Unraveling the Causal Association Between microRNAs and Ulcerative Colitis: A Multi-Stage Mendelian Randomization Study with External Validation

Unraveling the Causal Association Between microRNAs and Ulcerative Colitis: A Multi-Stage Mendelian Randomization Study with External Validation | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Unraveling the Causal Association Between microRNAs and Ulcerative Colitis: A Multi-Stage Mendelian Randomization Study with External Validation Jin He, Haimin Jin, Weijian Chu This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8634046/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 3 You are reading this latest preprint version Abstract Background Circulating microRNAs (miRNAs) are frequently dysregulated in ulcerative colitis (UC), yet whether these alterations represent causes or consequences of the disease remains unclear. This study aimed to systematically investigate the causal effect of plasma miRNAs on UC risk using a comprehensive multi-stage Mendelian randomization (MR) framework. Methods We employed a two-sample MR design utilizing genetic instruments for miRNAs derived from independent cis- and trans-eQTL datasets (Nikpay et al. and Framingham Heart Study). Outcome summary statistics were obtained from the FinnGen (n = 499,380) and EBI GWAS Catalog (n = 458,440) cohorts. Analyses included inverse variance weighted estimation, rigorous sensitivity checks, Bayesian colocalization to assess pleiotropy, and reverse MR. Findings were further biologically interpreted through target gene functional enrichment and triangulated with external transcriptomic validation (GSE122618). Results The miR-130b family (miR-130b-3p and miR-130b-5p) demonstrated consistent protective effects against UC across discovery and validation phases (e.g., miR-130b-5p: OR = 0.91–0.92, P < 0.05). Conversely, miR-101-3p, miR-1908-5p, miR-27a-3p, and let-7a-5p were identified as significant risk factors. Reverse MR analyses provided no robust evidence that UC liability influences these miRNA levels. Bayesian colocalization strongly supported shared causal variants for miR-27a-3p (PP.H4 = 0.61) and miR-1908-5p. Functional enrichment linked miR-130b targets to focal adhesion and MAPK/Wnt signaling pathways critical for mucosal homeostasis. Conclusion This study provides compelling genetic evidence that specific circulating miRNAs causally influence UC susceptibility. The identification of protective (miR-130b family) and risk-increasing miRNAs advances our understanding of UC pathogenesis, highlighting these molecules as promising biomarkers and potential therapeutic targets. miRNAs ulcerative colitis Mendelian randomization Bayesian colocalization pleiotropy Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Ulcerative colitis (UC) is a chronic, relapsing inflammatory bowel disease (IBD) characterized by continuous mucosal inflammation of the colon and rectum, leading to bloody diarrhea, abdominal pain, and substantial impairment in quality of life. Recent estimates indicate that the prevalence of IBD has exceeded 0.3% in many Western countries, with UC prevalence reaching up to 286–505 per 100,000 individuals in North America and Northern Europe [ 1 ]. In the United States alone, 2.4–3.1 million adults are living with IBD, imposing a considerable health‑care and economic burden [ 2 – 4 ]. Despite advances in biologic and small‑molecule therapies, a significant proportion of patients exhibit primary non‑response, loss of response, or treatment‑limiting adverse events, and many ultimately require colectomy [ 5 ]. These limitations highlight the need to better understand the molecular mechanisms that drive UC and to identify novel biomarkers and therapeutic targets. MicroRNAs (miRNAs) are ~ 22-nucleotide non-coding RNAs that repress gene expression post-transcriptionally by binding to complementary sequences in target mRNAs. They are now recognized as key regulators of immune cell differentiation, cytokine production, epithelial barrier function, and cell death pathways [ 6 ]. Multiple profiling studies have demonstrated widespread miRNA dysregulation in colonic mucosa from patients with active UC. Early microarray work identified altered expression of at least 11 miRNAs in UC tissue, including downregulation of miR-192 and upregulation of miR-155; miR-192 was shown to directly regulate the chemokine MIP-2α in colonic epithelial cells [ 7 ]. Subsequent studies confirmed increased mucosal levels of miR-21, miR-31, miR-155, miR-125b and others, which modulate IL-13 signaling, NF-κB activation, mucus production, and goblet cell differentiation, linking specific miRNAs to key elements of UC pathophysiology [ 8 ]. Pediatric UC cohorts have further revealed distinct miRNA signatures, such as downregulation of epithelial miR-4284 with reciprocal upregulation of its target CXCL5, underscoring the relevance of miRNA-mediated networks across age groups [ 9 ]. Beyond the intestinal mucosa, circulating and fecal miRNAs have emerged as promising non-invasive biomarkers for IBD. Several case–control studies reported increased levels of serum or plasma miR-16, miR-21, miR-155, miR-199a-5p, miR-223 and other miRNAs in patients with UC compared with healthy controls, some of which correlate with clinical or endoscopic activity[ 10 ]. Genome-wide profiling of blood cell fractions and microvesicles identified multi-miRNA circulating signatures that can distinguish UC from healthy individuals with high accuracy [ 11 ]. Additional work has shown that circulating miRNA panels can differentiate Crohn’s colitis from UC, discriminate active disease from remission, and track treatment responses [ 12 ]. Collectively, these observations support the utility of circulating miRNAs as diagnostic and monitoring tools in UC. However, because these studies are observational, they cannot determine whether altered miRNA levels are causal drivers of disease onset, reflect downstream inflammation, or represent epiphenomena related to treatment and comorbidities. Mendelian randomization (MR) offers a powerful epidemiologic framework to address causality by using germline genetic variants as instrumental variables (IVs) for modifiable exposures. Under core assumptions, genetic proxies for an exposure (e.g. biomarker levels) are randomly allocated at conception and are largely independent of confounders and reverse causation, allowing estimation of causal effects on disease outcomes using observational data [ 13 ]. With the advent of large GWAS, two-sample MR using summary statistics has been widely applied to UC and IBD to interrogate the causal roles of gut microbiota, circulating inflammatory proteins, blood metabolites, glucose, immune cell subsets, and druggable proteins. These studies have identified specific microbial genera, cytokines (e.g. IL-10 receptor subunits, chemokines), and metabolic pathways that appear to influence UC risk and may mediate immune–metabolic crosstalk [ 14 ]. Nevertheless, MR investigations focusing on non-coding RNAs in UC remain scarce, and the contribution of circulating miRNAs to UC susceptibility has not been systematically evaluated. In this study, we employed a comprehensive, multi-stage study design to systematically investigate the causal effect of plasma miRNAs on the risk of UC. Unlike previous studies, we integrated a two-sample MR approach with rigorous sensitivity analyses, including Bayesian colocalization, to distinguish true causality from linkage disequilibrium. Furthermore, to elucidate the biological plausibility, we mapped the downstream target genes of causal miRNAs to immune-related pathways. Uniquely, we triangulated our genetic findings with external transcriptomic validation using independent expression data from the Gene Expression Omnibus (GEO). By combining genetic causal inference with functional annotation and real-world expression profiling, this study aims to identify robust, causally relevant miRNA biomarkers for UC, offering new perspectives on disease pathogenesis and potential therapeutic interventions. Materials and methods Study Design and Data Sources We employed a two-sample MR design to systematically investigate the causal effect of circulating miRNAs on the risk of UC. To ensure the robustness and replicability of our findings, a two-stage strategy comprising a discovery phase and a validation phase was implemented. Genetic instruments for miRNAs were obtained from two independent eQTL datasets based on the hg38 genome assembly. The Nikpay et al. dataset (n = 710) served as the primary source [ 15 ], utilizing cis-eQTLs for the discovery phase and trans-eQTLs for the validation phase, while cis-eQTLs from the Framingham Heart Study (FHS, n = 5239) were employed as an additional validation source [ 16 ]. Cis-acting IVs were strictly defined as SNPs located within 250 kb of the corresponding miRNA gene region. For the outcome phenotypes, summary-level data for UC were sourced from two large-scale cohorts of European ancestry to minimize sample overlap. The primary analysis utilized data from the FinnGen consortium (Release 12), comprising 7,220 cases and 492,160 controls defined by strict clinical endpoints. To corroborate the findings, a replication dataset was accessed from the EBI GWAS Catalog (Accession ID: GCST90473823), which included 6,158 cases and 452,282 controls. Instrument Selection and Statistical Analysis The MR analysis was conducted using the TwoSampleMR package in R software. We performed a preliminary screening of IVs using a significance threshold of P < 5e − 6 to capture potential causal variants. To ensure independence among the selected instruments, SNPs were clumped using a strict linkage disequilibrium (LD) threshold of r2 < 0.001 within a 10,000 kb window. Palindromic SNPs with intermediate allele frequencies were inferred based on allele frequency information. To avoid weak instrument bias, the strength of each genetic instrument was assessed using the F-statistic (F = β2/SE2), and any SNPs with an F-statistic < 10 were excluded from the analysis. The Wald ratio method was applied for miRNAs instrumented by a single SNP, while the inverse variance weighted (IVW) method was employed for exposures with multiple valid instruments. In the discovery phase, miRNA-UC associations with a P-value < 0.05 were identified as potential causal candidates and were subsequently advanced to the validation phase using independent IVs or outcome datasets. Sensitivity Analyses and Pleiotropy Assessment To verify the validity of the causal assumptions, we performed rigorous sensitivity analyses. The Steiger directionality test was conducted to confirm the direction of causality, ensuring that the genetic variants explained significantly more variance in the miRNA expression than in the UC risk. Furthermore, to address potential horizontal pleiotropy, we utilized the LDlinkR package to query the LDtrait tool. We systematically screened the identified causal SNPs against the GWAS Catalog for associations with known confounders of UC, including smoking behaviors, body mass index (BMI), other autoimmune diseases (e.g., Crohn’s disease, psoriasis), and antibiotic usage. SNPs showing significant associations with these confounding traits were flagged to exclude the possibility that the observed estimates were driven by alternative pathways independent of the miRNA exposure. Reverse Mendelian Randomization Analysis To investigate the possibility of reverse causality—specifically, whether genetic predisposition to UC causally influences the expression levels of the identified miRNAs—we performed a reverse MR analysis. In this analysis, UC was treated as the exposure and the candidate miRNAs identified in the forward analysis served as the outcomes. Genetic instruments for UC were extracted from the same two independent GWAS datasets used in the forward analysis (FinnGen and GCST90473823). Significant SNPs associated with UC were selected using a genome-wide significance threshold of P < 5e − 8. To ensure independence, these instruments were clumped using a LD threshold of r2 < 0.001 within a 10,000 kb window, based on the 1000 Genomes Project European reference panel. Summary statistics for the outcome miRNAs were retrieved from the full genome-wide association statistics of the Nikpay et al. study. The causal effects were estimated using the IVW, Weighted Median, and Wald Ratio methods, consistent with the forward analysis. A P-value < 0.05 in the reverse direction was considered indicative of potential reverse causality or bidirectional effects. Bayesian Colocalization Analysis To distinguish whether the identified associations between miRNAs and UC were driven by a shared causal variant (pleiotropy) or by distinct variants in linkage disequilibrium, we performed Bayesian colocalization analysis using the coloc R package. We focused on the genomic regions surrounding the lead SNPs identified in the MR analysis. For each candidate miRNA, we defined a 400 kb window (± 200 kb) centered on the top cis-eQTL. Since the miRNA summary statistics were based on the hg19 genome assembly while the UC data utilized hg38, we performed a genomic coordinate conversion using the LiftOver tool with the hg19-to-hg38 chain file to ensure precise alignment. We extracted the summary statistics (beta coefficients, standard errors, and minor allele frequencies) for all SNPs within these regions from both the miRNA eQTL dataset (Nikpay et al.) and the UC GWAS datasets (FinnGen and GCST90473823). The coloc.abf function was employed to calculate the posterior probabilities (PP) for five distinct hypotheses: H0 (no association with either trait), H1 (association with miRNA only), H2 (association with UC only), H3 (association with both traits but distinct causal variants), and H4 (association with both traits sharing a single causal variant). A high posterior probability for H4 (PP.H4 > 0.8) was considered strong evidence of colocalization, supporting a shared genetic mechanism underlying both miRNA expression and UC risk. Target Prediction and Functional Enrichment Analysis To elucidate the potential biological mechanisms linking the identified causal miRNAs to UC, we performed a downstream functional enrichment analysis. First, experimentally validated target genes of the significant miRNAs were retrieved from the multiMiR database (version 2.3.0), focusing exclusively on interactions supported by strong experimental evidence ("validated" table) to ensure reliability. The resulting gene list was then mapped to Entrez IDs using the org.Hs.eg.db package. Subsequently, we conducted Gene Ontology (GO) Biological Process and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses using the clusterProfiler package in R. Statistical significance was determined using the Benjamini-Hochberg (BH) method to adjust P-values for multiple testing, with a threshold of adjusted P < 0.05. To specifically investigate the relevance of these targets to UC pathogenesis, we prioritized and visualized pathways containing keywords associated with intestinal inflammation and immune regulation, such as "Inflammatory bowel disease", "Th17 cell differentiation", "Cytokine-cytokine receptor interaction", and "NF-kappa B signaling pathway". External Validation Using Transcriptomic Data To corroborate the causal associations identified by Mendelian randomization, we performed an external validation using independent transcriptomic expression profiling data retrieved from the GEO database. We utilized the dataset GSE122618, which contains plasma miRNA expression profiles from patients with UC and healthy controls. The dataset comprised 39 UC samples and 7 healthy control samples. Raw count data were imported into R and subjected to rigorous preprocessing. To account for differences in sequencing depth across samples, raw counts were normalized to Counts Per Million (CPM) and subsequently log2-transformed (log2 (CPM + 1)) to stabilize variance and ensure comparability. We specifically extracted the expression levels of the candidate miRNAs identified in the MR analysis. The differential expression of these miRNAs between the UC and control groups was evaluated using the non-parametric Wilcoxon rank-sum test (Mann-Whitney U test), as the data did not necessarily follow a normal distribution. Visualizations were generated using the ggpubr package, and a P-value < 0.05 was considered statistically significant for validation. Results Identification of Causal miRNAs in the Discovery Phase In the discovery phase, we utilized cis-acting eQTLs from the Nikpay dataset to screen for miRNAs causally associated with UC (Supplementary materials: Table S1 ). All selected IVs exhibited strong predictive power for miRNA expression, with F-statistics ranging from 22.13 to 200.46 (mean F = 50.8), well above the conventional threshold of 10, indicating a low risk of weak instrument bias. Analysis of the FinnGen dataset identified seven miRNAs significantly associated with UC risk (P < 0.05, Fig. 2 ). Specifically, genetically predicted higher levels of miR-101-3p (OR = 4.03, 95% CI: 1.96–8.27, P = 1.43e − 4) and miR-1908-5p (OR = 1.61, 95% CI: 1.29–2.01, P = 2.83e − 5) were strongly associated with an increased susceptibility to UC based on the Wald ratio method. Conversely, miR-130b-3p (OR = 0.81, 95% CI: 0.68–0.97, P = 0.019) and miR-130b-5p (OR = 0.67, 95% CI: 0.48–0.94, P = 0.020) demonstrated significant