkers associated with PPD. Our MR analyses provide compelling evidence, indicating a causal relationship between elevated levels of choline, malate, guanosine, tryptophan, and a reduced risk of PPD. Choline, a vital water-soluble B-vitamin, serves as a crucial component of human cell membranes, playing a pivotal role in maintaining the nervous system, cell structure, and lipid metabolism. Abnormal levels of choline have been linked to neurodegenerative and neuropsychiatric disorders(Sarter and Parikh, 2005). Clinical trials(Sun et al., 2023) and animal studies(Liu et al., 2023) consistently demonstrate elevated levels of choline in the plasma of patients with major depressive disorder or rodents exhibiting severe depression, supporting the positive causal relationship between choline and PPD. Malate, also known as malic acid, participates in energy metabolism and neurotransmitter synthesis, showcasing antioxidant properties. Studies have demonstrated increased malic acid levels in the plasma and liver of rats experiencing depression, highlighting a positive correlation with depressive symptoms(Hernandez-Baixauli et al., 2021; Liu et al., 2021). Our study reaffirmed a similar causal relationship.

Interestingly, existing research indicates that Guanosine can improve symptoms in depressed mice(Camargo et al., 2023; Dos et al., 2023). Plasma levels of Tryptophan significantly decrease in PPD patients or in PPD rats(Sha et al., 2022; Zhao et al., 2022). These findings contrast with our results, possibly due to differences in diseases, models, confounding factors, and reverse causality. Further investigation is crucial for clarifying and reconciling these differences. Additionally, there is limited research on the correlation of two metabolites (hyodeoxycholate and 2-hydroxybutyrate) in depressive disorder and PPD. This represents a promising avenue for future research and warrants further investigation.

Mediation analysis results

Our mediation analysis provides genetic evidence for the association between GM and blood metabolites. Mediation analysis revealed that the impact of Odoribacter on PPD indirectly through Hyodeoxycholate with a mediation proportion of 16.8%. Odoribacter is commonly associated with the production of beneficial short-chain fatty acids such as butyric acid, which contributes to anti-inflammatory effects, enhanced intestinal barrier function, and influences the composition and diversity of the intestinal microbial community(Dai et al., 2024). Compared to healthy pregnant women, the abundance of Odoribacter was significantly increased in pregnant women with fetal growth restriction(Tang et al., 2024). Increased abundance of Odoribacter may signal a microbiota imbalance that could be associated with elevated levels of inflammation, potentially affecting brain function and mood states(Malan-Muller et al., 2022; Zang et al., 2023).

Studies have shown that Odoribacter can participate in the metabolism of bile acids and can dehydrogenate bile acids to form secondary bile acids(Granado-Serrano et al., 2019; Zeng et al., 2022; Zhang et al., 2024). Hyodeoxycholate, a secondary bile acid derived from bile acids, acts on specific bile acid receptors, such as the farnesoid X receptor (FXR) and the G protein-coupled bile acid receptor(Xu et al., 2023). In the brain, FXR activation inhibits neuroinflammation and provides neuroprotection by decreasing the expression of NLRP3 inflammasome, caspase-1, and IL-1β(Bao et al., 2021). During the postpartum period, women's hormone levels change, which may affect the metabolism of bile acids, leading to elevated serum levels of bile acids. Excessive bile acids can be toxic to cells in the brain, leading to increased levels of inflammation, affecting neurological functioning, and indirectly influencing moods and behaviors(Dicks, 2023; Hualin et al., 2023).

Based on these analyses, we hypothesize that Odoribacter affects PPD through a neuroimmune response pathway, and that part of this effect is mediated by Hyodeoxycholate.

Validation set

By validating the causal relationships of 13 GM taxa associated with PPD, we found that only Alphaproteobacteria remained causally associated with PPD, thus increasing the robustness of the previous results. Additionally, Bifidobacteriales (IVW, p = 0.080), Bifidobacteriaceae (IVW, p = 0.080), Bifidobacterium (IVW, p = 0.054), and Clostridia (IVW, p = 0.099) were not causally associated with PPD. But we found a trend of statistical differences. No statistical significance (p < 0.05) may be related to their small sample sizes (196 GM sample sizes: 14306; Bifidobacteriales, Bifidobacteriaceae, and Bifidobacterium sample sizes: 7738; Clostridia sample sizes: 8956, see Supplementary Table 1). Other GM taxa were also not found to be causally related to PPD for two reasons: one being the small sample size of the validation data, and the other being the possibility that the previous causality was a false-positive result caused by Type I errors. The 196 GM GWAS data were obtained from the MiBioGen consortium, which conducted the largest, multiracial, full-genome meta-analysis of GM to date. Therefore, RuminococcaceaeUCG011 and Prevotella7 were not validated by appropriate data. So the significant or potential causal relationship between 196 GM taxa and PPD is highly plausible.

