Investigating the Causal Relationship and Shared Genetic Basis Between Major Depression Disorder and Eight Types of Gastrointestinal Diseases Authors: Fu Sun, Haochang Li, Bin Gong, Shirong Hui, Ran He, Meijie Yu, Yihao Li, Sheng Yang, Peng Huang, Sreeja Dattachoudhury Published: 2025 ABSTRACT AbstractBackgroundPrevious studies on the causal relationship or shared genetic basis between major depression disorder (MDD) and gastrointestinal (GI) diseases covered either a limited range and not comprehensive enough.MethodsWe used linkage disequilibrium score regression (LDSC) to estimate the heritability for nine traits and the genetic correlation (r g ) between MDD and eight GI diseases, respectively. We further conducted a two‐sample Mendelian randomization (MR) analysis to identify the causal relationship between MDD and eight GI diseases. Finally, based on the result of MR, we performed stratified LDSC (S‐LDSC) to estimate the partitioned heritability and significantly enriched tissues, transcriptome‐wide association study (TWAS) to define shared genes, and colocalization analysis to define the pleiotropic single‐nucleotide polymorphisms (SNPs) and genes.ResultsFor the heritability, heritability of all nine traits was significant. For genetic correlation, six GI diseases showed significant correlations with MDD. For the result of MR, we revealed the causal relationship between MDD and acute appendicitis (OR = 1.09), irritable bowel syndrome (IBS) (OR = 1.14), and ulcer of esophagus (OR = 1.24). Additionally, we found no significant overlapping tissues after S‐LDSC. Finally, we defined three shared genes: PRSS16, ZNF602P, and ZNF204P by TWAS and nine pleiotropic genes C4A, FLOT1, LINC00243, MICB, and PRSS16 by colocalization analysis between MDD and acute appendicitis.ConclusionsOur findings provided the evidence of genetic association, causal relationship, and shared pleiotropic genes between MDD and GI diseases especially acute appendicitis, offering new insights into our understanding of their shared genetic basis.Keywords: causal relationship, gastrointestinal diseases, major depression disorder, shared genetic basis 1. Introduction Gastrointestinal (GI) diseases are a collection of diseases originating from digestive dysfunctions, posing a significant threat to human health [1]. The pathogenesis of GI diseases is exceedingly complex, involving multiple contributing factors. Based on cohort and case‐control studies, emerging studies have suggested that major depression disorder (MDD) is a significant risk factor for certain GI disorders [2–4], including irritable bowel syndrome (IBS) [5], gastroesophageal reflux disease (GORD) [6], and peptic ulcers [7]. However, observational studies face limitations, such as confounding factors and the lack of randomized design and genetic basis evidence [8]. Therefore, stronger evidence especially the genetic evidence needed to confirm the causal relationship between MDD and GI diseases. In the past 20 years, genome‐wide association studies (GWASs) have effectively identified the association between genetic variants and numerous complex traits [9], providing novel insights into the pathogenesis of diseases [10]. In terms of MDD, the two largest GWASs discovered 44 and 102 risk loci associated with the disorder, respectively [11, 12]. Amount summary statistics of GWAS provides the researcher more opportunity to investigate the heritability, stratified heritability, causal inference between two complex diseases, and the association between gene expression trait based on single‐nucleotide polymorphisms (SNPs). Linkage disequilibrium score regression (LDSC) estimates trait heritability and global genetic correlations between traits [13]. The stratified LDSC (S‐LDSC) explores heritability enrichment in specific cell and tissue types from SNPs [14]. Mendelian randomization (MR) uses genetic variants to explain causal relationships between exposures and outcomes [15]. Transcriptome‐wide association study (TWAS) integrates GWASs with expression quantitative trait loci (eQTLs) to investigate the relationship between tissue‐specific gene expression and trait variation [16]. Colocalization analysis employs Bayesian statistical methods to detect shared genetic variants in common genomic regions across traits [17]. Several studies have used these methods to explore the shared genetic basis of MDD and GI diseases. For example, a two‐sample MR study has demonstrated a causal relationship between MDD and an increased risk of 12 out of 24 GI diseases examined [18]. Another comprehensive analysis has further revealed the pleiotropic genes and the role of the gut–brain axis in mediating the relationship between MDD and GORD, IBS, peptic ulcer disease (PUD), and