protective effects. In the GCST dataset, six miRNAs were identified as potential causal determinants. Notably, miR-27a-3p (OR = 9.47, P = 8.68e − 4) and let-7a-5p (OR = 9.54, P = 0.002) exhibited substantial risk effects. Most importantly, the protective effect of miR-130b-3p was consistently replicated in the GCST cohort using the IVW method (OR = 0.82, 95% CI: 0.67–0.99, P = 0.044), reinforcing its potential role as a robust protective factor against UC. The Steiger directionality test confirmed the validity of the causal direction for all significant associations (all Steiger P < 0.05), ruling out reverse causality in this phase. Validation of Causal Associations Using Alternative Instruments To further substantiate the robustness of our findings, we conducted a validation analysis using alternative sets of IVs derived from trans-eQTLs (Nikpay et al.) and cis-eQTLs from the FHS cohort (Fig. 3 , Supplementary materials: Table S2 ). The strength of these instruments remained robust across all analyses (F-statistics > 22). In the validation using trans-eQTLs from the Nikpay dataset, miR-4326 showed a significant risk effect on UC in the FinnGen cohort (OR = 1.22, 95% CI: 1.07–1.39, P = 0.003), although this association was not statistically significant in the GCST cohort (P = 0.563). Other candidates, including miR-130b-3p, did not reach statistical significance using trans-IVs in either outcome dataset, likely due to the inherent complexity and potential pleiotropy of trans-acting variants. However, validation using cis-eQTLs from the FHS dataset yielded compelling results. miR-130b-5p was significantly associated with a reduced risk of UC in both the FinnGen cohort (OR = 0.92, 95% CI: 0.86–1.00, P = 0.045) and the GCST cohort (OR = 0.91, 95% CI: 0.84–0.99, P = 0.038). This finding aligns with the protective effect observed for the miR-130b family in the discovery phase. Additionally, miR-130b-3p showed a trend towards protection in the FinnGen cohort (OR = 0.94, 95% CI: 0.89–0.99, P = 0.031), further supporting the potential therapeutic relevance of the miR-130b cluster. The Steiger directionality test confirmed the correct causal direction for all significant estimates (all P < 0.05). Assessment of Reverse Causality To investigate the potential reverse causal effects of UC on the identified candidate miRNAs, we performed a reverse Mendelian Randomization analysis using genetic instruments associated with UC from two independent large-scale GWAS datasets (FinnGen and GCST). In the analysis using the FinnGen UC dataset as the exposure, the IVW method showed no significant causal association between genetic liability to UC and the expression levels of any tested miRNAs (all P > 0.05, Supplementary materials: Table S3 ). Sensitivity analysis using the weighted median method yielded consistent non-significant estimates, further suggesting that UC liability does not influence the levels of these miRNAs in this cohort. Validation analysis using the GCST UC dataset largely corroborated these findings. The majority of miRNAs showed no significant association with UC liability. Although nominal associations were observed for miR-6810-3p (β = 0.070, SE = 0.033, P = 0.032) and miR-130b-5p (β = 0.074, SE = 0.037, P = 0.048) under the IVW method in the GCST cohort, these associations were not statistically significant in the weighted median sensitivity analysis (P = 0.051 and P = 0.099, respectively). Furthermore, given the lack of replication in the FinnGen dataset (β = 0.027, P = 0.318 for miR-6810-3p; β = -0.046, P = 0.070 for miR-130b-5p), the evidence for a reverse causal effect remains weak and inconsistent. Collectively, the reverse MR analyses across two independent cohorts provide no robust evidence to support a reverse causal pathway from UC to the identified miRNAs, thereby strengthening the validity of the causal effects observed in the forward MR analysis. Bayesian Colocalization Analysis To distinguish between shared causal variants (pleiotropy) and distinct variants in linkage disequilibrium, we performed Bayesian colocalization analysis for the identified miRNA-UC associations (Fig. 4 , Supplementary materials: Table S4 ). A high PP.H4 indicates a shared causal variant. In the analysis of the GCST cohort, miR-27a-3p exhibited the strongest evidence for colocalization, with a PP.H4 of 0.608, suggesting a potential shared genetic mechanism at the 19p13.11 locus (lead SNP rs2594718). Similarly, let-7a-5p and let-7d-5p showed moderate support for colocalization in the GCST dataset (PP.H4 = 0.409 and 0.430, respectively). In the FinnGen cohort, miR-1908-5p displayed suggestive evidence of colocalization (PP.H4 = 0.583) at the 11q12.1 locus, whereas its colocalization probability in the GCST cohort was low. However, for miR-130b-3p and miR-130b-5p, which demonstrated consistent protective effects in the MR analysis, the evidence for colocalization was limited in both the FinnGen (PP.H4 ≈ 0.11) and GCST (PP.H4 ≈ 0.04) cohorts. This relatively low PP.H4, combined with low PP.H3 values (probability of distinct causal variants), suggests that while the MR assumptions hold, the current GWAS summary statistics may lack sufficient resolution to definitively pinpoint a single shared variant, or the association may be driven by multiple variants in the region. Other candidates, such as miR-101-3p in FinnGen, showed high support for distinct causal variants (PP.H3 = 0.95), indicating that the observed association might be partly driven by linkage disequilibrium rather than direct pleiotropy. Integrative Functional Enrichment Analysis of miR-130b-3p/5p Targets To comprehensively characterize the regulatory network of miR-130b, we predicted the downstream targets for both miR-130b-3p and miR-130b-5p, identifying 5,149 and 346 potential target genes (Supplementary materials: Table S5 ), respectively. Given the overlapping biological contexts of these distinct miRNA strands, we performed a combined functional enrichment analysis to delineate their collective biological impact. Hierarchical clustering of GO terms highlighted that the predicted targets are functionally interconnected, clustering into five distinct biological modules (Fig. 5 ). A substantial proportion of targets were enriched in clusters related to structural organization and adhesion (e.g., "actin filament organization," "cell-substrate adhesion"), suggesting a critical role in regulating cellular morphology and motility. Furthermore, a distinct cluster enriched in cell cycle progression terms (e.g., "Wnt signaling pathway," "chromosome segregation") indicated that miR-130b may function as a proliferative regulator. Notably, we also observed specific enrichment in neurodevelopmental processes (e.g., "axonogenesis," "forebrain development"), hinting at a potential tissue-specific function in the nervous system. Consistent with the GO findings, KEGG pathway analysis demonstrated that the targets were predominantly enriched in pathways governing cell-matrix interactions and signal transduction (Fig. 6 ). "Focal adhesion" was identified as the most significantly enriched pathway (lowest adjusted P-value), reinforcing the GO results regarding cell adhesion. Additionally, the analysis highlighted broad regulatory potential in multiple signaling cascades, including the MAPK, FoxO, and Insulin signaling pathways. It is noteworthy that several cancer-related pathways (e.g., "Proteoglycans in cancer," "Prostate cancer," "Renal cell carcinoma") were highly represented, implying that miR-130b-3p/5p dysregulation may be a pivotal event in tumorigenesis and cancer progression. External Validation in Plasma Samples from UC Patients To evaluate whether the identified causal miRNAs exhibit detectable alterations in circulating biofluids, we further examined their expression levels in plasma samples from an independent cohort of UC patients and healthy controls. In this plasma-based validation, none of the candidate miRNAs reached the conventional threshold for statistical significance (P < 0.05, Supplementary materials: Table S6 , Figure S1 ). However, notable trends were observed for the miR-130b family. Specifically, miR-130b-3p showed a trend towards upregulation in UC patients compared to controls (log2 FC = 0.37, P = 0.075), as did miR-130b-5p (log2 FC = 0.58, P = 0.080). Other candidates, such as miR-671-3p and miR-27a-3p, also displayed elevated expression levels in the UC group but lacked statistical support (P > 0.10). The absence of significant differences in circulating miRNAs, in contrast to tissue-specific associations, may reflect the dilution effect in peripheral blood or distinct regulatory dynamics between local intestinal tissue and systemic