Gut microbiome dysbiosis is associated with aldosterone overproduction in idiopathic hyperaldosteronism Authors: Yuanliang Jiang, Qinglian Zeng, Linqiang Ma, Yifan He, Shuangshuang Zhu, Yunjie Xiong, Ying Jing, Qi Zhang, Shuangxin Qi, Jinbo Hu, Qifu Li, Shumin Yang Published: 2026 ABSTRACT SummaryIdiopathic hyperaldosteronism (IHA) is a significant cause of secondary hypertension. Emerging evidence links the gut microbiota to endocrine and cardiovascular diseases. We integrated human genetics, multi-omics, and fecal microbiota transplantation (FMT) to investigate the association between IHA and gut microbiome. Two-sample Mendelian randomization identified seven bacterial taxa associated with IHA. Metagenomic analysis of 30 patients with IHA and 30 healthy control subjects revealed lower microbial diversity and depletion of protective genera identified by Mendelian randomization. Plasma metabolomics showed alterations in tryptophan metabolism and other metabolic pathways that overlapped with microbial functional changes. Tryptophan-derived metabolites correlated with the abundance of protective genera. FMT from patients with IHA increased serum aldosterone and elevated 5-hydroxy-L-tryptophan in germ-free mice. These findings support an association between gut microbiome dysbiosis, metabolic alterations, and aldosterone dysregulation, highlighting the gut microbiome as a potential target for the diagnosis and treatment of IHA.Keywords: idiopathic hyperaldosteronism, gut microbiome, Mendelian randomization, metabolome, fecal microbiota transplantation Introduction Primary aldosteronism (PA) is characterized by autonomous aldosterone secretion from the adrenal cortex, clinically manifesting as hypertension and/or hypokalemia.1,2 It represents one of the most common forms of secondary hypertension, affecting up to 10% of hypertensive patients.3,4 PA is classified into unilateral and bilateral subtypes. Bilateral PA, accounting for approximately 65% of cases, is primarily caused by idiopathic hyperaldosteronism (IHA).3,5 Despite advances in diagnosis and treatment, the pathogenic mechanisms driving excessive aldosterone synthesis in IHA remain largely unknown. Accumulating research indicates that the gut microbiota plays a pivotal role in maintaining host homeostasis by modulating immune responses and metabolic pathways.6 Its dysbiosis has been implicated in the pathophysiology of various cardiovascular and metabolic disorders, including hypertension,7 coronary artery disease,8 and obesity.9 Emerging evidence suggests a potential link between gut microbial alterations and PA. An observational study has reported distinct gut microbial profiles and reduced diversity in PA patients compared with control subjects and those with essential hypertension (EH).10 Furthermore, experimental manipulation of the gut microbiota in mice has been shown to change aldosterone levels.11 However, existing research primarily focuses on PA in general and offers correlative insights, with limited exploration of IHA. Crucially, whether gut dysbiosis is a causal driver or a secondary consequence of PA or IHA remains unclear. To address this, we conducted a comprehensive multi-omics study integrating human genetics, deep phenotyping, and direct experimental validation. We first employed a two-sample Mendelian randomization (MR) analysis using large-scale genome-wide association study (GWAS) data to infer causal relationships between specific gut microbiota and IHA.12 Subsequently, we performed metagenomic sequencing and untargeted metabolomics on a well-characterized cohort of IHA patients and matched healthy control subjects (HCs) to delineate associated alterations in microbial composition, function, and host metabolism. Finally, to establish direct causal evidence, we transplanted fecal microbiota from IHA patients into germ-free (GF) mice and assessed aldosterone production and related phenotypic outcomes. In summary, our study aims to investigate the associations between gut microbial dysbiosis, host metabolism, and IHA, potentially paving the way for microbiota-targeted diagnostic and therapeutic strategies. Results MR analysis revealed a causal relationship between gut microbiota and IHA Following the application of our inclusion criteria for instrumental variables (IVs), 2,137 single-nucleotide polymorphisms (SNPs) were retained for MR analysis (Table S1). These IVs corresponded to microbial taxa across five taxonomic levels: 9 phyla, 16 classes, 20 orders, 35 families, and 131 genera. The inverse variance weighted (IVW) method identified seven taxa significantly associated with IHA (Figure 1). Of these, five taxa demonstrated protective associations with IHA: o_Desulfovibrionales, g_Ruminococcus1, g_RuminococcaceaeUCG003, f_Peptostreptococcaceae, g_Eubacteriumnodatumgroup. Conversely, f_Prevotellaceae and g_RuminococcaceaeUCG014 were positively associated with IHA risk. Causal relationship between seven bacterial taxa and IHA through MR analysis Forest plot of the MR analysis for the relationship between gut microbiota taxa and IHA. p values were derived from the inverse-variance weighted method, with p < 0.05 considered statistically significant. CI, confidence interval; OR, odds ratio. Leave-one-out analysis demonstrated the robustness of the causal estimates, as the effects remained largely consistent upon iteratively excluding individual SNPs (Figure S1A). Causal estimates derived from the MR-Egger method were comparable with those obtained using the IVW method (Figure S1B). The Cochran Q test for the IVW method and Rücker Q test for the MR-Egger method indicated no significant heterogeneity among the IVs (Table S2). Furthermore, MR-Egger intercept analysis revealed no evidence of directional horizontal pleiotropy (Table S3). Additional sensitivity analyses were performed using weighted median, weighted mode, and MR-Egger methods. Effect estimates were generally consistent in direction with the primary IVW results (Table S4). Furthermore, MR-PRESSO global tests detected no evidence of significant horizontal pleiotropy for any of the seven taxa associated with IHA (all p > 0.05; Table S5). Depletion of commensal taxa in IHA gut microbiome To investigate gut microbiome alterations in IHA, we recruited 30 IHA patients and 30 HC, performing metagenomic sequencing on their fecal samples. The main clinical characteristics of the study population are summarized in Table 1, while detailed clinical information is presented in Table S6. Metagenomic sequencing generated a total of 857.03 Gb of data, with an average of 9.46 × 107 high-quality reads per sample (Table S7). Rarefaction analysis indicated sufficient sequencing depth, with the curve approaching saturation (Figure 2A). The IHA group showed a significantly lower species number (Figure 2B). Alpha diversity, measured by the Shannon index, was relatively lower in the IHA group (Figure 2C). Principal coordinate analysis (PCoA) based on Bray-Curtis distance revealed a clear separation between the IHA and HC groups at the species level (PERMANOVA, p = 0.005; Figure 2D). Table 1The main clinical characteristics of study populationCharacteristicHC (n = 30)IHA (n = 30)pFemale, n (%)a17 (56.67)19 (63.33)0.598Age (y)b49.47 ± 8.5149.53 ± 9.360.977BMI (kg/m2)b25.03 ± 2.8725.89 ± 3.230.282SBP (mmHg)b109.93 ± 8.94137.55 ± 11.75<0.001DBP (mmHg)b72.45 ± 6.7585.80 ± 8.44<0.001PAC (pg/mL)b122.39 ± 61.75234.97 ± 108.89<0.001PRC (uIU/mL)b19.53 ± 18.905.47 ± 5.88<0.001ARR (pg/uIU)b25.99 ± 46.26123.25 ± 156.960.003Serum potassium (mmol/L)b4.15 ± 0.253.95 ± 0.350.010Data are presented as mean ± SD or number (%). HC, healthy control subjects; IHA, idiopathic hyperaldosteronism; BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; PAC, plasma aldosterone concentration; PRC, plasma renin concentration; ARR, aldosterone-renin ratio.aChi-square test.bStudent’s t test. The main clinical characteristics of study population Data are presented as mean ± SD or number (%). HC, healthy control subjects; IHA, idiopathic hyperaldosteronism; BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; PAC, plasma aldosterone concentration; PRC, plasma renin concentration; ARR, aldosterone-renin ratio. Chi-square test. Student’s t test. Alterations in gut microbial composition in IHA (A) Rarefaction curves showing the number of genes in IHA (n = 30) and HC (n = 30) samples. The curve approaching saturation indicates that the sequencing depth is sufficient. (B and C) Violin plots showing species richness and α-diversity (Shannon index) in IHA and HC groups. Statistical significance was determined using the Wilcoxon rank-sum test; ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001. (D) Principal coordinate analysis (PCoA) plot based on Bray-Curtis distance showing separation between the IHA and HC groups at the species level. Statistical significance was determined using the PERMANOVA, p = 0.005. (E–H) Relative abundance distributions of top 20 taxa at the phylum, order, family, and genus levels in two groups. Differentially enriched or depleted taxa in IHA compared with HC were identified using the Wilcoxon rank-sum test (p < 0.05). Red and blue labels indicate increased and decreased abundance in IHA, respectively. At the phylum level, p_Bacillota (formerly p_Firmicutes, mean relative abundance: 63.95%) was the most abundant taxon across all samples, followed by p_Bacteroidota (formerly p_Bacteroidetes, 24.92%) (Figure 2E). Further analysis at the order, family, and genus levels identified o_Eubacteriales (32.52%), f_Lachnospiraceae (21.66%), and g_Faecalibacterium (12.88%) as the predominant taxa, respectively (Figures 2F–2H). Among the top 20 most abundant taxa, differential enrichment patterns were observed between two groups. The HC group exhibited significantly higher relative abundances of p_Bacillota and several of its subordinate taxa, including g_Ruminococcus, g_Roseburia, g_Eubacterium, g_Coprococcus, g_Dorea, and g_Lachnospira (Figures 2E–2H). In contrast, the IHA group showed enrichment of p_Pseudomonadota (formerly p_Proteobacteria) across multiple taxonomic levels (Figures 2E–2H). We further examined the seven microbial taxa previously identified as causal factors for IHA in the MR analysis. Five of the seven taxa were detected in our cohort: o_Desulfovibrionales, f_Peptostreptococcaceae, f_Prevotellaceae, g_Ruminococcus, and g_Eubacterium (Table S8). Notably, both g_Ruminococcus and g_Eubacterium, which serve as protective factors against IHA, showed significantly reduced abundances in the IHA group, supporting the MR findings. Functional remodeling of gut microbiota in IHA To further explore microbial compositional differences between IHA and HC, we performed linear discriminant analysis (LDA) effect size (LEfSe) analysis. At the genus level, LEfSe analysis identified 7 genera enriched in IHA and 19 genera enriched in HC (LDA >2.0, p < 0.05; Table S9). To assess potential clinical confounding, we further performed multivariable regression analyses adjusting for sex, age, body mass index (BMI), systolic blood pressure (SBP), serum potassium, creatinine, smoking status, and alcohol consumption. The associations of the key genera g_Ruminococcus and g_Eubacterium with IHA remained statistically significant after adjustment. At the species level, 18 species enriched in the IHA group and 99 species enriched in the HC group (LDA >2.0, p < 0.05; Figure 3A and Table S10). Alterations in microbial species composition and functional profiles in IHA (A) Cladogram showing differentially abundant microbial species between IHA and HC groups identified by LEfSe analysis. A total of 117 species showed differential abundance, including 18 species enriched in IHA and 99 species enriched in HC (LEfSe LDA score >2.0, Wilcoxon rank-sum test, p < 0.05). (B) Gut microbial co-occurrence network constructed using SparCC to visualize potential interactions among species. The network comprises 116 nodes and 1,267 edges. Node size corresponds to relative abundance, with red and blue colors denoting taxa upregulated or downregulated in IHA, respectively. Edge colors indicate positive (red) or negative (blue) correlations. (C) Principal coordinate analysis (PCoA) plot based on Bray-Curtis distance showing separation of microbial functional profiles between the IHA and HC groups. Statistical significance was determined using the PERMANOVA, p = 0.004. (D) LEfSe analysis identifying differentially enriched microbial pathways between IHA and HC groups. Twenty level 3 pathways were enriched in IHA, whereas 18 pathways were enriched in HC (LEfSe LDA score >2.0; Wilcoxon rank-sum test, p < 0.05). Font colors indicate the corresponding enriched groups at level 2 pathways (red, IHA; blue, HC). Here, we constructed a gut microbial co-occurrence network using SparCC to visualize potential interactions among differential abundant species (Figure 3B). The network comprised 116 nodes and 1,267 edges. IHA-enriched species were found to be primarily concentrated in genera such as g_Clostridium, g_Eubacterium, g_Roseburia, g_Blautia, g_Ruminococcus, g_Enterocloster, and so on. Notably, s_Ruminococcus sp. CAG:403 under g_Ruminococcus was enriched in IHA. In contrast, the HC group showed significant enrichment of 15 species, including s_Clostridioides difficile (within f_Peptostreptococcaceae), 9 species under g_Eubacterium, and 5 species under g_Ruminococcus (Table S10). We assessed the gut microbial functions across the study cohort using the KEGG database, identifying a total of 7,027 Kyoto Encyclopedia of Genes and Genomes (KEGG) Orthology genes. PCoA based on Bray-Curtis distance indicated significant functional differences between the IHA and HC groups (PERMANOVA, p = 0.004; Figure 3C). LEfSe analysis of microbial functions showed that 20 pathways were enriched in the IHA group, while 18 pathways were enriched in the HC group (Figure 3D and Table S11). When summarized at the level 2 pathway classification, IHA microbiota were primarily associated with carbohydrate metabolism, metabolism of other amino acids, and nervous system (Figure 3D and Table S12). Specifically, these enriched pathways encompassed multiple biological processes. Valine, leucine, and isoleucine degradation (ko00280) converted branched-chain amino acids into acetyl-CoA or succinyl-CoA, which enter the tricarboxylic acid cycle as a key energy source.13 The enrichment of lipopolysaccharide biosynthesis (ko00540) directly suggested increased endotoxin production, potentially driving systemic inflammation.14 Moreover, tryptophan metabolism (ko00380) generated various signaling molecules such as kynurenine, serotonin, and indoles, deeply involved in host immunity and neuroendocrine functions.15,16 In summary, dysregulated gut microbial functions in IHA may contribute to disease pathogenesis through multiple interconnected mechanisms. Host metabolic signature linked to microbial dysfunction in IHA In the metabolomics analysis, a total of 1,103 metabolites were annotated, with 628 identified in positive ion mode and 475 in negative ion mode. The score plots of the orthogonal partial