nonalcoholic fatty liver disease (NAFLD) [19]. However, while the former study covered a wide range of diseases, it only explored the causal relationship; the latter study employed more comprehensive methods but was limited to a few GI diseases. Therefore, it is necessary to investigate in depth the complex genetic mechanisms underlying the comorbidity between MDD and GI diseases, to provide additional insights into the prevention of GI diseases. In our study, we adopted a comprehensive approach integrating various methodologies including LDSC, MR, S‐LDSC, TWAS, and colocalization analysis to elucidate the shared genetic basis and potential causal relationship underlying MDD and eight GI diseases (i.e., Crohn disease, gastric ulcer, GORD, gastroduodenal ulcer, ulcerative colitis, acute appendicitis, IBS, and ulcer of esophagus). We aimed to deeply understand of the shared genetics between MDD and GI diseases. The detailed flowchart in this study is shown in Figure 1. This study was conducted using publicly available GWAS summary statistics; thus, individual‐level data were not accessible, and a conventional participant flow diagram illustrating individual inclusion and exclusion is not applicable. The comprehensive analytical workflow, spanning from data acquisition to final result generation, is visually summarized in Figure 1. The analytical flowchart in our study. We collected MDD summary datasets from PGC and GI diseases from FinnGen R9. First, we used LDSC to estimate heritability and genetic correlation. Then, we used MR and a series of sensitivity analyses to define the causal relationship between MDD and each GI disease. Finally, integrating gene expression and eQTL summary statistics of GTEx, we used S‐LDSC, FUSION, and coloc to define the enrichment tissues, shared significant genes, and shared pleiotropic genes, respectively. MDD, major depression disorder; GI, gastrointestinal; LDSC, linkage disequilibrium score regression; MR, Mendelian randomization; eQTL, expression quantitative trait loci; S‐LDSC, stratified linkage disequilibrium score regression. 2. Methods 2.1. Data Collection and Processing In the analysis, we used three kinds of data resources: summary statistics for the MDD, summary statistics for eight GI diseases, and reference LD panel. Specifically, the GWAS summary statistics for MDD were obtained from a meta‐analysis conducted by the Psychiatric Genomics Consortium (https://pgc.unc.edu/), which included case–control studies from multiple cohorts of European ancestry [11]. The sample comprised 135,458 cases and 344,901 controls. The PGC MDD GWAS meta‐analysis reported minimal evidence of heterogeneity across the contributing cohorts, as detailed in the original publication. The summary statistics of eight GI diseases were downloaded from FinnGen GWAS summary statistics (R9) (https://storage.googleapis.com/finngen-public-data-r9/summary_stats/), a large Finnish cohort study that combines genetic data with digital health record data from national health registries of European ancestry [20], including Crohn disease, gastric ulcer, GORD, gastroduodenal ulcer, ulcerative colitis, acute appendicitis, IBS and ulcer of esophagus. The specific number of cases and controls for each GI disease is provided in Table 1. The underlying study design for both data sources is case–control. Power calculations for the MR analysis were not performed a priori as the analysis utilized publicly available summary statistics with fixed sample sizes. However, the large sample sizes for most traits (e.g., MDD: N = 480,359; acute appendicitis: N = 375,028) provide substantial power to detect genetic correlations and causal effects of moderate magnitude. Missing genetic data in the GWAS summary statistics were handled by the original studies (PGC and FinnGen). In our analysis, we carried out strict quality control for these data. Specifically, we filtered out SNPs (1) with minor allele fraction (MAF) < 0.01, (2) with INFO < 0.3, (3) that are not included in the reference panel, and (4) that are duplicated, which are standard quality control procedures to minimize the impact of poorly imputed or rare variants​ [21–23]. Phenotypic data were complete in the summary statistics as they represent case–control status. In this study, we considered 1000 Genomes Project Phase3 European population SNPs as the reference panel. For the reference panel, we used 503 EUR individuals to estimate the LD matrix [24]. Table 1Summary for the MDD and eight gastrointestinal diseases.DiseaseNo. of casesNo. of controlsnSNPs h 2 p Major depressive disorder135,458344,9018,710,6660.02170.0014Crohn disease1655375,4459,199,2530.00830.0016Gastric ulcer5935320,3879,198,6500.00430.0014Gastroesophageal reflux disease26,184320,3879,198,7610.02300.0018Gastroduodenal ulcer9216320,3879,198,7690.00620.0015Ulcerative colitis5034371,5309,199,2180.01670.0026Acute appendicitis28,745346,2839,199,2490.01210.0019Irritable bowel syndrome9323301,9319,198,9700.00920.0017Ulcer of esophagus2064320,3879,198,6310.00500.0014 Summary for the MDD and eight gastrointestinal diseases. 