No causal relationship was found between blood metabolites and PPD in the validation of the causal relationships of 6 blood metabolites associated with PPD. Since the sample size of data before and after validation did not differ significantly (see Supplementary Table 1), further large-scale data, clinical trials, or animal studies are needed to validate our results.

Advantages and Limitation

Using MR analysis, our study systematically identified the causal relationships between GM, blood metabolites, and PPD for the first time, offering comprehensive screening data for these associations. Unlike traditional observational studies, this method significantly reduces bias from confounding variables and the risk of reverse causality. The identified causalities were validated using different GWAS databases for GM and blood metabolites, enhancing the robustness of the results. Additionally, we explored the mechanism linking GM, blood metabolites, and PPD through mediation analysis, uncovering that Odoribacter influences PPD indirectly via Hyodeoxycholate. This finding provides new insights for future research.

This study comes with certain limitations. Firstly, a significant portion of the GWAS summary data originated from European cohorts, while only a fraction of the GM information was derived from other ethnic backgrounds. This discrepancy might introduce a degree of bias into our observations. Secondly, due to constraints in the foundational dataset, our investigation was confined to the genus level, bypassing finer subdivisions such as species or strains. Nevertheless, the application of advanced shotgun metagenomic sequencing techniques holds promise for unveiling more precise and nuanced insights. Moreover, it's important to note that no experiments were conducted to validate the causal associations identified in this study. Therefore, mechanistic studies will be imperative in the future to authenticate these relationships.

To our knowledge, this is the first study to comprehensively assess the causal relationships between the GM, blood metabolites, and PPD. These findings highlight the importance of elucidating the underlying mechanisms linking gut microbiota, blood metabolites, and PPD. These results provide novel insights into microbiome-based therapies and metabolite-targeted interventions for PPD.

Acknowledgments

We express our gratitude for the time and effort contributed by participants during the phase of clinical data collection.

Funding Statement

This research was supported by the National Natural Science Foundation of China (82204958).

CRediT authorship contribution statement

Zhan Gao: Writing – original draft, Visualization, Validation, Methodology, Investigation, Formal analysis. Runze Zhou: Writing – original draft, Visualization, Validation, Methodology. Haotian Qian: Methodology, Formal analysis. Chendong Xu: Data curation. Mingzhou Gao: Project administration. Xi Huang: Methodology. ZhiQiang Chen: Writing – review & editing, Supervision, Investigation, Conceptualization.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data sharing is not applicable to this article.

Conflicts of Interest

The authors declare no conflict of interest.