least-squares discriminant analysis model revealed significant separation between the IHA and HC groups (Figure 4A), with permutation testing confirming model validity (Figure 4B). Of these, 74 metabolites were identified as significantly different between the two groups (variable importance in projection [VIP] >1.5, p < 0.05; Table S13), with 22 downregulated and 52 upregulated in IHA compared with HC (Figure 4C). A clustering heatmap of these metabolites further highlighted distinct patterns between IHA patients and HCs (Figure 4D). Plasma metabolic signatures associated with IHA (A) Orthogonal partial least squares discriminant analysis (OPLS-DA) score plots display clear discrimination of IHA from HC. (B) Permutation tests evaluating the robustness of the OPLS-DA model. R2 represents the goodness of fit, whereas Q2 represents the predictive ability of the model. A negative Q2 intercept indicates the absence of model overfitting. (C) Volcano plot displays 22 decreased and 52 increased metabolites in IHA compared with HC (OPLS-DA VIP >1.5, Student’s t test, p < 0.05). (D) Clustered heatmap showing the expression patterns of the 74 differential metabolites in IHA patients and HC individuals. (E) Metabolic pathways analysis reveals five significantly altered KEGG pathways on MetaboAnalyst. Statistical significance was determined using the hypergeometric test, p < 0.05. Circle size represents pathway impact, while circle color indicates pathway enrichment significance based on -log10(P). Larger circles denote greater pathway impact, and redder colors indicate higher statistical significance. Red font shows those reflected in both the gut microbiome and plasma metabolome. KEGG pathway analysis of the differential metabolites revealed five significantly altered pathways (Figure 4E). Notably, consistent dysregulation was observed in galactose metabolism, tryptophan metabolism, and phenylalanine, tyrosine, and tryptophan biosynthesis, which were also reflected in both the gut microbiome and plasma metabolome. Key metabolites enriched in these three pathways included 2-oxoadipic acid, 5-hydroxy-l-tryptophan (5-HTP), alpha-D-glucose, D-sorbitol, and DL-phenylalanine. Of particular interest is 5-HTP, which exhibited the greatest fold change among these compounds. Similarly, the association between 5-HTP and IHA remained significant after multivariable adjustment for sex, age, BMI, SBP, serum potassium, creatinine, smoking status, and alcohol consumption (Table S13). We performed functional intervention experiments in vitro using the human adrenocortical carcinoma cell line HAC15. Across various concentrations, 5-HTP did not induce a statistically significant alteration in aldosterone secretion by HAC15 cells (p > 0.05; Figure S2). These findings suggest that 5-HTP may not directly act on adrenocortical cells to modulate aldosterone synthesis. This negative result in vitro is consistent with previous reports.17 However, studies in both humans and rats have demonstrated that 5-HTP can elevate aldosterone levels in vivo,17,18 underscoring a potential association between altered 5-HTP metabolism and aldosterone regulation. Perturbed gut microbiota interactions with host metabolism related to IHA We investigated the associations between IHA-enriched gut microbiota and circulating metabolites. Spearman correlation analysis revealed that the abundances of g_Eubacterium and g_Ruminococcus were negatively correlated with plasma aldosterone concentration (PAC), aldosterone-renin ratio (ARR), SBP, and diastolic blood pressure (DBP), but positively correlated with plasma renin concentration (Figure 5). Among the three overlapping pathways identified from both metagenomics and metabolomics, only the tryptophan metabolism showed significant negative correlations with these two protective genera. Of the seven key metabolites in the five altered pathways, the two derived from tryptophan metabolism were significantly associated with two genera: 2-oxoadipic acid positively correlated with their abundance, whereas 5-HTP showed negative correlations. The Spearman correlation analysis revealed that a total of 38 species were significantly correlated with PAC, including eight species that belonged to the seven bacterial taxa identified by MR analysis (Table S14). Associations between gut microbiota and host metabolic alterations related to IHA Correlation heatmap showing the associations between differentially abundant genera and clinical characteristics, key metabolites, and metabolic pathways. ko00400, phenylalanine, tyrosine, and tryptophan biosynthesis; ko00052, galactose metabolism; ko00380, tryptophan metabolism. Statistical significance was determined using Spearman correlation. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001. ARR, aldosterone-renin ratio; PAC, plasma aldosterone concentration; SBP, systolic blood pressure; DBP, diastolic blood pressure; PRC, plasma renin concentration; BMI, body mass index. FMT from IHA patients elevated aldosterone levels in GF mice Based on previous findings, we hypothesized a potential causal relationship between gut microbiota and IHA. To further validate our hypothesis, we transplanted pooled fecal microbiota from IHA patients and HCs into GF mice. Microbiota donors included two IHA patients and two HCs (Table S15). A total of 20 male mice were divided into two groups: IHA microbiota colonization (FMT_IHA) and HC microbiota colonization (FMT_HC), receiving gavage three times per week for four weeks (Figure 6A). Phenotyping began at the ninth week. Throughout the experiment, there were no significant differences between the two groups in body weight (Figure 6B). Fecal microbiota transplantation induces IHA-associated phenotypic changes in GF mice (A) Experimental procedure of FMT. Twenty male GF C57BL/6JGpt mice were randomly assigned to FMT_IHA (n = 10) and FMT_HC (n = 10) groups. Mice received FMT three times per week for four weeks, followed by a four-week period without FMT. Phenotypic assessments were performed from week 9 onward. (B) Body weight dynamics of FMT_IHA and FMT_HC mice over the experimental period. (C and D) Violin plots showing the number of ASVs and α-diversity (Shannon index) of gut microbiota. Statistical significance was determined using the Wilcoxon rank-sum test. ns, not significant. (E) Principal coordinate analysis (PCoA) based on Bray-Curtis distance showing separation of gut microbial community structures between FMT_IHA and FMT_HC groups. Statistical significance was determined using the PERMANOVA, p = 0.001. (F and G) Serum aldosterone (F) and corticosterone (G) levels were significantly elevated in FMT_IHA mice. (H–J) Serum ACTH (H), renin concentration (I), and renal Renin1 mRNA expression (J) showed no significant differences between the two groups. (K) Targeted metabolomic analysis showing serum 5-HTP levels significantly elevate in FMT_IHA mice. (L) Scatterplot showing the positive correlation between serum 5-HTP and aldosterone levels. (M–O) SBP (M), DBP (N), and heart rate (O) showed no significant differences between the two groups. (P and Q) Echocardiography-derived stroke volume (P) and cardiac output (Q) were reduced in FMT_IHA mice. Data are presented as mean ± SD. Statistical significance for (F)–(K) and (M)–(Q) was determined using the Student’s t test. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001; ns, not significant. FMT, fecal microbiota transplantation; ACTH, adrenocorticotropic hormone; SBP, systolic blood pressure; DBP, diastolic blood pressure; CO, cardiac output. We performed 16S rRNA gene sequencing on the mouse gut microbiota. Although the FMT_IHA group showed decreases in amplicon sequence variants (ASV) number (Figure 6C) and Shannon index (Figure 6D), these differences were not statistically significant. The PCoA scores plot based on Bray-Curtis distances revealed a significant distinction in gut microbial structure between FMT_IHA and FMT_HC mice (PERMANOVA, p = 0.001; Figure 6E), with the former clustering more closely with their respective human donors at the genus level. PICRUSt2 functional prediction identified 234 KEGG pathways associated with IHA, with 223 significantly upregulated and 11 downregulated in FMT_IHA mice (p < 0.05; Table S16). Notably, galactose metabolism and tryptophan metabolism exhibited significant alterations in the FMT_IHA group, echoing the changes observed in both microbial function and circulating metabolome in IHA patients. Compared with the FMT_HC group, the FMT_IHA mice showed significantly higher serum aldosterone levels (483.97 ± 126.89 ng/mL vs. 278.17 ± 97.79 ng/mL, Figure 6F). Serum corticosterone levels were likewise significantly increased in the FMT_IHA group (Figure 6G), indicating altered adrenal endocrine function following microbiota transplantation. In contrast, no significant difference in plasma adrenocorticotropic hormone levels was observed between the two groups (Figure 6H). No significant differences were detected in serum renin concentration (Figure 6I) or renal Renin1 mRNA expression, which encodes mouse renin (Figure 6J), between the two groups, suggesting that the elevation in aldosterone levels is likely independent of renin expression. Targeted metabolomic profiling of the tryptophan pathway demonstrated significantly elevated serum 5-HTP levels in the FMT_IHA group compared with the FMT_HC group (724.66 ± 287.91 ng/mL vs. 342.32 ± 130.28 ng/mL, Figure 6K and Table S17). Moreover, 5-HTP levels showed a significant positive correlation with aldosterone (Figure 6L), corroborating the findings from the human cohort study. Blood pressure measurements showed that SBP, DBP, and heart rate were higher in the FMT_IHA group, but the differences were not statistically significant (Figures 6M–6O). Echocardiography revealed decreased stroke volume and cardiac output in the FMT_IHA group (Figures 6P and 6Q), while there were no differences in other indicators (Figure S3A). Histological examination of adrenal glands by H&E staining revealed no obvious morphological abnormalities or gross structural differences between FMT_HC and FMT_IHA mice (Figure S3B). Discussion Based on previous research, we hypothesize that gut microbial dysbiosis may underlie the pathogenesis of IHA. By integrating two-sample MR, metagenomic sequencing, and plasma metabolomics, we identified a plausible association between gut microbial dysbiosis, host metabolic alterations, and aldosterone excess in IHA. This framework is critically complemented by the FMT experiment that transcends correlative observations to support a pathophysiological contribution of gut dysbiosis to IHA. This comprehensive study provides, for the first time, converging and multi-faceted evidence supporting a contributory role for the gut microbiota in the development of IHA. Using large-scale GWAS data, our MR analysis identified seven gut microbial taxa causally associated with IHA. Taxa such as g_Ruminococcus1 and g_Eubacteriumnodatumgroup were identified as protective factors, consistent with their established roles as major producers of short-chain fatty acids (SCFAs).19,20 A reduction in these taxa would likely diminish the anti-inflammatory effects of SCFAs, which are mediated through enhancement of intestinal epithelial barrier integrity.21 Conversely, f_Prevotellaceae was identified as a risk factor, as its colonization has been shown to exacerbate intestinal inflammation and thereby promote gut dysbiosis.22,23 It has also been reported that f_Prevotellaceae may be involved in estrogen signaling via GABAergic neurons.24 A prior MR study reported causal associations between hyperaldosteronism and microbial taxa including g_Intestinimonas, g_Prevotella9, g_Veillonella, and g_Parasutterella.25 Another study linked hyperaldosteronism to c_Methanobacteria, f_Lactobacillaceae, f_Methanobacteriaceae, g_Lactobacillus, o_Methanobacteriales, g_Parasutterella, g_Peptococcus, and g_Veillonella.26 These findings differ markedly from our results, which may be attributed to variations in the included populations, SNP selection criteria, and other factors. Their study utilized the FinnGen database, in which the enrolled PA cases likely included a mix of unilateral PA rather than IHA. In contrast, our cohort consisted exclusively of IHA cases without lateralization, as confirmed by adrenal venous sampling. Collectively, our findings provide a genetically informed prioritized list of microbial taxa, suggesting that alterations in the gut microbiome are not merely correlated with IHA but may contribute to its pathogenesis. Although none of the associations remained significant after multiple-testing correction, the identified taxa were not selected solely on the basis of MR evidence. Their relevance was further supported by independent metagenomic and FMT analyses, providing convergent evidence across multiple study designs. Metagenomic sequencing revealed microbial compositional alterations in IHA patients, accompanied by lower α-diversity. This aligns with a previous study in PA patients,10 which also observed decreased abundances of genera such as g_Coprococcus, g_Dorea, and g_Adlercreutzia. In line with the findings from MR analysis, the two genera identified as protective factors (g_Ruminococcus and g_Eubacterium) were significantly reduced in IHA patients. They also exhibit a negative correlation with PAC, ARR, SBP, and DBP, thereby providing cross-validation between genetic causal inference and observed microbiome alterations. Another important consideration is whether the gut microbial alterations observed in IHA are disease-specific or reflect broader changes associated with hypertension. Previous studies comparing patients with PA and those with EH have reported distinct microbial compositions,10 suggesting that aldosterone excess may contribute to microbial alterations beyond the effects of elevated blood pressure alone. Furthermore, recent evidence has identified gut microbial species associated with PACs and demonstrated the ability of certain bacteria to metabolize aldosterone directly.27 These findings support the biological plausibility of aldosterone-related microbiota signatures. Beyond compositional shifts, IHA was associated with functional remodeling of the gut microbiome. KEGG functional analysis indicated enrichment of pathways related to carbohydrate metabolism, amino acid metabolism, inflammation, and neuroendocrine signaling. Specifically, enrichment of lipopolysaccharide biosynthesis and biosynthesis of siderophore group nonribosomal peptides suggests enhanced pro-inflammatory potential,14,28,29 while alteration in other glycan degradation point to disrupted energy homeostasis.13 Galactose metabolism, involved in energy metabolism, is associated with various metabolic and cardiovascular diseases.30,31 The enrichment of tryptophan metabolism and GABAergic synapse implicates the involvement of the gut microbiota in neuroendocrine regulation,15,16,32 a mechanism highly relevant to aldosterone synthesis.33,34 To further characterize tryptophan metabolism, targeted metabolomic profiling was performed in FMT mice. In addition to