2.2. Heritability and Genetic Correlation We used LDSC to estimate the global heritability for each disease and the genetic correlation (r g ) between MDD and eight GI diseases, respectively. Also, we used S‐LDSC to estimate the partitioned heritability and significantly enriched tissues. Specifically, integrated the summary statistics of eQTLs from GTEx V8 (http://gusevlab.org/projects/fusion/#gtex-v8-multi-tissue-expression) [25], S‐LDSC defined the associated tissues among 25 selected tissues using the top significant genes [14]. Considering the MDD and GI diseases, we selected 15 tissues from the nervous system (i.e., brain amygdala, brain anterior cingulate cortex, brain caudate basal ganglia, brain cerebellar hemisphere, brain cerebellum, brain cortex, brain frontal cortex, brain hippocampus, brain hypothalamus, brain nucleus accumbens basal ganglia, brain putamen basal ganglia, brain spinal cord cervical, brain substantia nigra, and nerve tibial, pituitary), nine tissues from digestive system (i.e., colon sigmoid, colon transverse, esophagus gastroesophageal junction, esophagus mucosa, esophagus muscularis, liver, pancreas, small intestine terminal ileum, and stomach), and whole blood. We defined the differentially expressed genes (DEGs) by the Wilcoxon test and set the SNP annotation as 1 if it located within 100 KB of DEGs. Following Ref. [26], we added the focal functional categories to the baseline model and model of top 5% or 10% genes. Finally, we used S‐LDSC to estimate the enrichment for each tissue. 2.3. MR Analysis To define the causal relationship between MDD to eight GI diseases, we used the two‐sample MR analysis. Given that the exposure and outcome data were derived from distinct consortia and predominantly nonoverlapping populations, the overlap of individuals between these two samples is expected to be minimal or null, thus satisfying the key assumption of two‐sample MR. The crucial step of MR was performed to select genetic variants as instruction variables (IVs) for exposure. Following Refs. [27–29], we retained SNPs with (1) strong association (p value < 1 × 10−5) in MDD with r 2 less than 0.001 in the range of 10 Mb and (2) strong IVs with F‐statistics > 10 [30]. To ensure that the genetic variant–exposure associations were comparable between the two samples, we performed rigorous data harmonization by aligning all effect alleles to the same reference strand, filtering out SNPs with palindrome alleles and SNPs that were strongly correlated with outcome (p < 1 × 10−5), thereby minimizing biases arising from differences between consortia. To obtain robust result, we carried out five models: MR‐Egger method [31], weighted median (WM) method [32], inverse variance weighting (IVW) method [33], simple mode method, and weighted mode method [34]. We mainly considered the IVW method as the main result [33, 35]. In parallel, we performed three sensitivity analyses: (i) heterogeneity test—to test the heterogeneity among IVs, (ii) MR pleiotropy residual sum and outlier (MR‐PRESSO)—to find outliers and test the level of pleiotropy, and (iii) leave‐one‐out (LOO) test—to observe whether the result significantly changed after removing each SNP [36, 37]. For these sensitivity analyses, we considered 0.05 as the significant level of statistic tests. In addition, we performed reverse‐direction MR to assess potential reverse causal effects of GI diseases on MDD. Parameters of selecting IVs and analysis procedure, as well as sensitivity analyses are the same as those in the forward direction MR analysis. 2.4. TWAS Based on the results of MR, we only defined shared significant genes among the causal traits. FUSION, defining the top eQTL with the best prediction model, was used to fit TWAS [38]. We used the same eQTL summary statistics from 25 tissues and the same reference panel in S‐LDSC. For each trait pair, we needed the shared genes with statistical significance. 2.5. Colocalization Integrating summary statistics of eQTLs, we used the colocalization analysis to define the pleiotropic SNPs for each trait. The colocalization analysis includes five model assumptions: (1) H0: no association, (2) H1: only association with Trait 1, (3) H2: only association with Trait 2, (4) H3: association with Trait 1 and Trait 2, distinct causal variants, and (5) H4: association with Trait 1 and Trait 2, shared causal variants [39]. Following Ref. [40], we considered the result of PP.H4 above 0.75 as significant results. The eQTLs we used in the colocalization analysis are consistent with those used in TWAS. We selected around 1 MB of the starting location of the significant candidate genes found in TWAS as the candidate region for the colocalization analysis. 