Bao, H., Li, H., Jia, Y., Xiao, Y., Luo, S., Zhang, D., Han, L., Dai, L., Xiao, C., Feng, L., Feng, Y., Yang, Y., Wang, H., Wang, G., Du J, 2021. Ganoderic acid A exerted antidepressant-like action through FXR modulated NLRP3 inflammasome and synaptic activity. Biochem Pharmacol 188, 114561. https://doi.org/10.1016/j.bcp.2021.114561.
Bowden, J., Holmes, M.V., 2019. Meta-analysis and Mendelian randomization: A review. Res Synth Methods 10, 486-496. https://doi.org/10.1002/jrsm.1346.
Brann, E., Malavaki, C., Fransson, E., Ioannidi, M.K., Henriksson, H.E., Papadopoulos, F.C., Chrousos, G.P., Klapa, M.I., Skalkidou, A., 2021. Metabolic Profiling Indicates Diversity in the Metabolic Physiologies Associated With Maternal Postpartum Depressive Symptoms. Front Psychiatry 12, 685656. https://doi.org/10.3389/fpsyt.2021.685656.
Camargo, A., Bettio, L., Rosa, P.B., Rosa, J.M., Alte, G.A., Rodrigues, A., 2023. The antidepressant-like effect of guanosine involves the modulation of adenosine A(1) and A(2A) receptors. Purinerg Signal 19, 387-399. https://doi.org/10.1007/s11302-022-09898-8.
Cao, Y., Lu, H., Xu, W., Zhong, M., 2023. Gut microbiota and Sjogren's syndrome: a two-sample Mendelian randomization study. Front Immunol 14, 1187906. https://doi.org/10.3389/fimmu.2023.1187906.
Chao, G., Zhang, S., 2020. The characteristics of intestinal flora of IBS-D with different syndromes. Immun Inflamm Dis 8, 615-628. https://doi.org/10.1002/iid3.348.
Chen, D.L., Dai, Y.C., Zheng, L., Chen, Y.L., Zhang, Y.L., Tang, Z.P., 2021. Features of the gut microbiota in ulcerative colitis patients with depression: A pilot study. Medicine 100, e24845. https://doi.org/10.1097/MD.0000000000024845.
Dai, H., Hou, T., Wang, Q., Hou, Y., Wang, T., Zheng, J., Lin, H., Zhao, Z., Li, M., Wang, S., Zhang, D., Dai, M., Zheng, R., Lu, J., Xu, Y., Chen, Y., Ning, G., Wang, W., Bi, Y., Xu, M., 2023. Causal relationships between the gut microbiome, blood lipids, and heart failure: a Mendelian randomization analysis. Eur J Prev Cardiol 30, 1274-1282. https://doi.org/10.1093/eurjpc/zwad171.
Dai, J., Jiang, M., Wang, X., Lang, T., Wan, L., Wang, J., 2024. Human-derived bacterial strains mitigate colitis via modulating gut microbiota and repairing intestinal barrier function in mice. Bmc Microbiol 24, 96. https://doi.org/10.1186/s12866-024-03216-5.
Dicks, L., 2023. Our Mental Health Is Determined by an Intrinsic Interplay between the Central Nervous System, Enteric Nerves, and Gut Microbiota. Int J Mol Sci 25. https://doi.org/10.3390/ijms25010038.
Dos, S.R., de Lima, R.S., Gibbert, P.C., de Arruda, C.M., Doneda, D.L., de Matos, Y., Viola, G.G., Rios, S.F., de Lima, E., Da, S.B.Z., Vandresen-Filho, S., 2023. Guanosine treatment prevents lipopolysaccharide-induced depressive-like behavior in mice. J Psychiatr Res 164, 296-303. https://doi.org/10.1016/j.jpsychires.2023.06.022.
Emdin, C.A., Khera, A.V., Kathiresan, S., 2017. Mendelian Randomization. Jama-J Am Med Assoc 318, 1925-1926. https://doi.org/10.1001/jama.2017.17219.
Granado-Serrano, A.B., Martin-Gari, M., Sanchez, V., Riart, S.M., Berdun, R., Ludwig, I.A., Rubio, L., Vilaprinyo, E., Portero-Otin, M., Serrano, J., 2019. Faecal bacterial and short-chain fatty acids signature in hypercholesterolemia. Sci Rep-Uk 9, 1772. https://doi.org/10.1038/s41598-019-38874-3.
Hartwig, F.P., Davey, S.G., Bowden, J., 2017. Robust inference in summary data Mendelian randomization via the zero modal pleiotropy assumption. Int J Epidemiol 46, 1985-1998. https://doi.org/10.1093/ije/dyx102.
Hemani, G., Zheng, J., Elsworth, B., Wade, K.H., Haberland, V., Baird, D., Laurin, C., Burgess, S., Bowden, J., Langdon, R., Tan, V.Y., Yarmolinsky, J., Shihab, H.A., Timpson, N.J., Evans, D.M., Relton, C., Martin, R.M., Davey, S.G., Gaunt, T.R., Haycock, P.C., 2018. The MR-Base platform supports systematic causal inference across the human phenome. Elife 7. https://doi.org/10.7554/eLife.34408.