circulation. Discussion This comprehensive two-sample Mendelian randomization study provides genetic evidence that multiple circulating miRNAs are involved in UC susceptibility. Across discovery and validation phases, we observed a pattern in which members of the miR-130b family (miR-130b-3p and miR-130b-5p) showed consistent protective associations, whereas several other miRNAs—most prominently miR-101-3p, miR-1908-5p, miR-27a-3p, let-7a-5p, miR-103a-3p, and miR-4326—were associated with increased UC risk in at least one dataset. Additional candidates, including miR-6891-3p, miR-6810-3p, miR-671-3p, and let-7d-5p, showed suggestive associations and biologically plausible links to mucosal immunity or inflammatory signaling. Together with colocalization, enrichment, reverse-MR and expression analyses, these findings extend previous observational work on miRNAs in inflammatory bowel disease (IBD) and highlight a mechanistically diverse set of miRNAs that may contribute to UC pathogenesis. The most robust and internally consistent finding in our study is the protective association of the miR-130b family. In the discovery phase, genetically predicted higher levels of miR-130b-3p and miR-130b-5p were associated with reduced UC risk in FinnGen (OR ~ 0.67–0.81). This protective pattern was replicated for miR-130b-3p in the GCST dataset and for miR-130b-5p using independent cis-eQTLs from the Framingham Heart Study in both UC GWAS cohorts. The consistency across instruments (cis-eQTLs in two cohorts) and outcomes, as well as Steiger tests supporting the miRNA→UC direction, make reverse causation unlikely. Functionally, miR-130b has been implicated in several inflammatory and tissue-remodeling contexts. In adipose tissue, miR-130b promotes obesity-associated inflammation and insulin resistance by repressing PPAR-γ and skewing macrophages toward a pro-inflammatory M1 phenotype, linking it to metabolic inflammation [ 17 ]. In contrast, in diabetic nephropathy, miR-130b mimics attenuated renal tubulointerstitial fibrosis by inhibiting Snail-driven epithelial–mesenchymal transition (EMT) and restoring E-cadherin expression, thereby limiting fibrotic remodeling [ 18 ]. More recently, miR-130b-3p has been shown to dampen Th2-type airway inflammation in asthma by targeting HMGB1 and suppressing the HMGB1–TLR4–DRP1 axis [ 19 ]. These apparently divergent effects suggest that miR-130b is a context-dependent regulator of inflammatory and EMT-related pathways, with potential to be either detrimental (e.g. in metabolic inflammation or colorectal cancer) or protective (e.g. in fibrotic kidney disease and allergic airway inflammation) depending on cellular context and target engagement. Our MR results, which integrate genetically proxied lifelong higher circulating miR-130b levels, align more closely with its anti-inflammatory/fibrosis-limiting roles. In chronic intestinal inflammation, where barrier injury, EMT-like changes, and innate immune activation coexist, up-regulation of miR-130b may preferentially restrain profibrotic and HMGB1/TLR-driven inflammatory circuits, outweighing any adverse effects seen in tumorigenic settings. This interpretation is supported by our enrichment analysis, which highlighted miR-130b targets in pathways such as autophagy, FoxO signaling pathway, and MAPK signaling pathway—pathways centrally implicated in UC pathophysiology. Although plasma expression of miR-130b-3p/5p trended higher rather than lower in UC patients in our external cohort, this discrepancy is biologically plausible: genetically protective miRNAs can be secondarily up-regulated as a compensatory response in active disease, and cross-sectional plasma levels may not faithfully recapitulate lifetime genetically determined exposure. In contrast to the miR-130b cluster, several miRNAs showed risk-increasing effects, with variable support across datasets and instruments. In the FinnGen discovery analysis, miR-101-3p and miR-1908-5p emerged as strong risk factors, while in the GCST replication, miR-27a-3p and let-7a-5p displayed large risk-enhancing ORs. In the trans-eQTL–based validation using the Nikpay dataset, miR-4326 showed a significant risk effect in FinnGen. These signals, although not uniformly replicated across all analyses, are biologically coherent with known roles of these miRNAs in epithelial homeostasis, cytokine signaling, and autoimmunity. miR-101-3p is widely recognized as a tumor-suppressor miRNA in several cancers, including colorectal cancer (CRC), where it directly targets the PGE₂ receptor EP4 and represses PI3K/AKT and other oncogenic pathways [ 20 ]. In colon cancer specimens, down-regulation of miR-101 is associated with increased EP4 expression, enhanced proliferation and migration, and poor prognosis [ 20 ]. Moreover, miR-101 is part of a diagnostic panel (miR-19a, miR-21, miR-31, miR-146a, miR-375) that differentiates UC from Crohn’s disease across colon tissue, blood, and saliva, implicating it in IBD pathogenesis [ 21 ]. Our MR finding that genetically higher plasma miR-101-3p levels increase UC risk is therefore counterintuitive relative to its tumor-suppressive role, but may reflect context-specific, isoform-specific (-3p vs -5p) or compartment-specific effects. It is conceivable that systemic elevation of miR-101-3p modulates immune cell or stromal targets differently from local epithelial miR-101 in the colon, or that the MR signal partially captures linkage with nearby inflammatory regulators. miR-1908-5p represents another notable risk-associated miRNA in our analysis. Integrative eQTL work has previously shown that Crohn’s disease and rheumatoid arthritis risk SNPs can significantly alter miR-1908-5p expression in lymphoblastoid cell lines, with enrichment of predicted targets among autoimmune GWAS loci, suggesting a potential role in systemic autoimmunity [ 22 ]. A recent review summarized its broad involvement in cancer, fibrosis, obesity, and rheumatoid arthritis, often converging on NF-κB activation and TGF-β1/Smad signaling [ 23 ]. Our MR results, together with this prior evidence, support a model in which miR-1908-5p promotes pro-inflammatory and profibrotic circuits relevant to mucosal immunity and UC. The large risk effects for miR-27a-3p and let-7a-5p in the GCST cohort are in line with their known roles in inflammation and intestinal pathology. miR-27a-3p is consistently up-regulated in CRC tissue and cell lines and promotes proliferation and survival partly by targeting BTG1 and modulating ERK/MEK signaling [ 24 ]. Experimental work in osteoblasts also shows that miR-27a-3p represses GLP1R and AMPK signaling, reducing autophagy and differentiation while increasing inflammatory cytokine production, linking it to metabolic-inflammatory regulation [ 25 ]. Our colocalization analysis indicated the strongest evidence for a shared causal variant at the miR-27a-3p locus in the GCST dataset (PP.H4 ≈ 0.61), supporting a direct genetic mechanism at 19p13.11. let-7a-5p belongs to the broadly expressed let-7 family, which generally acts as a tumor suppressor and differentiation-promoting factor in the intestine. In mouse models, global intestinal loss of let-7 miRNAs leads to adenocarcinoma formation and stem-like transcriptional programs, accompanied by up-regulation of Hmga1/2 and other oncogenic targets [ 26 ]. However, several studies have demonstrated anti-inflammatory roles for let-7 family members: let-7a ameliorates ConA-induced hepatitis by inhibiting IL-6–dependent Th17 differentiation, and let-7a-5p reduces TNF-α, IL-1β and IL-6 expression and Ras/MAPK activation in chronic rhinosinusitis fibroblasts [ 27 ]. In pediatric UC, tissue levels of let-7 are reduced along with miR-124, contributing to enhanced STAT3 phosphorylation [ 28 ]. Thus, our MR evidence that genetically increased plasma let-7a-5p heightens UC risk may again reflect systemic vs mucosal compartmentalization, family-member specificity, or complex feedback loops: higher circulating let-7a-5p could be a marker of a broader regulatory state that, while compensatory in some tissues, correlates with greater UC susceptibility overall. miR-103a-3p and let-7d-5p, which showed nominal associations in our discovery analyses and/or colocalization, also have compelling links to intestinal inflammation and fibrosis. In Crohn’s disease, exosomal miR-103a-3p derived from creeping fat adipose-derived stem cells promotes intestinal fibrosis by targeting TGFBR3, activating Smad2/3, and driving fibroblast activation; its expression in diseased intestine correlates with the severity of fibrosis [ 29 ]. Beyond the gut, miR-103a-3p promotes fibrosis and inflammation in non-alcoholic fatty liver disease by repressing HBP1 [ 30 ]. let-7d-5p, in a