2.6. Statistical Analysis We used LDSC (Version 1.0.1) to estimate heritability (h 2), genetic correlation, and tissue type of SNP heritability enrichment in GWAS summary statistics. We used TwoSampleMR (Version 0.5.7) to perform two‐sample MR analysis and coloc (Version 5.2.3) to perform the colocalization analysis. We used FUSION (Version 0.0.1) to perform TWAS. All the analyses are performed by R software (Version 4.3.0). We performed multiple corrections to the results of genetic correlation, MR (including direct and reverse) and TWAS by using the false discovery rate (FDR) [41]. We considered 0.05 as the significant level of our statistic tests as well as FDR correction. 3. Results 3.1. Heritability and Genetic Correlation Summary statistics for the major depressive disorder and eight GI diseases, including sample sizes and heritability estimates, are presented in Table 1. We estimated the heritability for nine traits (Table 1) and genetic correlation (Figure 2 and Table S1) for eight trait pairs. For the heritability, heritability of all nine traits was significant. Specifically, h 2 for MDD was 0.0217 with p = 0.0014. h 2 for the eight GI diseases ranged from 0.0167 (p = 0.0026) for ulcerative colitis to 0.043 (p = 0.0014) for gastric ulcer. For genetic correlation, there were six significant results between MDD and eight GI diseases and the results were still significant after FDR. Specifically, the genetic correlation between MDD and eight GI diseases ranged from 0.2003 (p = 0.0005) to 0.5446 (p = 5.71 × 10−11). The genetic correlation of nine traits using linkage disequilibrium score regression. The heat map shows r g (lower triangle) and its p value (upper triangle). The different colors represent different values of r g and p. The red color shows a more positive correlation, while the yellow color shows a more negative correlation. 3.2. MR Analysis We strictly followed the procedures in our methods, and we eventually obtained 55 IVs of MDD, which includes effect alleles, beta coefficients, standard errors, and p values on the log‐odds scale (Table S2). The mean F‐statistic for these instruments was 28.7 (range: 15.3–45.2), substantially above the conventional threshold of 10, indicating a low risk of weak instrument bias. Based on the IVW model, we defined MDD was a risky factor for five GI diseases by FDR threshold, including acute appendicitis (OR = 1.08, 95% CI: 1.02–1.15, p = 0.006, FDR = 0.024), gastroduodenal ulcer (OR = 1.10, 95% CI: 1.01–1.20, p = 0.029, FDR = 0.041), GORD (OR = 1.09, 95% CI: 1.02–1.16, p = 0.011, FDR = 0.028), IBS (OR = 1.13, 95% CI: 1.04–1.23, p = 0.005, FDR = 0.024), and ulcer of esophagus (OR = 1.24, 95% CI: 1.03–1.50, p = 0.025, FDR = 0.041) (Table S3). To assess the validity of the MR assumptions, particularly, the absence of horizontal pleiotropy, we conducted comprehensive sensitivity analyses. For three causal associations that passed multiple testing correction (MDD on acute appendicitis, IBS, and esophageal ulcer), Cochran’s Q statistic indicated no significant heterogeneity among the variant‐specific estimates (all p > 0.05), suggesting that a fixed‐effect IVW model was appropriate. The MR‐Egger intercept test revealed no evidence of directional pleiotropy (all p > 0.05). Furthermore, the MR‐PRESSO global test detected no significant outliers (all p > 0.05), and the results remained consistent after correcting for potential outliers. The results of all direct sensitivity analyses are shown in Table S4. Besides, the WM and MR‐Egger methods yielded effect estimates that were directionally consistent with the IVW method, providing additional confidence in the primary findings. The F‐statistics for all instruments were well above 10 (mean: 28.7), minimizing concerns about weak instrument bias. We used the pictures including the forest plot, scatter plot, and funnel plot to show the results of the MR analysis. We displayed the result of MDD on acute appendicitis (Figure 3); others are shown in our Supporting information (Supporting information 1). The results of the MR analysis for MDD on acute appendicitis. (a) Forest plot for the MR effect size of each IV, with each point representing the causal effect by IVW if only using this SNP. The red line at bottom represents the effect size of the IVW method. (b) The scatter plot of causal effects of MDD on acute appendicitis. The vertical and horizontal black lines show