elevated 5-HTP levels, significant alterations were observed in 5-hydroxyindoleacetic acid and indole-3-acetic acid, whereas serotonin, kynurenine, indole-3-acetamide, indolelactic acid, 3-indoleacrylic acid, 3-indolepropionic acid, and indoxyl sulfate were unchanged between groups. These findings suggest selective perturbations in specific branches of tryptophan metabolism rather than global remodeling of the entire pathway. Among the identified metabolic alterations, particular attention was drawn to 5-HTP and tryptophan-related metabolism. The systemic consequences of gut microbial dysbiosis were evident in the host plasma metabolome. Differential metabolite reflected perturbations in galactose metabolism, tryptophan metabolism, and phenylalanine, tyrosine, and tryptophan biosynthesis—pathways that overlapped with the alterations in microbial function. The putative protective taxa g_Ruminococcus and g_Eubacterium showed correlations with tryptophan-derived metabolites, particularly 2-oxoadipic acid and 5-HTP, suggesting a potential mechanistic link between microbial metabolism and adrenal hormone regulation. Notably, 5-HTP stands out as a special metabolite, exhibiting substantial fold changes across study groups. Previous studies have reported that 5-HTP can elevate aldosterone levels in healthy individuals, IHA patients, and rats.17,18,35 Previous studies have proposed that the endocrine effects of 5-HTP may involve the hypothalamic-pituitary-adrenal (HPA) axis and/or the renin-angiotensin-aldosterone system (RAAS).36,37 However, the present study did not directly evaluate these pathways, and their relevance to IHA remains uncertain. Moreover, one study has shown that injection of 5-HTP increased the expression of mineralocorticoid receptors in the rat amygdala compared with controls.38 These findings suggest that dysregulation of microbial tryptophan metabolism may contribute to alterations in circulating metabolites such as 5-HTP. However, the current data do not establish whether 5-HTP directly regulates adrenal aldosterone synthesis or acts indirectly through neural, immune, HPA axis, RAAS-related, or other systemic pathways. Therefore, 5-HTP should be regarded as a candidate metabolite associated with the microbiota-host interaction in IHA. The FMT experiment provided functional evidence supporting a contribution of the gut microbiota to IHA-related phenotypes. GF mice colonized with the gut microbiota from IHA patients exhibited significantly elevated serum aldosterone level compared with those receiving HC microbiota, indicating that the IHA-associated microbial ecosystem possesses the capacity to stimulate host steroidogenesis. Previous study has shown that plasma and urinary aldosterone levels increased in both male and female mice following antibiotic treatment.11 Echocardiography showed reduced stroke volume and cardiac output in the FMT_IHA group, which may indicate early cardiac functional impairment associated with aldosterone excess. Aldosterone excess has been implicated in myocardial remodeling, fibrosis, oxidative stress, and impaired ventricular compliance, which may contribute to early reductions in cardiac performance. The lack of significant differences in other echocardiographic parameters may be related to the relatively short modeling period, during which time structural cardiac changes may not yet have fully developed. In addition, measurements obtained under anesthesia may have been more sensitive to functional than structural differences between groups.39,40 Importantly, FMT mice reproduced several endocrine and metabolic alterations observed in human IHA, including elevated aldosterone and 5-HTP levels and selective alterations in tryptophan-related metabolites. However, not all clinical features of IHA were reproduced, indicating that the FMT model captures only part of the disease phenotype. Nevertheless, the concomitant elevation of 5-HTP and aldosterone in FMT_IHA mice should not be interpreted as evidence of a direct causal relationship, as mechanistic intervention studies were not performed in the present work. These findings provide in vivo evidence that the gut microbiota from IHA patients can promote aldosterone production, supporting a potential contribution of gut dysbiosis to hyperaldosteronism. Clinically, the gut microbiome might serve as novel biomarkers for risk stratification and disease monitoring in IHA. More importantly, they raise the possibility that microbiota-targeted interventions—such as dietary modifications, probiotics, or precision FMT—could potentially complement existing therapeutic strategies for hyperaldosteronism. Although the present study demonstrates associations among gut microbiota composition, circulating metabolites, and adrenal endocrine alterations, the biological pathways mediating these interactions remain undefined. And several features of human IHA were not fully reproduced in the FMT model. Potential mechanisms may involve endocrine, neural, immune, inflammatory, barrier-related, or other systemic processes. In conclusion, this study provides convergent evidence supporting a potential contribution of the gut microbiome to the pathophysiology of IHA. We provide evidence spanning from genetic causal inference, detailed molecular phenotyping in humans to direct experimental validation in GF mice. Our findings position the gut microbiota as a potential target for novel diagnostic and therapeutic strategies in IHA, thereby paving the way for new avenues in the mechanistic research of adrenal and other endocrine diseases. Limitations of the study This study has several limitations. First, the MR analysis was constrained by differences in ancestry and microbial profiling methods between the exposure and outcome GWAS datasets. The MR analysis was intended to identify candidate microbial taxa potentially associated with IHA, and the biological relevance of these taxa was further evaluated through metagenomic and FMT analyses. Second, although medications affecting RAAS were withdrawn before enrollment, residual effects of prior antihypertensive treatment may have persisted. Because the human cohort compared patients with IHA with healthy control subjects rather than matched hypertensive control subjects, we could not fully distinguish microbiome and metabolome alterations specific to IHA from those related to hypertension or associated clinical characteristics. Furthermore, because diabetes and impaired glucose commonly coexists with IHA,41 diabetic patients were not excluded. As a result, we cannot fully exclude the potential influence of metabolic comorbidities and related medications on gut microbiota composition. Future studies should include larger sample sizes and more stringent control of medication exposure, hypertension, and metabolic comorbidities on gut microbiota composition. Third, the FMT experiment employed pooled donor microbiota from a limited number of human donors, which reduced inter-individual variability but precluded assessment of donor-specific effects and reproducibility across independent donor inocula. While the FMT model reproduced key features of IHA, the mechanisms linking gut microbiota-derived metabolites to aldosterone overproduction remain unclear. Although 5-HTP was elevated in both IHA patients and FMT_IHA mice, it did not directly stimulate aldosterone secretion in HAC15 cells, indicating that any potential effect may be indirect or context-dependent. To minimize confounding effects related to hormonal fluctuations, only male mice were included. In