Hernandez-Baixauli, J., Puigbo, P., Abasolo, N., Palacios-Jordan, H., Foguet-Romero, E., Sunol, D., Galofre, M., Caimari, A., Baselga-Escudero, L., Bas, J., Mulero, M., 2021. Alterations in Metabolome and Microbiome Associated with an Early Stress Stage in Male Wistar Rats: A Multi-Omics Approach. Int J Mol Sci 22. https://doi.org/10.3390/ijms222312931.
Hualin, X., Yupin, X., Guoqiang, Z., Xukun, F., Hongmei, L., 2023. Intrahepatic cholestasis of pregnancy worsening perinatal depressive tendency: A follow-up study from the second trimester to the sixth week postpartum. Heliyon 9, e15845. https://doi.org/10.1016/j.heliyon.2023.e15845.
Kurilshikov, A., Medina-Gomez, C., Bacigalupe, R., Radjabzadeh, D., Wang, J., Demirkan, A., Le Roy, C.I., Raygoza, G.J., Finnicum, C.T., Liu, X., Zhernakova, D.V., Bonder, M.J., Hansen, T.H., Frost, F., Ruhlemann, M.C., Turpin, W., Moon, J.Y., Kim, H.N., Lull, K., Barkan, E., Shah, S.A., Fornage, M., Szopinska-Tokov, J., Wallen, Z.D., Borisevich, D., Agreus, L., Andreasson, A., Bang, C., Bedrani, L., Bell, J.T., Bisgaard, H., Boehnke, M., Boomsma, D.I., Burk, R.D., Claringbould, A., Croitoru, K., Davies, G.E., van Duijn, C.M., Duijts, L., Falony, G., Fu, J., van der Graaf, A., Hansen, T., Homuth, G., Hughes, D.A., Ijzerman, R.G., Jackson, M.A., Jaddoe, V., Joossens, M., Jorgensen, T., Keszthelyi, D., Knight, R., Laakso, M., Laudes, M., Launer, L.J., Lieb, W., Lusis, A.J., Masclee, A., Moll, H.A., Mujagic, Z., Qibin, Q., Rothschild, D., Shin, H., Sorensen, S.J., Steves, C.J., Thorsen, J., Timpson, N.J., Tito, R.Y., Vieira-Silva, S., Volker, U., Volzke, H., Vosa, U., Wade, K.H., Walter, S., Watanabe, K., Weiss, S., Weiss, F.U., Weissbrod, O., Westra, H.J., Willemsen, G., Payami, H., Jonkers, D., Arias, V.A., de Geus, E., Meyer, K.A., Stokholm, J., Segal, E., Org, E., Wijmenga, C., Kim, H.L., Kaplan, R.C., Spector, T.D., Uitterlinden, A.G., Rivadeneira, F., Franke, A., Lerch, M.M., Franke, L., Sanna, S., D'Amato, M., Pedersen, O., Paterson, A.D., Kraaij, R., Raes, J., Zhernakova, A., 2021. Large-scale association analyses identify host factors influencing human gut microbiome composition. Nat Genet 53, 156-165. https://doi.org/10.1038/s41588-020-00763-1.
Kurki, M.I., Karjalainen, J., Palta, P., Sipila, T.P., Kristiansson, K., Donner, K.M., Reeve, M.P., Laivuori, H., Aavikko, M., Kaunisto, M.A., Loukola, A., Lahtela, E., Mattsson, H., Laiho, P., Della, B.P.P., Lehisto, A.A., Kanai, M., Mars, N., Ramo, J., Kiiskinen, T., Heyne, H.O., Veerapen, K., Rueger, S., Lemmela, S., Zhou, W., Ruotsalainen, S., Parn, K., Hiekkalinna, T., Koskelainen, S., Paajanen, T., Llorens, V., Gracia-Tabuenca, J., Siirtola, H., Reis, K., Elnahas, A.G., Sun, B., Foley, C.N., Aalto-Setala, K., Alasoo, K., Arvas, M., Auro, K., Biswas, S., Bizaki-Vallaskangas, A., Carpen, O., Chen, C.Y., Dada, O.A., Ding, Z., Ehm, M.G., Eklund, K., Farkkila, M., Finucane, H., Ganna, A., Ghazal, A., Graham, R.R., Green, E.M., Hakanen, A., Hautalahti, M., Hedman, A.K., Hiltunen, M., Hinttala, R., Hovatta, I., Hu, X., Huertas-Vazquez, A., Huilaja, L., Hunkapiller, J., Jacob, H., Jensen, J.N., Joensuu, H., John, S., Julkunen, V., Jung, M., Junttila, J., Kaarniranta, K., Kahonen, M., Kajanne, R., Kallio, L., Kalviainen, R., Kaprio, J., Kerimov, N., Kettunen, J., Kilpelainen, E., Kilpi, T., Klinger, K., Kosma, V.M., Kuopio, T., Kurra, V., Laisk, T., Laukkanen, J., Lawless, N., Liu, A., Longerich, S., Magi, R., Makela, J., Makitie, A., Malarstig, A., Mannermaa, A., Maranville, J., Matakidou, A., Meretoja, T., Mozaffari, S.V., Niemi, M., Niemi, M., Niiranen, T., O, D.C., Obeidat, M.E., Okafo, G., Ollila, H.M., Palomaki, A., Palotie, T., Partanen, J., Paul, D.S., Pelkonen, M., Pendergrass, R.K., Petrovski, S., Pitkaranta, A., Platt, A., Pulford, D., Punkka, E., Pussinen, P., Raghavan, N., Rahimov, F., Rajpal, D., Renaud, N.A., Riley-Gillis, B., Rodosthenous, R., Saarentaus, E., Salminen, A., Salminen, E., Salomaa, V., Schleutker, J., Serpi, R., Shen, H.Y., Siegel, R., Silander, K., Siltanen, S., Soini, S., Soininen, H., Sul, J.H., Tachmazidou, I., Tasanen, K., Tienari, P., Toppila-Salmi, S., Tukiainen, T., Tuomi, T., Turunen, J.A., Ulirsch, J.C., Vaura, F., Virolainen, P., Waring, J., Waterworth, D., Yang, R., Nelis, M., Reigo, A., Metspalu, A., Milani, L., Esko, T., Fox, C., Havulinna, A.S., Perola, M., Ripatti, S., Jalanko, A., Laitinen, T., Makela, T.P., Plenge, R., McCarthy, M., Runz, H., Daly, M.J., Palotie, A., 2023. FinnGen provides genetic insights from a well-phenotyped isolated population. Nature 613, 508-518. https://doi.org/10.1038/s41586-022-05473-8.
Lawlor, D.A., Harbord, R.M., Sterne, J.A., Timpson, N., Davey, S.G., 2008. Mendelian randomization: using genes as instruments for making causal inferences in epidemiology. Stat Med 27, 1133-1163. https://doi.org/10.1002/sim.3034.
Li, J., Zhang, C., Tang, J., He, M., He, C., Pu, G., Liu, L., Sun, J., 2024. Causal associations between gut microbiota, metabolites and asthma: a two-sample Mendelian randomization study. Bmc Pulm Med 24, 72. https://doi.org/10.1186/s12890-024-02898-x.
Li, Z., Lai, J., Zhang, P., Ding, J., Jiang, J., Liu, C., Huang, H., Zhen, H., Xi, C., Sun, Y., Wu, L., Wang, L., Gao, X., Li, Y., Fu, Y., Jie, Z., Li, S., Zhang, D., Chen, Y., Zhu, Y., Lu, S., Lu, J., Wang, D., Zhou, H., Yuan, X., Li, X., Pang, L., Huang, M., Yang, H., Zhang, W., Brix, S., Kristiansen, K., Song, X., Nie, C., Hu, S., 2022. Multi-omics analyses of serum metabolome, gut microbiome and brain function reveal dysregulated microbiota-gut-brain axis in bipolar depression. Mol Psychiatr 27, 4123-4135. https://doi.org/10.1038/s41380-022-01569-9.
Li, Z., Zhu, G., Lei, X., Tang, L., Kong, G., Shen, M., Zhang, L., Song, L., 2023. Genetic support of the causal association between gut microbiome and COVID-19: a bidirectional Mendelian randomization study. Front Immunol 14, 1217615. https://doi.org/10.3389/fimmu.2023.1217615.
Lin, S.H., Brown, D.W., Machiela, M.J., 2020. LDtrait: An Online Tool for Identifying Published Phenotype Associations in Linkage Disequilibrium. Cancer Res 80, 3443-3446. https://doi.org/10.1158/0008-5472.CAN-20-0985.
Liskiewicz, P., Kaczmarczyk, M., Misiak, B., Wronski, M., Baba-Kubis, A., Skonieczna-Zydecka, K., Marlicz, W., Bienkowski, P., Misera, A., Pelka-Wysiecka, J., Kucharska-Mazur, J., Konopka, A., Loniewski, I., Samochowiec, J., 2021. Analysis of gut microbiota and intestinal integrity markers of inpatients with major depressive disorder. Prog Neuro-Psychoph 106, 110076. https://doi.org/10.1016/j.pnpbp.2020.110076.
Liu, X., Wang, S., Wang, G., 2022. Prevalence and Risk Factors of Postpartum Depression in Women: A Systematic Review and Meta-analysis. J Clin Nurs 31, 2665-2677. https://doi.org/10.1111/jocn.16121.
Liu, X., Zhang, B., Tian, J., Han, Y., 2023. Plasma metabolomics reveals the intervention mechanism of different types of exercise on chronic unpredictable mild stress-induced depression rat model. Metab Brain Dis. https://doi.org/10.1007/s11011-023-01310-7.
Liu, X.J., Liu, H.L., Zhao, D., Wei, F.X., Wang, Y.Z., Lv, M., Chen, S.J., Li, S.Y., Qin, X.M., 2021. Hepatic metabolomics of the compatibility effect of Xiaoyaosan on CUMS-induced depression based on the TCM theory of "Treating Diseases via Regulating the Liver's Function". J Pharmaceut Biomed 201, 114123. https://doi.org/10.1016/j.jpba.2021.114123.
Long, Y., Tang, L., Zhou, Y., Zhao, S., Zhu, H., 2023. Causal relationship between gut microbiota and cancers: a two-sample Mendelian randomisation study. Bmc Med 21, 66. https://doi.org/10.1186/s12916-023-02761-6.