the 95% CI of the estimated effect of IVs on MDD and acute appendicitis. The x axis represents the SNP effect on MDD, and the y axis represents the SNP effect on acute appendicitis. (c) The funnel plot of the causal effect of MDD on acute appendicitis. Each point represents the individual causal effect of every IV. The red line represents the causal effect estimate of the IVW method. MR, Mendelian randomization; MDD, major depression disorder; IV, instrumental variable; IVW, inverse variance weighting; SNP, single‐nucleotide polymorphism; CI, confidence interval. (a) (b) (c) Finally, we performed reverse‐direction MR to explore whether there was potential reverse causality (Supporting information 2). Using the same criteria in MR, we obtained 47 IVs for acute appendicitis, 28 IVs for Crohn’s disease, 23 IVs for gastric ulcer, 50 IVs for GORD, 20 IVs for gastroduodenal ulcer, 22 IVs for IBS, 13 IVs for ulcer of esophagus, and 80 IVs for ulcerative colitis (Table S2). For the reverse result, we found GORD was a risky factor on MDD (OR: 1.09; 95% CI: 1.03–1.16; p = 0.005, FDR = 0.041), and other results did not show significance (Table S3). As for sensitivity analyses, GORD on MDD did not show obvious heterogeneity and pleiotropy, as well as outliers (p > 0.05), and all reverse sensitivity analyses are shown in Table S5. 3.3. SNP Heritability Enrichment of Different Tissues in GWAS Summary Statistics Based on the result of MR, we further defined the partitioned heritability of MDD and three GI diseases by S‐LDSC (Figure 4 and Table S6). For MDD, there were five significant tissue types of SNP heritability enrichment when using top 10% of marker genes and the most significant tissue was liver, while six significant results when using top 5% of marker genes and the most significant tissue was brain cortex. For acute appendicitis, there were four significant tissue types of SNP heritability enrichment when using top 10% of marker genes and the most significant tissue was esophagus gastroesophageal junction, while three significant results when using top 5% of marker genes and the most significant tissue was colon sigmoid. For IBS, there was only one significant tissue type of SNP heritability enrichment when using top 10% as well as top 5% of marker genes. For ulcer of esophagus, there were no significant tissue type of SNP heritability enrichment when using top 10% as well as top 5% of marker genes. However, there were no significant overlapping tissues between MDD and GI diseases. SNP heritability enrichment in different tissues. The x axis represents the negative log10 p values of coefficient Z scores, while the y axis represents the selected tissues from GTEx. The criterion in the legend represents the different defined focal functional categories, the dark color represents using top 10% of marker genes, while the light color represents 5% of selected marker genes. The gray dotted line represents the significant threshold 0.05. (a) Tissue type of SNP heritability enrichment of 25 tissues in MDD. (b) Tissue type of SNP heritability enrichment of 25 tissues in acute appendicitis. (c) Tissue type of SNP heritability enrichment of 25 tissues in irritable bowel syndrome. (d) Tissue type of SNP heritability enrichment of 25 tissues in ulcer of esophagus. MDD, major depression disorder; SNP, single‐nucleotide polymorphism. 3.4. Summary for Shared Genetics Between Significant Trait Pairs Based on the results of MR, we performed the TWAS analysis for MDD and three GI diseases. Based on the FDR, we defined 57 significant genes for MDD, 218 significant genes for acute appendicitis, one significant gene for IBS, and one significant gene for ulcer of esophagus (Figure 5 and Table S7). Furthermore, we found that MDD and acute appendicitis have three shared genes: PRSS16, ZNF602P, and ZNF204P (Table 2). Z score of TWAS of the same genes of MDD and acute appendicitis in different tissues. (a) The expression of genes PRSS16, ZNF204P, and ZNF602P of MDD in 25 tissues. (b) The expression of genes PRSS16, ZNF204P, and ZNF602P of acute appendicitis in 25 tissues. The blue color means that the Z score has a negative value, while the red color means that the Z score has a positive value. The higher absolute value of Z represents stronger expression of genes. The asterisk represents that the value of P after the FDR of the TWAS Z score is significant for this gene in tissues. MDD, major depression disorder; TWAS, transcriptome‐wide association study. (a) (b) Table 2The shared genes of TWAS for MDD and acute appendicitis.DiseaseGeneTissue p FDRMDD PRSS16 Brain cerebellar hemisphere3.82 × 10−6 0.006MDD PRSS16 Brain cerebellum3.70 × 10−5 0.028MDD PRSS16 Esophagus mucosa5.28 × 10−6 0.026MDD PRSS16 Whole