addition, blood pressure was assessed using the tail-cuff method rather than radiotelemetry, which may be less sensitive for detecting subtle changes. Therefore, the absence of significant blood pressure differences in the FMT model should be interpreted cautiously. Future studies incorporating mechanistic interventions, strain-level analyses, both sexes, and longitudinal designs are warranted. Resource availability Lead contact Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Shumin Yang (443068494@qq.com). Materials availability This study did not generate new unique reagents. Data and code availability •Human metagenomic sequencing data are available in the NCBI Sequence Read Archive (SRA) under BioProject accession number PRJNA1494378. Mouse 16S rRNA sequencing data are available under BioProject accession number PRJNA1494439. Human untargeted metabolomics raw data have been deposited in MetaboLights under accession number MTBLS15210. Human metagenomic sequencing data are available in the NCBI Sequence Read Archive (SRA) under BioProject accession number PRJNA1494378. Mouse 16S rRNA sequencing data are available under BioProject accession number PRJNA1494439. Human untargeted metabolomics raw data have been deposited in MetaboLights under accession number MTBLS15210. •This study does not report original code. This study does not report original code. •Any additional information required to reanalyze the data reported in this study is available from the lead contact upon request. Any additional information required to reanalyze the data reported in this study is available from the lead contact upon request. Acknowledgments This research was financially supported by Chongqing Technology Innovation and Application Development Special Key Project (CSTB2025TIAD-KPX0098 and CSTB2024TIAD-KPX0039), Chongqing Medical Scientific Research Project (Joint project of Chongqing Health Commission and Science and Technology Bureau) (2025GGXM004), The Role of CMPF in Idiopathic Hyperaldosteronism: Mechanistic and Fundamental Research (KJZD-K202500409). Author contributions S.Y., Q.L., J.H., and Y. Jiang designed the study. Y. Jiang, Y. Jing, and Q. Zhang recruited subjects. Y. Jiang, S.Z., Y.X., and S.Q. conducted experiments. Y. Jiang, Q. Zeng, L.M., and Y.H. analyzed the data. S.Y., Y. Jiang, Q. Zeng, and L.M. drafted the manuscript. S.Y., Q.L., and J.H. supervised the project. All authors contributed to the article and approved the submitted version. Declaration of interests The authors declare no competing interests. STAR★Methods Key resources table REAGENT or RESOURCESOURCEIDENTIFIERBiological samplesHuman fecal samplesThis studyN/AHuman plasma samplesThis studyN/AMouse fecal samplesThis studyN/AMouse serum samplesThis studyN/AMouse kidney tissuesThis studyN/AMouse adrenal tissuesThis studyN/AChemicals, peptides, and recombinant proteins5-hydroxy-L-tryptophanMedChemExpress (MCE)HY- B1716Critical commercial assaysMouse Renin ELISA KitElabscienceE-EL-M0061Mouse ACTH ELISA KitBeyotimePA009FastPure Stool DNA Isolation KitMJYHT10-100NEXTFLEX Rapid DNA-Seq KitBioo ScientificNOVA-5144NovaSeq X Series 25B Reagent KitIllumina20104706PrimeScript™ RT Master Mix KitTaKaRaRR036AExperimental models: Cell linesHuman adrenocortical carcinoma cell line HAC15ATCCCRL-3301Experimental models: Organisms/strainsGerm-free C57BL/6J miceGemPharmatechN/AOligonucleotides16S V3-V4 Forward (338F) Primer: 5′-ACTCCTACGGGAGGCAGCAG-3′This studyN/A16S V3-V4 Reverse (806R) Primer: 5′-GGACTACHVGGGTWTCTAAT-3′This studyN/AGAPDH, Mouse, Forward Primer: 5′-TACTGGTGTCTTCACCACCATGG-3′This studyN/AGAPDH, Mouse, Reverse Primer: 5′-AGTTGTCATGGATGACCTTGG-3′This studyN/ARen1, Mouse, Forward Primer: 5′-ACGGGTCCGACTTCACCAT-3′This studyN/ARen1, Mouse, Reverse Primer: 5′-TGCCTAGAACACCGTCAAACT-3′This studyN/ADeposited dataHuman metagenomic sequencing dataThis studyNCBI Sequence Read Archive (SRA): PRJNA1494378Mouse 16S rRNA sequencing dataThis studyNCBI Sequence Read Archive (SRA): PRJNA1494439Human untargeted metabolomics dataThis studyMetabolights: MTBLS15210Software and algorithmsR software (v4.4.1)R Core Teamhttps://www.r-project.org/GraphPad Prism (v8.0.2)GraphPadhttps://www.graphpad.com/MetaboAnalyst (v6.0)Xia Labhttps://www.metaboanalyst.ca/ Experimental model and study participant details Human study participants This study was performed at the First Affiliated Hospital of Chongqing Medical University. A total of 60 participants were enrolled, including 30 patients with IHA and 30 HCs. Participants were allocated into the IHA and HC groups according to clinical diagnosis and eligibility criteria. The IHA group consisted of patients diagnosed with IHA based on established diagnostic criteria, whereas HCs were recruited from individuals undergoing routine health examinations who met the predefined inclusion criteria. Patients with IHA and HCs were aged 18–75 years. Exclusion criteria included: (a) other secondary hypertension (e.g., renal artery stenosis, pheochromocytoma, Cushing’s syndrome); (b) current hypertensive urgency (SBP ≥180 mmHg and/or DBP ≥120 mmHg); (c) recent use (within 1 month) of antibiotics, probiotics, laxatives, or immunosuppressants; (d) active infection or fever; (e) major chronic diseases (e.g., severe anemia, leukemia, asthma, epilepsy, inflammatory bowel disease); (f) history of major abdominal surgery; (g) positive fecal occult blood; (h) severe hepatic insufficiency (AST/ALT >100 IU/L); (i) renal insufficiency (eGFR <30 mL/min/1.73 m2); and (j) severe heart failure (NYHA class III–IV). Blood pressure was measured in a seated resting position by physicians. Two readings were recorded at 5-min intervals using an electronic blood pressure monitor (Omron, Japan), and the average of the two readings was used for analysis. IHA diagnosis followed established criteria. Prior to screening, antihypertensive drugs were discontinued or adjusted under physician supervision: diuretics for ≥4 weeks; angiotensin-converting enzyme inhibitors, angiotensin II receptor blockers, dihydropyridine calcium channel blockers (CCBs), and β-blockers for ≥2 weeks. Non-dihydropyridine CCBs and/or α-blockers were permitted for uncontrolled hypertension. Participants maintained a normal-sodium diet, and hypokalemia patients received oral potassium supplementation to keep serum potassium within reference range (3.5–5.0 mmol/L). Morning fasting blood samples were collected for plasma aldosterone concentration (PAC) and plasma renin concentration (PRC) measurements. PA screening was based on the aldosterone-renin ratio (ARR, cutoff ≥20 pg/uIU), confirmed by captopril challenge test (CCT, post-CCT PAC cutoff ≥110 pg/mL) and/or saline infusion test (SIT, post-SIT PAC cutoff ≥80 pg/mL). Then, IHA was diagnosed by adrenal venous sampling (lateralization index <2). HCs were required to be normotensive (SBP <130 mmHg and DBP <80 mmHg), free of chronic diseases (including those specified in exclusion criteria), without medication use within the past 30 days, and exhibit normal baseline laboratory parameters (including standard blood/urine/feces tests, hepatic/renal function, serum electrolytes, and fasting blood glucose). This study was approved by the Ethical Committee of the First Affiliated Hospital of Chongqing Medical University (No. 2024–002-01). Written informed consent was signed by all participants. All procedures adhered to the principles outlined in the Declaration of Helsinki.42 Germ-free mouse model Pooled fecal microbiota from two randomly selected IHA patients and two HCs were transplanted into GF mice. Fecal microbiota