Lukic, I., Getselter, D., Ziv, O., Oron, O., Reuveni, E., Koren, O., Elliott, E., 2019. Antidepressants affect gut microbiota and Ruminococcus flavefaciens is able to abolish their effects on depressive-like behavior. Transl Psychiat 9, 133. https://doi.org/10.1038/s41398-019-0466-x.
Luo, M., Cai, J., Luo, S., Hong, X., Xu, L., Lin, H., Chen, X., Fu, W., 2023. Causal effects of gut microbiota on the risk of chronic kidney disease: a Mendelian randomization study. Front Cell Infect Mi 13, 1142140. https://doi.org/10.3389/fcimb.2023.1142140.
Malan-Muller, S., Valles-Colomer, M., Foxx, C.L., Vieira-Silva, S., van den Heuvel, L.L., Raes, J., Seedat, S., Lowry, C.A., Hemmings, S., 2022. Exploring the relationship between the gut microbiome and mental health outcomes in a posttraumatic stress disorder cohort relative to trauma-exposed controls. Eur Neuropsychopharm 56, 24-38. https://doi.org/10.1016/j.euroneuro.2021.11.009.
Payne, J.L., Maguire, J., 2019. Pathophysiological mechanisms implicated in postpartum depression. Front Neuroendocrin 52, 165-180. https://doi.org/10.1016/j.yfrne.2018.12.001.
Qu, Z., Wu, S., Zheng, Y., Bing, Y., Liu, X., Li, S., Li, W., Zou, X., 2023. Fecal metabolomics combined with metagenomics sequencing to analyze the antidepressant mechanism of Yueju Wan. J Pharmaceut Biomed 238, 115807. https://doi.org/10.1016/j.jpba.2023.115807.
Ramsteijn, A.S., Jasarevic, E., Houwing, D.J., Bale, T.L., Olivier, J.D., 2020. Antidepressant treatment with fluoxetine during pregnancy and lactation modulates the gut microbiome and metabolome in a rat model relevant to depression. Gut Microbes 11, 735-753. https://doi.org/10.1080/19490976.2019.1705728.
Rees, J., Wood, A.M., Burgess, S., 2017. Extending the MR-Egger method for multivariable Mendelian randomization to correct for both measured and unmeasured pleiotropy. Stat Med 36, 4705-4718. https://doi.org/10.1002/sim.7492.
Rodriguez-Martin, I., Villanueva-Martin, G., Guillen-Del-Castillo, A., Ortego-Centeno, N., Callejas, J.L., Simeon-Aznar, C.P., Martin, J., Acosta-Herrera, M., 2023. Contribution of Telomere Length to Systemic Sclerosis Onset: A Mendelian Randomization Study. Int J Mol Sci 24. https://doi.org/10.3390/ijms242115589.
Sanderson, E., 2021. Multivariable Mendelian Randomization and Mediation. Csh Perspect Med 11. https://doi.org/10.1101/cshperspect.a038984.
Sarter, M., Parikh, V., 2005. Choline transporters, cholinergic transmission and cognition. Nat Rev Neurosci 6, 48-56. https://doi.org/10.1038/nrn1588.
Sha, Q., Madaj, Z., Keaton, S., Escobar, G.M., Smart, L., Krzyzanowski, S., Fazleabas, A.T., Leach, R., Postolache, T.T., Achtyes, E.D., Brundin, L., 2022. Cytokines and tryptophan metabolites can predict depressive symptoms in pregnancy. Transl Psychiat 12, 35. https://doi.org/10.1038/s41398-022-01801-8.
Shin, S.Y., Fauman, E.B., Petersen, A.K., Krumsiek, J., Santos, R., Huang, J., Arnold, M., Erte, I., Forgetta, V., Yang, T.P., Walter, K., Menni, C., Chen, L., Vasquez, L., Valdes, A.M., Hyde, C.L., Wang, V., Ziemek, D., Roberts, P., Xi, L., Grundberg, E., Waldenberger, M., Richards, J.B., Mohney, R.P., Milburn, M.V., John, S.L., Trimmer, J., Theis, F.J., Overington, J.P., Suhre, K., Brosnan, M.J., Gieger, C., Kastenmuller, G., Spector, T.D., Soranzo, N., 2014. An atlas of genetic influences on human blood metabolites. Nat Genet 46, 543-550. https://doi.org/10.1038/ng.2982.
Silverman, M.E., Reichenberg, A., Savitz, D.A., Cnattingius, S., Lichtenstein, P., Hultman, C.M., Larsson, H., Sandin, S., 2017. The risk factors for postpartum depression: A population-based study. Depress Anxiety 34, 178-187. https://doi.org/10.1002/da.22597.
Skrivankova, V.W., Richmond, R.C., Woolf, B., Yarmolinsky, J., Davies, N.M., Swanson, S.A., VanderWeele, T.J., Higgins, J., Timpson, N.J., Dimou, N., Langenberg, C., Golub, R.M., Loder, E.W., Gallo, V., Tybjaerg-Hansen, A., Davey, S.G., Egger, M., Richards, J.B., 2021. Strengthening the Reporting of Observational Studies in Epidemiology Using Mendelian Randomization: The STROBE-MR Statement. Jama-J Am Med Assoc 326, 1614-1621. https://doi.org/10.1001/jama.2021.18236.