neonatal necrotizing enterocolitis (NEC) model, is down-regulated in inflamed intestinal tissue, and its over-expression suppresses TLR4/NF-κB signaling and pro-inflammatory cytokines via targeting LGALS3, thereby reducing epithelial apoptosis and mucosal injury [ 31 ]. These data suggest that miR-103a-3p may be broadly pro-fibrotic and pro-inflammatory in the intestine, whereas let-7d-5p appears anti-inflammatory. The direction and strength of their MR associations with UC in our study were more modest, but their inclusion among suggestive candidates and their enrichment in UC-relevant pathways warrant further follow-up. miR-4326 and miR-671-3p, although less well studied in IBD, both converge on signaling axes that intersect with mucosal homeostasis. miR-4326 is up-regulated in lung cancer and promotes proliferation by targeting APC2, a negative regulator of the Wnt/β-catenin pathway [ 32 ]. Given that aberrant Wnt signaling is pivotal in intestinal epithelial renewal and colorectal carcinogenesis, genetically elevated miR-4326 could theoretically perturb epithelial barrier dynamics and regeneration, making its risk association with UC in our trans-eQTL MR (FinnGen) mechanistically plausible. miR-671-3p has been described both as an oncogenic factor in glioma and as a regulator of inflammation and matrix metabolism in osteoarthritis, where it targets TRAF3 and modulates chondrocyte apoptosis and pro-inflammatory cytokine production [ 33 ]. TRAF3 is a key adaptor in TNF receptor and Toll-like receptor pathways, suggesting that miR-671-3p could influence innate immune activation in the gut as well. Among the discovery-phase miRNAs, miR-6891-3p and miR-6810-3p stand out as largely unexplored in the IBD literature, yet their genomic context and predicted functions are suggestive. miR-6891 is encoded within intron 4 of HLA-B in the MHC class I region; prior functional work has shown that its 5p strand (miR-6891-5p) regulates IgA heavy chain transcripts (IGHA1/2), and that elevated miR-6891-5p expression in lymphoblastoid cells from IgA-deficient patients suppresses IgA production [ 34 ]. Secretory IgA is central to mucosal barrier immunity, and selective IgA deficiency is epidemiologically linked to IBD and celiac disease. Although our MR implicates the 3p strand, the shared precursor locus suggests that MIR6891-derived miRNAs could mechanistically couple HLA variation, IgA homeostasis, and UC risk. miR-6810-3p, while not previously connected to intestinal inflammation, has been detected in human plasma and is highly enriched in brain tissue and appendix according to transcriptomic atlases. A recent population-based profiling of plasma miRNAs and persistent organic pollutants suggested that miR-6810-3p targets autophagy-related pathways [ 35 ]. Given the central role of autophagy genes (e.g., ATG16L1, IRGM) in Crohn’s disease and their contributions to Paneth cell function and bacterial handling, a causal link between genetically determined miR-6810-3p levels and UC could point toward an underappreciated autophagy-modulating miRNA in mucosal immunity. At present, however, the evidence is preliminary and driven primarily by MR statistics and in-silico functional prediction rather than direct experimental data. Our Bayesian colocalization analysis helped distinguish likely shared causal variants from associations potentially driven by linkage disequilibrium. The strongest support for colocalization (PP.H4 > 0.6) was seen for miR-27a-3p in the GCST dataset and for miR-1908-5p in FinnGen, whereas the miR-130b locus showed relatively low PP.H4 but also low PP.H3, suggesting either multiple causal variants or limited resolution of current GWAS summary statistics. This pattern is not unexpected in regions with complex LD structure and modest effect sizes; it argues for cautious interpretation and motivates fine-mapping using denser genotyping or whole-genome sequencing data in UC cohorts. Although the reverse-direction MR results were not the primary focus of this work, the absence of strong and consistent UC→miRNA effects for most candidates (as suggested by Steiger directionality and the lack of robust reverse IVW estimates) supports our main inference that the identified miRNAs are more likely upstream determinants rather than downstream consequences of genetic UC liability. Nonetheless, disease activity, treatment, and environmental exposures can modulate miRNA levels independently of germline predisposition, which likely explains why our external plasma validation (GSE122618) did not show statistically significant differences despite suggestive trends for miR-130b-3p/5p, miR-671-3p, and miR-27a-3p. The apparent disconnect between MR and cross-sectional expression findings underscores a broader point: MR estimates reflect the effect of lifelong genetic perturbation of miRNA levels, averaged across tissues and disease stages, whereas transcriptomic snapshots capture short-term, tissue- and context-specific regulation. For example, protective miRNAs may be up-regulated as a counter-regulatory response in established disease, while genetically harmful miRNAs could be partially suppressed by therapy or feedback inhibition. In addition, many of our instruments derive from whole-blood or plasma eQTLs; their relationship to miRNA abundance in the colonic epithelium, lamina propria lymphocytes and mesenchymal cells is imperfect, and some genetically determined variation may be cell-type specific. Key strengths of this study include the two-stage design with independent discovery and validation cohorts, the use of multiple instrument sets (cis and trans eQTLs across Nikpay and FHS), careful LD clumping and weak-instrument filtering, extensive sensitivity analyses (Steiger tests, pleiotropy screening via LDtrait, reverse MR), and Bayesian colocalization. The integration of experimentally validated miRNA–target data and pathway enrichment (highlighting IBD, Th17, NF-κB and cytokine signaling pathways) provides mechanistic plausibility that is often lacking in purely statistical MR screens. Our work also extends miRNA-focused MR beyond cardiometabolic traits to complex immune-mediated disease. Several limitations should be acknowledged. First, all GWAS and eQTL data were derived from individuals of European ancestry, limiting generalizability to other populations where both miRNA eQTL architecture and UC genetics may differ. Second, some miRNAs were instrumented by a small number of SNPs, and while all F-statistics exceeded 10, power to detect modest causal effects and to robustly assess pleiotropy remained limited. Third, the reliance on blood/plasma eQTLs may not capture tissue-specific miRNA regulation in the gut; future work using intestinal eQTL datasets, once available at scale, will be critical. Fourth, colocalization results for several key miRNAs (e.g., miR-130b-3p/5p) were inconclusive, reflecting LD complexity and limited SNP density near short miRNA genes. Finally, our external expression validation cohort was relatively small and heterogeneous with respect to disease activity, which reduces power to detect modest differences in circulating miRNA levels. Conclusion In conclusion, this study provides compelling genetic evidence that specific circulating miRNAs causally influence UC susceptibility. The identification of protective (miR-130b family) and risk-increasing (miR-101-3p, miR-1908-5p, miR-27a-3p, let-7a-5p) miRNAs advances our understanding of UC pathogenesis and highlights potential biomarkers and therapeutic targets. These findings support the emerging paradigm of miRNAs as key regulators in inflammatory bowel disease and provide a foundation for developing precision medicine approaches in UC management. Declarations Ethics Approval and Consent to Participate This study is a secondary analysis of publicly available summary-level data. The original studies (FinnGen, Framingham Heart Study, and other GWAS cohorts) received ethical approval from their respective institutional review boards, and informed consent was obtained from all participants. No separate ethical approval or informed consent was required for the present study, as it utilized anonymized, aggregate data. Consent for Publication Not applicable. Availability of Data and Materials The datasets presented in this study can be found in online repositories. The summary statistics for Ulcerative Colitis were obtained from the FinnGen consortium (Release 12, https://www.finngen.fi/en) and the EBI GWAS Catalog (Accession ID: GCST90473823). The miRNA eQTL data were sourced from the publications by Nikpay et al. and the Framingham Heart Study. The external validation transcriptomic dataset is available in the Gene Expression Omnibus (GEO) under