blood3.06 × 10−6 0.024Acute appendicitis PRSS16 Brain cerebellum1.81 × 10−4 0.042Acute appendicitis PRSS16 Whole blood1.97 × 10−4 0.046MDD ZNF602P Brain cerebellum9.87 × 10−5 0.042MDD ZNF602P Brain substantia nigra4.71 × 10−5 0.031MDD ZNF602P Liver7.83 × 10−5 0.046MDD ZNF602P Stomach5.76 × 10−5 0.048Acute appendicitis ZNF602P Brain caudate basal ganglia1.18 × 10−4 0.038Acute appendicitis ZNF602P Brain frontal cortex1.18 × 10−4 0.028Acute appendicitis ZNF602P Esophagus mucosa6.66 × 10−6 0.006Acute appendicitis ZNF602P Small intestine terminal ileum1.52 × 10−6 0.001MDD ZNF204P Brain cerebellar hemisphere5.78 × 10−5 0.035MDD ZNF204P Brain cerebellum5.11 × 10−5 0.031MDD ZNF204P Brain spinal cord cervical c‐15.36 × 10−6 0.008MDD ZNF204P Esophagus muscularis2.02 × 10−6 0.009Acute appendicitis ZNF204P Colon transverse2.99 × 10−5 0.012 The shared genes of TWAS for MDD and acute appendicitis. After TWAS, we further performed the colocalization analysis to find whether there were shared pleiotropic genes between MDD and three GI diseases. Based on the FDR, we defined 26 significant pleiotropic genes for MDD, and two significant pleiotropic genes for acute appendicitis when using candidate genes of TWAS of MDD. We also defined 23 significant pleiotropic genes for MDD and 96 significant pleiotropic genes for acute appendicitis when using candidate genes of TWAS of acute appendicitis. Besides, we defined one pleiotropic gene for IBS when using candidate genes of TWAS of IBS and one pleiotropic gene for ulcer of esophagus when using candidate genes of TWAS of ulcer of esophagus (Table S8). And we found that five genes C4A, FLOT1, LINC00243, MICB, and PRSS16 shared the same tissues between MDD and acute appendicitis (Table 3). Table 3The shared pleiotropic genes and SNPs of colocalization for MDD and acute appendicitis.DiseaseTissuePP.H4Lead SNPGeneMDDBrain amygdala0.856rs2534684 MICB Acute appendicitisBrain amygdala0.881rs1980496 MICB MDDBrain anterior cingulate cortex0.776rs2894221 MICB Acute appendicitisBrain anterior cingulate cortex0.982rs3130924 MICB MDDBrain caudate basal ganglia0.940rs2534684 MICB Acute appendicitisBrain caudate basal ganglia0.933rs3130924 MICB MDDBrain hypothalamus0.931rs2534684 MICB Acute appendicitisBrain hypothalamus0.812rs2894221 MICB MDDBrain putamen basal ganglia0.936rs2534684 MICB Acute appendicitisBrain putamen basal ganglia0.861rs3130924 MICB MDDColon transverse0.844rs3130660 FLOT1 Acute appendicitisColon transverse0.783rs3095151 FLOT1 MDDEsophagus mucosa0.951rs3130660 LINC00243 Acute appendicitisEsophagus mucosa0.891rs3131934 LINC00243 MDDEsophagus mucosa0.962rs34388707 PRSS16 Acute appendicitisEsophagus mucosa0.817rs34569203 PRSS16 MDDPituitary0.927rs2534684 MICB Acute appendicitisPituitary0.935rs3130924 MICB MDDStomach0.905rs3094125 FLOT1 Acute appendicitisStomach0.873rs3131934 FLOT1 The shared pleiotropic genes and SNPs of colocalization for MDD and acute appendicitis. 4. Discussion In our study, we conducted an integrated analysis to explore the shared genetic basis between MDD and eight GI diseases. For the six GI diseases genetically correlating to MDD, we further performed two‐sample MR analysis to examine their causal relationship. Based on multiple testing correction and a series of sensitivity analyses, as well as reverse‐direction MR, we defined a positive causal relationship between MDD and acute appendicitis (OR = 1.09), IBS (OR = 1.14), and ulcer of esophagus (OR = 1.24). Subsequently, we conducted S‐LDSC and TWAS to explore the expression of genes of MDD and the three GI diseases across various tissues. TWAS revealed three overlapping genes between MDD and acute appendicitis. Finally, the colocalization analysis identified five shared pleiotropic genes for MDD and acute appendicitis: C4A, FLOT1, LINC00243, MICB, and PRSS16. Our results suggest a potential causal association between MDD and an increased risk of IBS and ulcer of esophagus, which is consistent with findings from previous studies [18, 42, 43]. Moreover, we uncovered a causal relationship between MDD and acute appendicitis—a novel finding that requires further validation. In this analysis, given the relatively weak signals observed in the GWAS summary statistics and the limited number of available IVs, we employed a more lenient p value threshold of 1 × 10−5, rather than the conventional level of 5 × 10−8. While this relaxed threshold may increase the likelihood of false‐positive findings, we performed three sensitivity analyses as well as multiple testing corrections to enhance the robustness of our results and to balance statistical power with control of Type I error. The direction of the effect we observed (i.e., MDD increasing appendicitis risk) appears to contradict previous