Stewart, D.E., Vigod, S.N., 2019. Postpartum Depression: Pathophysiology, Treatment, and Emerging Therapeutics. Annu Rev Med 70, 183-196. https://doi.org/10.1146/annurev-med-041217-011106.
Sun, X.L., Ma, L.N., Chen, Z.Z., Xiong, Y.B., Jia, J., Wang, Y., Ren, Y., 2023. Search for serum biomarkers in patients with bipolar disorder and major depressive disorder using metabolome analysis. Front Psychiatry 14, 1251955. https://doi.org/10.3389/fpsyt.2023.1251955.
Tang, H., Li, D., Peng, J., Yang, W., Zhang, X., Li, H., 2024. Potential Association of Gut Microbial Metabolism and Circulating mRNA Based on Multiomics Sequencing Analysis in Fetal Growth Restriction. Mediat Inflamm 2024, 9986187. https://doi.org/10.1155/2024/9986187.
Thomann, A.K., Wustenberg, T., Wirbel, J., Knoedler, L.L., Thomann, P.A., Zeller, G., Ebert, M.P., Lis, S., Reindl, W., 2022. Depression and fatigue in active IBD from a microbiome perspective-a Bayesian approach to faecal metagenomics. Bmc Med 20, 366. https://doi.org/10.1186/s12916-022-02550-7.
Tian, X.Y., Xing, J.W., Zheng, Q.Q., Gao, P.F., 2021. 919 Syrup Alleviates Postpartum Depression by Modulating the Structure and Metabolism of Gut Microbes and Affecting the Function of the Hippocampal GABA/Glutamate System. Front Cell Infect Mi 11, 694443. https://doi.org/10.3389/fcimb.2021.694443.
Verbanck, M., Chen, C.Y., Neale, B., Do, R., 2018. Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nat Genet 50, 693-698. https://doi.org/10.1038/s41588-018-0099-7.
Vicariotto, F., Malfa, P., Torricelli, M., Lungaro, L., Caio, G., De Leo, V., 2023. Beneficial Effects of Limosilactobacillus reuteri PBS072 and Bifidobacterium breve BB077 on Mood Imbalance, Self-Confidence, and Breastfeeding in Women during the First Trimester Postpartum. Nutrients 15. https://doi.org/10.3390/nu15163513.
Wu, X., Lin, D., Li, Q., Cai, J., Huang, H., Xiang, T., Tan, H., 2023. Investigating causal associations among gut microbiota, gut microbiota-derived metabolites, and gestational diabetes mellitus: a bidirectional Mendelian randomization study. Aging (Albany Ny) 15, 8345-8366. https://doi.org/10.18632/aging.204973.
Xie, A., Ensink, E., Li, P., Gordevicius, J., Marshall, L.L., George, S., Pospisilik, J.A., Aho, V., Houser, M.C., Pereira, P., Rudi, K., Paulin, L., Tansey, M.G., Auvinen, P., Brundin, P., Brundin, L., Labrie, V., Scheperjans, F., 2022. Bacterial Butyrate in Parkinson's Disease Is Linked to Epigenetic Changes and Depressive Symptoms. Movement Disord 37, 1644-1653. https://doi.org/10.1002/mds.29128.
Xie, J., Huang, H., Liu, Z., Li, Y., Yu, C., Xu, L., Xu, C., 2023. The associations between modifiable risk factors and nonalcoholic fatty liver disease: A comprehensive Mendelian randomization study. Hepatology 77, 949-964. https://doi.org/10.1002/hep.32728.
Xu, H., Fang, F., Wu, K., Song, J., Li, Y., Lu, X., Liu, J., Zhou, L., Yu, W., Yu, F., Gao, J., 2023. Gut microbiota-bile acid crosstalk regulates murine lipid metabolism via the intestinal FXR-FGF19 axis in diet-induced humanized dyslipidemia. Microbiome 11, 262. https://doi.org/10.1186/s40168-023-01709-5.
Xu, Q., Sun, L., Chen, Q., Jiao, C., Wang, Y., Li, H., Xie, J., Zhu, F., Wang, J., Zhang, W., Xie, L., Wu, H., Zuo, Z., Chen, X., 2024. Gut microbiota dysbiosis contributes to depression-like behaviors via hippocampal NLRP3-mediated neuroinflammation in a postpartum depression mouse model. Brain Behav Immun 119, 220-235. https://doi.org/10.1016/j.bbi.2024.04.002.
Xu, X., Wu, L.Y., Wang, S.Y., Yan, M., Wang, Y.H., Li, L., Sun, Z.L., Zhao, J.X., 2024. Investigating causal associations among gut microbiota, metabolites, and psoriatic arthritis: a Mendelian randomization study. Front Microbiol 15, 1287637. https://doi.org/10.3389/fmicb.2024.1287637.