accession number GSE122618. All other data generated or analyzed during this study are included in this published article and its supplementary information files. Competing Interests The authors declare that they have no competing interests. Funding Not applicable Authors' Contributions Conceptualization and Methodology: W. Chu; Investigation and Formal Analysis: J. He and H. Jin; Resources: W. Chu; Writing – Original Draft: J. He; Writing – Review & Editing: H. Jin and W. Chu. All authors have read and agreed to the published version of the manuscript. Acknowledgments We gratefully acknowledge the participants and investigators of the FinnGen study, the Framingham Heart Study, and the authors of the original GWAS and eQTL studies for making their summary statistics publicly available. We also thank the contributors to the GEO database for providing the transcriptomic data used for validation. References Ng SC et al (2017) Worldwide incidence and prevalence of inflammatory bowel disease in the 21st century: a systematic review of population-based studies. Lancet 390(10114):2769–2778 Lewis JD et al (2023) Incidence, Prevalence, and Racial and Ethnic Distribution of Inflammatory Bowel Disease in the United States. Gastroenterology 165(5):1197–1205e2 Xu F et al (2018) Health-Risk Behaviors and Chronic Conditions Among Adults with Inflammatory Bowel Disease - United States, 2015 and 2016. MMWR Morb Mortal Wkly Rep 67(6):190–195 Dahlhamer JM et al (2016) Prevalence of Inflammatory Bowel Disease Among Adults Aged ≥ 18 Years - United States, 2015. MMWR Morb Mortal Wkly Rep 65(42):1166–1169 Vermeire S et al (2025) Obefazimod in patients with moderate-to-severely active ulcerative colitis: efficacy and safety analysis from the 96-week open-label maintenance phase 2b study. J Crohns Colitis, 19(5) Sokal-Dembowska A et al (2025) The Role of microRNAs in Inflammatory Bowel Disease. Int J Mol Sci, 26(10) Wu F et al (2008) MicroRNAs are differentially expressed in ulcerative colitis and alter expression of macrophage inflammatory peptide-2 alpha. Gastroenterology 135(5):1624–1635e24 Takagi T et al (2010) Increased expression of microRNA in the inflamed colonic mucosa of patients with active ulcerative colitis. J Gastroenterol Hepatol 25(Suppl 1):S129–S133 Koukos G et al (2015) A microRNA signature in pediatric ulcerative colitis: deregulation of the miR-4284/CXCL5 pathway in the intestinal epithelium. Inflamm Bowel Dis 21(5):996–1005 Paraskevi A et al (2012) Circulating MicroRNA in inflammatory bowel disease. J Crohns Colitis 6(9):900–904 Duttagupta R et al (2012) Genome-wide maps of circulating miRNA biomarkers for ulcerative colitis. PLoS ONE 7(2):e31241 Schönauen K et al (2018) Circulating and Fecal microRNAs as Biomarkers for Inflammatory Bowel Diseases. Inflamm Bowel Dis 24(7):1547–1557 Davey Smith G, Hemani G (2014) Mendelian randomization: genetic anchors for causal inference in epidemiological studies. Hum Mol Genet 23(R1):R89–98 Liu B et al (2022) Two-Sample Mendelian Randomization Analysis Investigates Causal Associations Between Gut Microbial Genera and Inflammatory Bowel Disease, and Specificity Causal Associations in Ulcerative Colitis or Crohn's Disease. Front Immunol 13:921546 Nikpay M et al (2019) Genome-wide identification of circulating-miRNA expression quantitative trait loci reveals the role of several miRNAs in the regulation of cardiometabolic phenotypes. Cardiovasc Res 115(11):1629–1645 Huan T et al (2015) Genome-wide identification of microRNA expression quantitative trait loci. Nat Commun 6:6601 Zhang M et al (2016) MiR-130b promotes obesity associated adipose tissue inflammation and insulin resistance in diabetes mice through alleviating M2 macrophage polarization via repression of PPAR-γ. Immunol Lett 180:1–8 Bai X et al (2016) MicroRNA-130b improves renal tubulointerstitial fibrosis via repression of Snail-induced epithelial-mesenchymal transition in diabetic nephropathy. Sci Rep 6:20475 Han X et al (2024) Mechanism of miR-130b-3p in relieving airway inflammation in asthma through HMGB1-TLR4-DRP1 axis. Cell Mol Life Sci 82(1):9 Chandramouli A et al (2012) MicroRNA-101 (miR-101) post-transcriptionally regulates the expression of EP4 receptor in colon cancers. Cancer Biol Ther 13(3):175–183 Schaefer JS et al (2015) MicroRNA signatures differentiate Crohn's disease from ulcerative colitis. BMC Immunol 16:5 Wohlers I, Bertram L, Lill CM (2018) Evidence for a potential role of miR-1908-5p and miR-3614-5p in autoimmune disease risk using integrative bioinformatics. J Autoimmun 94:83–89 Ghafouri-Fard S et al (2022) miR-1908: a microRNA with diverse functions in cancers and non-malignant conditions. Cancer Cell Int 22(1):281 Su C et al (2019) miR-27a-3p regulates proliferation and apoptosis of colon cancer cells by potentially targeting BTG1. Oncol Lett 18(3):2825–2834 Zeng Z et al (2021) MiR-27a-3p Targets GLP1R to Regulate Differentiation, Autophagy, and Release of Inflammatory Factors in Pre-Osteoblasts via the AMPK Signaling Pathway. Front Genet 12:783352 Madison BB et al (2015) Let-7 Represses Carcinogenesis and a Stem Cell Phenotype in the Intestine via Regulation of Hmga2. PLoS Genet 11(8):e1005408 Zhang Y et al (2013) MicroRNA let-7a ameliorates con A-induced hepatitis by inhibiting IL-6-dependent Th17 cell differentiation. J Clin Immunol 33(3):630–639 Koukos G et al (2013) MicroRNA-124 regulates STAT3 expression and is down-regulated in colon tissues of pediatric patients with ulcerative colitis. Gastroenterology, 145(4): p. 842 – 52.e2. Qian W et al (2023) Exosomal miR-103a-3p from Crohn's Creeping Fat-Derived Adipose-Derived Stem Cells Contributes to Intestinal Fibrosis by Targeting TGFBR3 and Activating Fibroblasts. J Crohns Colitis 17(8):1291–1308 Chu K, Gu J (2022) microRNA-103a-3p promotes inflammation and fibrosis in nonalcoholic fatty liver disease by targeting HBP1. Immunopharmacol Immunotoxicol 44(6):993–1003 Sun L et al (2020) Let-7d-5p suppresses inflammatory response in neonatal rats with necrotizing enterocolitis via LGALS3-mediated TLR4/NF-κB signaling pathway. Am J Physiol Cell Physiol 319(6):C967–C979 Xu G et al (2018) miR-4326 promotes lung cancer cell proliferation through targeting tumor suppressor APC2. Mol Cell Biochem 443(1–2):151–157 Lu G et al (2018) MicroRNA-671-3p promotes proliferation and migration of glioma cells via targeting CKAP4. Onco Targets Ther 11:6217–6226 Chitnis N et al (2017) An Expanded Role for HLA Genes: HLA-B Encodes a microRNA that Regulates IgA and Other Immune Response Transcripts. Front Immunol 8:583 Qu J et al (2025) Persistent organic pollutants and plasma microRNAs: A community-based profiling analysis. Environ Int 197:109328 Supplementary Files FigureS1.tif Figure S1. External validation of candidate miRNA expression in plasma samples from UC patients and healthy controls. TableS1.xlsx Table S1. Results of the Mendelian Randomization Analysis for the Association between Plasma miRNAs and Ulcerative Colitis in the Discovery Set TableS2.xlsx Table S2. Results of the Mendelian Randomization Analysis for the Association between Plasma miRNAs and Ulcerative Colitis in the Validation Set TableS3.xlsx Table S3. Results of the Reverse MR Analysis TableS4.xlsx Table S4. Results of the Colocalization Analysis TableS5.xlsx Table S5. Identification of Target Genes for Causal miRNAs TableS6.xlsx Table S6. Differential Expression of Plasma miRNAs in the GSE122618 Cohort Cite Share Download PDF Status: Under Review Version 1 posted Reviewers invited by journal 03 Feb, 2026 Editor assigned by journal 28 Jan, 2026 First submitted to journal 27 Jan, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8634046","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":584980069,"identity":"9831e6a2-f6ae-40dc-b4af-8638fb2200e4","order_by":0,"name":"Jin He","email":"","orcid":"","institution":"Hangzhou Hospital of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Jin","middleName":"","lastName":"He","suffix":""},{"id":584980070,"identity":"b43a05ef-06c0-449d-aed0-3e84b8cd8111","order_by":1,"name":"Haimin Jin","email":"","orcid":"","institution":"Hangzhou Hospital of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Haimin","middleName":"","lastName":"Jin","suffix":""},{"id":584980071,"identity":"d2f75ed9-8481-4380-b11d-49a22307ef8a","order_by":2,"name":"Weijian Chu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA5klEQVRIie3RMWvCQBTA8VcCr8szWS9Ekq9wItwk9KvcIdxkwUkyOChKHNRdv0VHxwThpmtnx5QOXevmpuhaSeLmcL/5/eG9OwDHeUKY/Oa5TJEw+SlKmY7rEx8GqvyzfjsA3eelNfVJDLLb2WZxL5wMRPg99xosBrmOWpkgnhudqglCsFjK6uRlaiL60sSLmTmoXRuY/fyoTjzQEY0M8T3qg7IInL3XJAgiIjwTNySGKvMaJATdziZDClckoFnC6PrISAHDPpPWUO0tyeb19pVvyLzieErHcbBYVyf/0GPjjuM4zl0XFDpGbLop5m4AAAAASUVORK5CYII=","orcid":"","institution":"Hangzhou Hospital of Traditional Chinese Medicine","correspondingAuthor":true,"prefix":"","firstName":"Weijian","middleName":"","lastName":"Chu","suffix":""}],"badges":[],"createdAt":"2026-01-19 01:27:45","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8634046/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8634046/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":101921851,"identity":"943bb955-38c8-4156-970a-b1807ac19b3c","added_by":"auto","created_at":"2026-02-05 04:55:45","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":651829,"visible":true,"origin":"","legend":"\u003cp\u003eThe workflow diagram of the study.