was prepared as reported in supplemental methods. Only male mice were used. Twenty male GF C57BL/6JGpt mice aged 7 ± 1 week were randomly divided into two groups of ten, housed five per cage. The environmental conditions were controlled at a temperature of 20°C–26°C, humidity of 40–70%, and a 12 h light/12 h dark cycle (light on from 8a.m. to 8p.m.). After a one-week adaptation period, all mice received FMT three times per week for four weeks, followed by an additional four-week period without FMT. Throughout the experimental period, body weight was monitored. All animal procedures performed in this study were approved by the Institutional Animal Care and Use of Chongqing Medical University (IACUC-CQMU-2025–0110). Cell culture The human adrenocortical carcinoma cell line HAC15 (CRL-3301) was obtained from the American Type Culture Collection (ATCC, USA). The cell line was authenticated by ATCC using short tandem repeat (STR) profiling and was reported to be free of mycoplasma contamination by the supplier. Cells were cultured in DMEM/F12 medium (Gibco, USA) supplemented with 10% bovine calf serum (SenBeiJia, China), 1% insulin-transferrin-selenium (Gibco, USA), and 1% penicillin-streptomycin solution (Hyclone, USA) at 37°C in a humidified atmosphere containing 5% CO2. Method details Two-sample Mendelian randomization The exposure data for the two-sample MR analysis were derived from the MiBiogen Consortium, a large-scale GWAS on human gut microbiota composition.12 The outcome data originated from our prior GWAS on IHA within the CONPASS (Chongqing Primary Aldosteronism Study) cohort.43 The GWAS data for IHA were derived from whole-genome sequencing of 261 IHA patients and 261 HC. The CONPASS cohort comprises hypertensive patients recruited from primary and tertiary care centers in Chongqing, China. Detailed characteristics of the CONPASS population have been reported previously.4,44,45,46 Instrumental variables (IVs) were selected as follows: (a) single nucleotide polymorphisms (SNPs) significantly associated with gut microbiota (p < 1×10−5); (b) SNPs in linkage disequilibrium were pruned using a clumping window of 10,000 kb and an R2 threshold <0.001 to ensure independence; (c) IVs with an F-statistic below 10 were excluded to avoid weak instrument bias, with the F-statistic calculated as F = Beta2/SE2.25,47 We performed the MR analysis using the TwoSampleMR package (v0.6.6) in R (v4.4.1). LD clumping was performed using the European reference panel implemented in TwoSampleMR, consistent with the ancestry of the exposure GWAS. During harmonization, palindromic SNPs with ambiguous strand orientation were removed using the standard TwoSampleMR pipeline. The primary causal estimate was obtained via the IVW method. Sensitivity analyses were performed using weighted median, weighted mode, MR-Egger, and MR-PRESSO methods. Heterogeneity among IVs was evaluated with Cochrane’s Q test for the IVW method and Rücker’s Q test for MR-Egger. Additionally, a leave-one-out sensitivity analysis was conducted to determine whether results were driven by individual influential or pleiotropic SNPs.48 Sample collection of study population Plasma samples were separated and stored at −80°C until liquid chromatography-tandem mass spectrometry (LC-MS) metabolomic profiling. Stool samples were collected from subjects following standardized protocol. Each sample was immediately divided into two aliquots under controlled conditions: (a) approximately 10 g was transferred to a sterile cryotube containing 30 mL of fecal microbiota preservation solution (0.9% NaCl, 30% glycerol, 0.1% L-cysteine, w/v), shaken 30 s to ensure homogenization; (b) the remaining sample was placed in a dry sterile cryotube. Both tubes were placed in ice packs within 5 min of collection, transported to the laboratory within 2 h, and stored at −80°C for fecal microbiota transplantation (FMT) or shotgun metagenomic sequencing. Metagenomic sequencing and analysis Fecal DNA was extracted using the FastPure Stool DNA Isolation Kit (MJYH, China). DNA concentration and purity were measured using SynergyHTX and NanoDrop2000, respectively, and integrity was verified by 1% agarose gel electrophoresis. The DNA was then fragmented to an average size of about 350 bp using Covaris M220 (Gene Company Limited, China). A paired-end library was prepared with the NEXTFLEX Rapid DNA-Seq Kit (Bioo Scientific, USA) and sequenced on the Illumina NovaSeq X Plus platform (Illumina, USA) using the NovaSeq X Series 25B Reagent Kit. Alpha diversity was quantified using the Shannon index. Beta diversity was evaluated through PCoA, with group differences statistically validated by PERMANOVA. Differential abundance analysis employed LEfSe to identify taxonomic features with biological consistency (LDA score >2, p < 0.05). The relationships between the microbiome, metabolites, and clinical parameters were assessed using Spearman correlations. Please refer to the supplementary methods for further details on metagenomic sequencing and analysis. LC-MS metabolomics analysis Untargeted plasma metabolomic profiling was performed on a Vanquish UHPLC system (Thermo Fisher Scientific, USA) coupled to an Orbitrap Exploris 480 mass spectrometer. The data processing flow was consistent with the protocol described in our previous study.49 Sum-normalization data were analyzed by the ropls package in R (v4.3.1), including OPLS-DA.Model validity was confirmed by 7-fold cross-validation and permutation test. Variable importance in projection (VIP) scores were calculated from OPLS-DA loadings. Differential metabolites were defined by VIP >1.5 and Student’s t test p < 0.05. KEGG pathway analysis of the differential metabolites was performed on MetaboAnalyst platform (https://www.metaboanalyst.ca/MetaboAnalyst/).50 Please refer to the supplemental methods for further details on metabolomics analysis. Animal experiments of fecal microbiota transplantation (FMT) Blood pressure was measured over a four-day period, which included three days of adaptation to the restraint and measurement procedure. Before each session, mice were allowed to acclimate undisturbed in the testing room for at least 30 min. Mice were gently restrained, and tail-cuff sensors (O-CUFF) and volume-pressure recording cuffs were applied. Data were collected using CODA, and only stable readings where systolic and diastolic pressures fell within consistent ranges, with tail blood volume exceeding 15 units, were selected. Five representative data points near the mean systolic pressure were chosen. For echocardiography, mice were anesthetized with isoflurane (2–3%) and oxygen mixture in an induction chamber, then placed supine on a platform with integrated electrocardiogram (ECG) leads. Ultrasound gel was applied to the chest area, and standard two-dimensional images (parasternal short-axis and apical four-chamber views) were captured and saved. Sample collection was initiated at approximately 10:00 a.m. during the light cycle. At the study endpoint, mice were anesthetized via intraperitoneal injection of ketamine/xylazine (100 mg/kg and 5 mg/kg). Blood samples were obtained via retro-orbital bleeding, and serum was separated by centrifugation at 3000g for 15 min. Serum and freshly collected fecal samples were stored at −80°C until further analysis. Following PBS perfusion, kidney and adrenal tissues were harvested for subsequent analyses. Serum aldosterone and corticosterone levels were quantified using ultra-performance liquid chromatography–tandem mass spectrometry following our previously published protocol.51,52 Additional details regarding FMT preparation, targeted tryptophan metabolomics, renin and ACTH measurements, PCR and adrenal histology are provided in the Supplemental Methods. 