Zang, Y., Lai, X., Li, C., Ding, D., Wang, Y., Zhu, Y., 2023. The Role of Gut Microbiota in Various Neurological and Psychiatric Disorders-An Evidence Mapping Based on Quantified Evidence. Mediat Inflamm 2023, 5127157. https://doi.org/10.1155/2023/5127157.
Zeng, Z., Zhou, Y., Xu, Y., Wang, S., Wang, B., Zeng, Z., Wang, Q., Ye, X., Jin, L., Yue, M., Tang, L., Zou, P., Zhao, P., Li, W., 2022. Bacillus amyloliquefaciens SC06 alleviates the obesity of ob/ob mice and improves their intestinal microbiota and bile acid metabolism. Food Funct 13, 5381-5395. https://doi.org/10.1039/d1fo03170h.
Zhang, M., Li, H., Tan, T., Lu, L., Mi, J., Rehman, A., Yan, Y., Ran, L., 2024. Anthocyanins from Lycium ruthenicum Murray attenuates high-fat diet-induced hypercholesterolemia in ApoE(-/-) mice are related to the modulation of gut microbiota and the ratio of conjugated to unconjugated bile acids in fecal bile acid profile. Food Sci Nutr 12, 2379-2392. https://doi.org/10.1002/fsn3.3923.
Zhang, S., Lu, B., Wang, G., 2023. The role of gut microbiota in the pathogenesis and treatment of postpartum depression. Ann Gen Psychiatr 22, 36. https://doi.org/10.1186/s12991-023-00469-8.
Zhao, R., Zhou, Y., Shi, H., Ye, W., Lyu, Y., Wen, Z., Li, R., Xu, Y., 2022. Effect of Gestational Diabetes on Postpartum Depression-like Behavior in Rats and Its Mechanism. Nutrients 14. https://doi.org/10.3390/nu14061229.
Zhao, X.H., Zhang, Z.H., 2020. Risk factors for postpartum depression: An evidence-based systematic review of systematic reviews and meta-analyses. Asian J Psychiatr 53, 102353. https://doi.org/10.1016/j.ajp.2020.102353.
Zheng, Q., Wang, S., Tian, X., Liu, W., Gao, P., 2023. Fecal microbiota transplantation confirmed that 919 Syrup reduced the ratio of erucamide to 5-AVAB in hippocampus to alleviate postpartum depression by regulating gut microbes. Front Immunol 14, 1203015. https://doi.org/10.3389/fimmu.2023.1203015.
Zhou, L., Wu, Z., Zhao, D., Wang, G., Xiao, L., Wang, H., Xie, Y., Sun, L., 2023. Brief pup separation during lactation confers resilience in behavioural deficits induced by chronic restraint stress in postpartum C57BL/6J dams. J Psychiatr Neurosci 48, E154-E170. https://doi.org/10.1503/jpn.220079.
Zhou, Y., Chen, C., Yu, H., Yang, Z., 2020. Fecal Microbiota Changes in Patients With Postpartum Depressive Disorder. Front Cell Infect Mi 10, 567268. https://doi.org/10.3389/fcimb.2020.567268.
Zou, X.Y., Zhang, M., Tu, W.J., Zhang, Q., Jin, M.L., Fang, R.D., Jiang, S., 2022. Bacillus subtilis inhibits intestinal inflammation and oxidative stress by regulating gut flora and related metabolites in laying hens. Animal 16, 100474. https://doi.org/10.1016/j.animal.2022.100474.
Zuo, M., Wang, Z., Li, W., Chen, S., Yuan, Y., Yang, Y., Mao, Q., Liu, Y., 2023. Causal effects of potential risk factors on postpartum depression: a Mendelian randomization study. Front Psychiatry 14, 1275834. https://doi.org/10.3389/fpsyt.2023.1275834.

No competing interests reported.

SupplementalMaterials.xlsx

Download PDF

Version 1

posted

You are reading this latest preprint version

Dissecting Causal Relationships Between Gut Microbiota, Blood Metabolites, and Postpartum Depression : A Mendelian Randomization Study

Status:

Version 1

Abstract

Figures

Introduction

Methods

Study design

Data sources

IVs selection

Mendelian randomization and mediation analysis

Results

Causal effects of GM on PPD

Causal effects of blood metabolites on PPD

Mediation analysis results

Validation set

Discussion

Causal effects of GM on PPD

Causal effects of blood metabolites on PPD

Mediation analysis results

Validation set

Advantages and Limitation

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1