\u003c/p\u003e","description":"","filename":"Figure1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8634046/v1/ed703118324170a843da0192.jpeg"},{"id":101921842,"identity":"338df97e-528b-4d80-bf54-6fa8b02058b6","added_by":"auto","created_at":"2026-02-05 04:55:44","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":646371,"visible":true,"origin":"","legend":"\u003cp\u003eForest plot of the causal effects of plasma miRNAs on ulcerative colitis (UC) risk in the discovery phase.\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-8634046/v1/19857625c8582054c218e554.png"},{"id":101921847,"identity":"4fa0ff04-e598-49f3-8038-655c84440c48","added_by":"auto","created_at":"2026-02-05 04:55:44","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":552017,"visible":true,"origin":"","legend":"\u003cp\u003eValidation of causal associations using alternative genetic instruments.\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-8634046/v1/fa7ac6d81660b36259b20652.png"},{"id":101921843,"identity":"e50afed5-1140-41d2-bc97-9eba196363f1","added_by":"auto","created_at":"2026-02-05 04:55:44","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":41632,"visible":true,"origin":"","legend":"\u003cp\u003eBayesian colocalization analysis of miRNA and UC susceptibility loci.\u003c/p\u003e","description":"","filename":"Binder13.png","url":"https://assets-eu.researchsquare.com/files/rs-8634046/v1/6b569e4d70bce93aaf210a14.png"},{"id":101943342,"identity":"e6a8c880-7b06-47c9-967a-87d4459adf7c","added_by":"auto","created_at":"2026-02-05 09:41:41","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":84337,"visible":true,"origin":"","legend":"\u003cp\u003eHierarchical clustering of Gene Ontology (GO) biological processes enriched for miR-130b targets.\u003c/p\u003e","description":"","filename":"Binder14.png","url":"https://assets-eu.researchsquare.com/files/rs-8634046/v1/1984702657ca8b434d014740.png"},{"id":101921852,"identity":"6302e4c4-f248-4029-a14b-4e0deca60554","added_by":"auto","created_at":"2026-02-05 04:55:45","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":48927,"visible":true,"origin":"","legend":"\u003cp\u003eKEGG pathway enrichment analysis of predicted target genes for miR-130b-3p and miR-130b-5p.\u003c/p\u003e","description":"","filename":"Binder15.png","url":"https://assets-eu.researchsquare.com/files/rs-8634046/v1/ce558eba77fb6f1a57eedff7.png"},{"id":102750173,"identity":"426815cc-0482-4ed8-b4e4-8e4d24f05cc2","added_by":"auto","created_at":"2026-02-16 09:17:57","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2780268,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8634046/v1/4c8bd40f-13e5-4b80-ad3e-2ff81a358412.pdf"},{"id":102745403,"identity":"eb53498d-bb8d-45de-8354-83bff95b9666","added_by":"auto","created_at":"2026-02-16 08:48:48","extension":"tif","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":861244,"visible":true,"origin":"","legend":"\u003cp\u003eFigure S1. External validation of candidate miRNA expression in plasma samples from UC patients and healthy controls.\u003c/p\u003e","description":"","filename":"FigureS1.tif","url":"https://assets-eu.researchsquare.com/files/rs-8634046/v1/79e0e29bb340e2ebed876fd8.tif"},{"id":101921845,"identity":"68dbf835-cd34-4a80-bd2a-2c7492dbb08a","added_by":"auto","created_at":"2026-02-05 04:55:44","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":30633,"visible":true,"origin":"","legend":"\u003cp\u003eTable S1. Results of the Mendelian Randomization Analysis for the Association between Plasma miRNAs and Ulcerative Colitis in the Discovery Set\u003c/p\u003e","description":"","filename":"TableS1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8634046/v1/84b21d8adb2f1a424d74a5df.xlsx"},{"id":101921844,"identity":"cd2be1f2-1c70-4d8c-ac44-94c759150126","added_by":"auto","created_at":"2026-02-05 04:55:44","extension":"xlsx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":14363,"visible":true,"origin":"","legend":"\u003cp\u003eTable S2. Results of the Mendelian Randomization Analysis for the Association between Plasma miRNAs and Ulcerative Colitis in the Validation Set\u003c/p\u003e","description":"","filename":"TableS2.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8634046/v1/544ee6b1a8c9052725971a9f.xlsx"},{"id":101921849,"identity":"1cf335f8-bba2-4e66-830b-da776e1d2cad","added_by":"auto","created_at":"2026-02-05 04:55:44","extension":"xlsx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":13960,"visible":true,"origin":"","legend":"\u003cp\u003eTable S3. Results of the Reverse MR Analysis\u003c/p\u003e","description":"","filename":"TableS3.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8634046/v1/39376f15c6cd451a26d9c5c2.xlsx"},{"id":101921854,"identity":"c71ce0e8-2306-4df2-a796-9d2bda33c3ee","added_by":"auto","created_at":"2026-02-05 04:55:47","extension":"xlsx","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":11705,"visible":true,"origin":"","legend":"\u003cp\u003eTable S4. Results of the Colocalization Analysis\u003c/p\u003e","description":"","filename":"TableS4.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8634046/v1/cb98c1d472f0197233eb8263.xlsx"},{"id":101943439,"identity":"1d42a03c-4979-443b-8a84-60a38d96d8bc","added_by":"auto","created_at":"2026-02-05 09:41:57","extension":"xlsx","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":497266,"visible":true,"origin":"","legend":"\u003cp\u003eTable S5. Identification of Target Genes for Causal miRNAs\u003c/p\u003e","description":"","filename":"TableS5.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8634046/v1/7f16ff3c2b4d72d38233bb09.xlsx"},{"id":101921848,"identity":"60c864b0-82d8-4342-8400-af4b24eb1e78","added_by":"auto","created_at":"2026-02-05 04:55:44","extension":"xlsx","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":10587,"visible":true,"origin":"","legend":"\u003cp\u003eTable S6. Differential Expression of Plasma miRNAs in the GSE122618 Cohort\u003c/p\u003e","description":"","filename":"TableS6.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8634046/v1/a2d10b8f0620c71529f3901d.xlsx"}],"financialInterests":"","formattedTitle":"Unraveling the Causal Association Between microRNAs and Ulcerative Colitis: A Multi-Stage Mendelian Randomization Study with External Validation","fulltext":[{"header":"Introduction","content":"\u003cp\u003eUlcerative colitis (UC) is a chronic, relapsing inflammatory bowel disease (IBD) characterized by continuous mucosal inflammation of the colon and rectum, leading to bloody diarrhea, abdominal pain, and substantial impairment in quality of life. Recent estimates indicate that the prevalence of IBD has exceeded 0.3% in many Western countries, with UC prevalence reaching up to 286\u0026ndash;505 per 100,000 individuals in North America and Northern Europe [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. In the United States alone, 2.4\u0026ndash;3.1\u0026nbsp;million adults are living with IBD, imposing a considerable health‑care and economic burden [\u003cspan additionalcitationids=\"CR3\" citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Despite advances in biologic and small‑molecule therapies, a significant proportion of patients exhibit primary non‑response, loss of response, or treatment‑limiting adverse events, and many ultimately require colectomy [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. These limitations highlight the need to better understand the molecular mechanisms that drive UC and to identify novel biomarkers and therapeutic targets.