research that suggested a potential protective effect of manic‐depressive psychosis against acute appendicitis [44]. This discrepancy could be attributed to several factors, including fundamental differences in phenotype definition (MDD vs. depression), the genetic nature of our analysis versus the observational design of prior work, or the utilization of distinct datasets and populations. Therefore, our results should not be interpreted as definitive proof of causation but rather as robust preliminary genetic evidence highlighting a previously unexplored relationship. Consequently, future investigations employing larger, more powerful datasets, such as the UK Biobank, and more stringent genetic instruments are essential to confirm or refute this putative causal link. In our TWAS analysis, we used eQTLs from the GTEx database as reference panels. Unlike other TWAS studies that focus on identifying candidate genes, we used TWAS to investigate gene expression associated with specific traits in different tissues. We identified three shared genes—PRSS16, ZNF602P, and ZNF204P—between MDD and acute appendicitis. The gene ZNF204P, also known as zinc finger protein 204, functions as a pseudogene. A study has revealed its involvement as a long noncoding RNA in hepatocellular carcinoma development [45]. Similarly, ZNF602P, or zinc finger protein 602, also functions as a pseudogene. PRSS16, or serine protease 16, plays an important role in human life processes and has been identified as a risk gene for schizophrenia [46]. However, the precise association between PRSS16 and MDD, as well as acute appendicitis, remains to be confirmed. Additionally, through the colocalization analysis, we identified five shared pleiotropic genes between MDD and acute appendicitis: C4A, FLOT1, LINC00243, MICB, and PRSS16. C4A, also called Complement C4A, plays a key role in the human immune system and has established links with depression [47] and IBD [48]. FLOT1, or Flotillin 1, has been confirmed as a significant contributor to MDD and GI malignancy [49], although its specific role in other GI diseases remains to be studied. LINC00243, also known as long intergenic nonprotein coding RNA 243, is involved in the development of certain cancers [50]. MICB, or MHC Class I Polypeptide‐Related Sequence B, is crucial for GI tract function [51] and has recently been identified as a risky gene for depression among perinatal populations [52]. In total, the genes identified in our study provide new insights into the shared genetic mechanisms underlying MDD and acute appendicitis. Indeed, previous studies have demonstrated that MDD can impact gut microbiota, which in turn affects GI health [53, 54]. Additionally, the microbiota–gut–brain axis has been shown to mediate the relationship between MDD and GI diseases [55]. However, most of these studies have primarily relied on indirect evidence. Unlike these studies, our research provides direct evidence of this connection. We have not only confirmed the causal relationship between MDD and three specific GI diseases, but have also identified shared genes between MDD and acute appendicitis. Our findings provide the basis for public health policies, underscoring the need for medical professionals to closely monitor the GI health of individuals diagnosed with MDD, with particular attention to potential emergencies. Compared with previous studies, our study not only expands the types of GI diseases studied but also uses a new combination of methods. The major strength of our study lies in its multifaceted analytical approach, providing a robust framework for understanding the complex genetic overlaps that link MDD and GI diseases. Specifically, LDSC determines the genetic correlation between traits, while MR analyses identify the causal relationship between MDD and GI diseases. Furthermore, S‐LDSC explores the possible gene expression of the trait. TWAS facilitates the exploration of gene expression in various tissues, and the colocalization analysis helps identify shared causal genetic variants. The integration of these methods enables us not only to identify causality but also to search for pleiotropic genes, thus offering valuable insights into potential therapeutic targets and directions for future research. Our findings, within the framework of MR and its core assumptions, provide evidence supporting a potential causal role of MDD on the risk of acute appendicitis, IBS, and esophageal ulcer. It is important to emphasize that the causal estimates from MR are valid only under the satisfaction of the three key instrumental variable assumptions that the genetic variants are strongly associated with the exposure (MDD), are not associated with any confounders of the exposure‐outcome