16S rRNA gene sequencing analysis Fecal DNA was extracted following methods of metagenomic studies, as well as quality and concentration of DNA. The bacterial 16S rRNA gene V3-V4 hypervariable regions were amplified in triplicate using primers 338F (5′-ACTCCTACGGGAGGCAGCAG-3′) and 806R (5′-GGACTACHVGGGTWTCTAAT-3′) on a T100 Thermal Cycler (Bio-Rad, USA). Amplicons were purified after electrophoresis on 2% agarose gels, quantified with Synergy HTX (BioTek, USA), and pooled in equimolar concentrations. Libraries were sequenced on an Illumina NextSeq 2000 platform (Illumina, USA). Raw sequences underwent quality control with fastp (v0.23.4),53 merged with FLASH (v1.2.11),54 and denoised with DADA2 (Qiime2 v2024.2).55 Resulting ASVs were rarefied to 20000 sequences per sample (Good’s coverage = 97.90%) to normalize sequencing depth. Taxonomic classification employed a Naive Bayes classifier trained on Qiime2 and the SILVA 16S rRNA database (v138). The metagenomic function was predicted by PICRUSt2 (v1.1.0)56 based on ASV representative sequences. Downstream analyses (alpha diversity: Shannon index; beta diversity: PCoA validated by PERMANOVA; differential abundance: LEfSe) followed identical workflows to the metagenomic dataset. Quantification and statistical analysis Statistical analyses were performed using R (v4.4.1) or GraphPad Prism (v8.0.2). Categorical variables are presented as numbers and frequencies. Continuous variables are expressed as mean ± SD, unless otherwise specified. Statistical tests were selected according to data distribution and experimental design, including the Chi-square test, Student’s t test, Wilcoxon rank-sum test (Mann-Whitney U test), hypergeometric test, and PERMANOVA as appropriate. Mendelian randomization analyses were performed using the TwoSampleMR package (v0.6.6), with details of instrumental variable selection, causal estimation methods, and sensitivity analyses described in the corresponding Methods section. For microbial diversity analyses, α-diversity was assessed using the Shannon index, and beta diversity was evaluated by principal coordinate analysis (PCoA) with group differences assessed using PERMANOVA. Differential abundance analyses were performed using LEfSe where applicable. For metabolomics analyses, multivariate statistical analyses were performed using the ropls package, and differential metabolites were identified based on predefined criteria. Correlations among microbiome, metabolomic, and clinical variables were evaluated using Spearman’s correlation analysis. Multivariable linear regression analyses were performed for differentially abundant microbial taxa and metabolites. Group status was included as the independent variable of interest, while gender, age, BMI, SBP, serum potassium, creatinine, smoking status, and alcohol consumption were included as covariates. Adjusted P were calculated for the group effect. All statistical details, including exact sample sizes (n), definition of biological replicates, statistical tests, and exact p values are provided in the corresponding figure legends or Results section,. For animal experiments, n represents the number of individual animals; for cell experiments, n represents independent biological replicates unless otherwise specified. Statistical significance was defined as p < 0.05. Footnotes Supplemental information can be found online at https://doi.org/10.1016/j.isci.2026.117426. Supplemental information Document S1. Figures S1–S3 and Methods S1–S10 and references Table S1. Instrumental variables used in MR analysis about the casual association between gut microbiota and IHA Table S2. Heterogeneity of gut microbiota instrumental variables Table S3. Directional horizontal pleiotropy assessed by intercept term in MR Egger regression Table S4. Sensitivity analyses of the associations between gut microbial taxa and IHA Table S5. MR-PRESSO global test for horizontal pleiotropy of microbial taxa associated with IHA Table S6. The detailed clinical characteristics of study population Table S7. Reads acquisition of metagenomic sequencing data Table S8. Overlapping microbiota between metagenomic sequencing and MR analysis Table S9. Significantly altered genera in metagenomic sequencing by LEfSe Table S10. Significantly altered species in metagenomic sequencing by LEfSe Table S11. Significant alterations in level 3 microbial functional pathways Table S12. Significant alterations in level 2 microbial functional pathways Table S13. Differential expressed metabolites in plasma between IHA and HC Table S14. Spearman correlation analysis between clinical characteristics and IHA-specific species Table S15. The clinical characteristics of microbiota donors Table S16. Significantly altered microbial functional pathways in mice by PICRUSt2 Table S17. Targeted metabolomic profiling of the tryptophan metabolic pathway References Associated Data Supplementary Materials Document S1. Figures S1–S3 and Methods S1–S10 and references Table S1. Instrumental variables used in MR analysis about the casual association between gut microbiota and IHA Table S2. Heterogeneity of gut microbiota instrumental variables Table S3. Directional horizontal pleiotropy assessed by intercept term in MR Egger regression Table S4. Sensitivity analyses of the associations between gut microbial taxa and IHA Table S5. MR-PRESSO global test for horizontal pleiotropy of microbial taxa associated with IHA Table S6. The detailed clinical characteristics of study population Table S7. Reads acquisition of metagenomic sequencing data Table S8. Overlapping microbiota between metagenomic sequencing and MR analysis Table S9. Significantly altered genera in metagenomic sequencing by LEfSe Table S10. Significantly altered species in metagenomic sequencing by LEfSe Table S11. Significant alterations in level 3 microbial functional pathways Table S12. Significant alterations in level 2 microbial functional pathways Table S13. Differential expressed metabolites in plasma between IHA and HC Table S14. Spearman correlation analysis between clinical characteristics and IHA-specific species Table S15. The clinical characteristics of microbiota donors Table S16. Significantly altered microbial functional pathways in mice by PICRUSt2 Table S17. Targeted metabolomic profiling of the tryptophan metabolic pathway Data Availability Statement •Human metagenomic sequencing data are available in the NCBI Sequence Read Archive (SRA) under BioProject accession number PRJNA1494378. Mouse 16S rRNA sequencing data are available under BioProject accession number PRJNA1494439. Human untargeted metabolomics raw data have been deposited in MetaboLights under accession number MTBLS15210. Human metagenomic sequencing data are available in the NCBI Sequence Read Archive (SRA) under BioProject accession number PRJNA1494378. Mouse 16S rRNA sequencing data are available under BioProject accession number PRJNA1494439. Human untargeted metabolomics raw data have been deposited in MetaboLights under accession number MTBLS15210. •This study does not report original code. This study does not report original code. •Any additional information required to reanalyze the data reported in this study is available from the lead contact upon request. Any additional information required to reanalyze the data reported in this study is available from the lead contact upon request.