relationship, and influence the outcome solely through the exposure (no horizontal pleiotropy). While we employed multiple sensitivity analyses to mitigate violations of these assumptions, particularly pleiotropy, the interpretability of our results remains contingent upon these assumptions holding true. There are certain limitations to our study. First, we did not aggregate all datasets covering relevant traits, such as the UK Biobank, which may affect the robustness of our findings due to smaller sample sizes. Furthermore, while we employed TWAS to identify gene expression, we did not subsequently validate these genes. Additionally, our colocalization analysis was restricted to genes identified through TWAS, potentially overlooking other significant genetic variants. In addition, we chose FDR correction instead of Bonferroni correction for multiple testing, which could increase the risk of false positives. The use of a lenient p value threshold (1e − 5) for IV selection, while necessary for power, might have introduced weak instrument bias or increased false positives, and the estimates should be interpreted with this caveat in mind. Although we mitigated this through sensitivity analyses, the possibility of residual bias cannot be fully excluded. Lastly, our study only analyzed the individuals of European descent. Therefore, the generalizability of our results to populations of different ancestries is limited. Furthermore, our findings represent the lifelong risk associated with genetic predisposition to MDD, as genetic instruments are fixed at birth. This may not directly generalize to the risk posed by MDD episodes occurring at specific life stages. Similarly, it remains to be tested whether the observed genetic correlations and causal relationships are consistent across different age groups (e.g., younger vs. older adults) or other population strata defined by sex or socioeconomic status. Future studies in diverse cohorts are necessary to assess the universality of the gut–brain axis mechanisms identified here. 5. Conclusion In conclusion, our study provides compelling evidence for the genetic association between MDD and GI diseases, particularly with acute appendicitis. We have identified shared pleiotropic genes between MDD and acute appendicitis, offering new insights into their shared genetic basis. Additionally, our findings could guide further research into the etiology and prognosis of acute appendicitis, potentially paving the way for novel therapeutic strategies. Ethics Statement The authors have nothing to report. Consent The authors have nothing to report. Disclosure All co‐authors have seen and agreed with the manuscript’s contents. Conflicts of Interest The authors declare no conflicts of interest. Author Contributions Conceptualization and supervision: Fu Sun, Peng Huang, and Sheng Yang. Data curation: Fu Sun, Haochang Li, Shirong Hui, Ran He, Yihao Li, and Meijie Yu. Formal analysis and visualization: Haochang Li, Bin Gong, Shirong Hui, and Ran He. Validation and writing–original draft: Fu Sun, Bin Gong, and Haochang Li. Writing–review and editing: Fu Sun, Haochang Li, Peng Huang, and Sheng Yang. Fu Sun and Haochang Li contributed equally to this study. Funding No funding was received for this research. Supporting Information Table S1: The results of LDSC. Table S2: The summary information about IVs. Table S3: The results of MR. Table S4: The results of the sensitivity analysis for direct MR. Table S5: The results of the sensitivity analysis for reverse MR. Table S6: The results of S‐LDSC. Table S7: The results of TWAS. Table S8: The results of colocalization. Supporting Information 1: The results of direct MR. Supporting Information 2: The results of reverse MR. Supporting information Supporting Information Additional supporting information can be found online in the Supporting Information section. Acknowledgments The authors have nothing to report. Sun, Fu , Li, Haochang , Gong, Bin , Hui, Shirong , He, Ran , Yu, Meijie , Li, Yihao , Yang, Sheng , Huang, Peng , Investigating the Causal Relationship and Shared Genetic Basis Between Major Depression Disorder and Eight Types of Gastrointestinal Diseases, Genetics Research, 2025, 6931243, 13 pages, 2025. 10.1155/genr/6931243 Academic Editor: Sreeja Dattachoudhury Contributor Information Peng Huang, Email: huangpeng@njmu.edu.cn. Sreeja Dattachoudhury, Email: sdattachou@wiley.com. Data Availability Statement Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study. References Associated Data Supplementary Materials Supporting Information Additional supporting information can be found online in the Supporting Information section. Data Availability Statement Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.