\u003c/p\u003e \u003cp\u003eMicroRNAs (miRNAs) are ~\u0026thinsp;22-nucleotide non-coding RNAs that repress gene expression post-transcriptionally by binding to complementary sequences in target mRNAs. They are now recognized as key regulators of immune cell differentiation, cytokine production, epithelial barrier function, and cell death pathways [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Multiple profiling studies have demonstrated widespread miRNA dysregulation in colonic mucosa from patients with active UC. Early microarray work identified altered expression of at least 11 miRNAs in UC tissue, including downregulation of miR-192 and upregulation of miR-155; miR-192 was shown to directly regulate the chemokine MIP-2α in colonic epithelial cells [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Subsequent studies confirmed increased mucosal levels of miR-21, miR-31, miR-155, miR-125b and others, which modulate IL-13 signaling, NF-κB activation, mucus production, and goblet cell differentiation, linking specific miRNAs to key elements of UC pathophysiology [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Pediatric UC cohorts have further revealed distinct miRNA signatures, such as downregulation of epithelial miR-4284 with reciprocal upregulation of its target CXCL5, underscoring the relevance of miRNA-mediated networks across age groups [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. Beyond the intestinal mucosa, circulating and fecal miRNAs have emerged as promising non-invasive biomarkers for IBD. Several case\u0026ndash;control studies reported increased levels of serum or plasma miR-16, miR-21, miR-155, miR-199a-5p, miR-223 and other miRNAs in patients with UC compared with healthy controls, some of which correlate with clinical or endoscopic activity[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Genome-wide profiling of blood cell fractions and microvesicles identified multi-miRNA circulating signatures that can distinguish UC from healthy individuals with high accuracy [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Additional work has shown that circulating miRNA panels can differentiate Crohn\u0026rsquo;s colitis from UC, discriminate active disease from remission, and track treatment responses [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. Collectively, these observations support the utility of circulating miRNAs as diagnostic and monitoring tools in UC. However, because these studies are observational, they cannot determine whether altered miRNA levels are causal drivers of disease onset, reflect downstream inflammation, or represent epiphenomena related to treatment and comorbidities.\u003c/p\u003e \u003cp\u003eMendelian randomization (MR) offers a powerful epidemiologic framework to address causality by using germline genetic variants as instrumental variables (IVs) for modifiable exposures. Under core assumptions, genetic proxies for an exposure (e.g. biomarker levels) are randomly allocated at conception and are largely independent of confounders and reverse causation, allowing estimation of causal effects on disease outcomes using observational data [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. With the advent of large GWAS, two-sample MR using summary statistics has been widely applied to UC and IBD to interrogate the causal roles of gut microbiota, circulating inflammatory proteins, blood metabolites, glucose, immune cell subsets, and druggable proteins. These studies have identified specific microbial genera, cytokines (e.g. IL-10 receptor subunits, chemokines), and metabolic pathways that appear to influence UC risk and may mediate immune\u0026ndash;metabolic crosstalk [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. Nevertheless, MR investigations focusing on non-coding RNAs in UC remain scarce, and the contribution of circulating miRNAs to UC susceptibility has not been systematically evaluated.\u003c/p\u003e \u003cp\u003eIn this study, we employed a comprehensive, multi-stage study design to systematically investigate the causal effect of plasma miRNAs on the risk of UC. Unlike previous studies, we integrated a two-sample MR approach with rigorous sensitivity analyses, including Bayesian colocalization, to distinguish true causality from linkage disequilibrium. Furthermore, to elucidate the biological plausibility, we mapped the downstream target genes of causal miRNAs to immune-related pathways. Uniquely, we triangulated our genetic findings with external transcriptomic validation using independent expression data from the Gene Expression Omnibus (GEO). By combining genetic causal inference with functional annotation and real-world expression profiling, this study aims to identify robust, causally relevant miRNA biomarkers for UC, offering new perspectives on disease pathogenesis and potential therapeutic interventions.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy Design and Data Sources\u003c/h2\u003e \u003cp\u003eWe employed a two-sample MR design to systematically investigate the causal effect of circulating miRNAs on the risk of UC. To ensure the robustness and replicability of our findings, a two-stage strategy comprising a discovery phase and a validation phase was implemented. Genetic instruments for miRNAs were obtained from two independent eQTL datasets based on the hg38 genome assembly. The Nikpay et al. dataset (n\u0026thinsp;=\u0026thinsp;710) served as the primary source [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e], utilizing cis-eQTLs for the discovery phase and trans-eQTLs for the validation phase, while cis-eQTLs from the Framingham Heart Study (FHS, n\u0026thinsp;=\u0026thinsp;5239) were employed as an additional validation source [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Cis-acting IVs were strictly defined as SNPs located within 250 kb of the corresponding miRNA gene region. For the outcome phenotypes, summary-level data for UC were sourced from two large-scale cohorts of European ancestry to minimize sample overlap. The primary analysis utilized data from the FinnGen consortium (Release 12), comprising 7,220 cases and 492,160 controls defined by strict clinical endpoints. To corroborate the findings, a replication dataset was accessed from the EBI GWAS Catalog (Accession ID: GCST90473823), which included 6,158 cases and 452,282 controls.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eInstrument Selection and Statistical Analysis\u003c/h3\u003e\n\u003cp\u003eThe MR analysis was conducted using the TwoSampleMR package in R software. We performed a preliminary screening of IVs using a significance threshold of P\u0026thinsp;\u0026lt;\u0026thinsp;5e\u0026thinsp;\u0026minus;\u0026thinsp;6 to capture potential causal variants. To ensure independence among the selected instruments, SNPs were clumped using a strict linkage disequilibrium (LD) threshold of r2\u0026thinsp;\u0026lt;\u0026thinsp;0.001 within a 10,000 kb window. Palindromic SNPs with intermediate allele frequencies were inferred based on allele frequency information. To avoid weak instrument bias, the strength of each genetic instrument was assessed using the F-statistic (F\u0026thinsp;=\u0026thinsp;β2/SE2), and any SNPs with an F-statistic\u0026thinsp;\u0026lt;\u0026thinsp;10 were excluded from the analysis. The Wald ratio method was applied for miRNAs instrumented by a single SNP, while the inverse variance weighted (IVW) method was employed for exposures with multiple valid instruments. In the discovery phase, miRNA-UC associations with a P-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 were identified as potential causal candidates and were subsequently advanced to the validation phase using independent IVs or outcome datasets.\u003c/p\u003e\n\u003ch3\u003eSensitivity Analyses and Pleiotropy Assessment\u003c/h3\u003e\n\u003cp\u003eTo verify the validity of the causal assumptions, we performed rigorous sensitivity analyses. The Steiger directionality test was conducted to confirm the direction of causality, ensuring that the genetic variants explained significantly more variance in the miRNA expression than in the UC risk. Furthermore, to address potential horizontal pleiotropy, we utilized the LDlinkR package to query the LDtrait tool. We systematically screened the identified causal SNPs against the GWAS Catalog for associations with known confounders of UC, including smoking behaviors, body mass index (BMI), other autoimmune diseases (e.g., Crohn\u0026rsquo;s disease, psoriasis), and antibiotic usage. SNPs showing significant associations with these confounding traits were flagged to exclude the possibility that the observed estimates were driven by alternative pathways independent of the miRNA exposure.\u003c/p\u003e\n\u003ch3\u003eReverse Mendelian Randomization Analysis\u003c/h3\u003e\n\u003cp\u003eTo investigate the possibility of reverse causality\u0026mdash;specifically, whether genetic predisposition to UC causally influences the expression levels of the identified miRNAs\u0026mdash;we performed a reverse MR analysis. In this analysis, UC was treated as the exposure and the candidate miRNAs identified in the forward analysis served as the outcomes. Genetic instruments for UC were extracted from the same two independent GWAS datasets used in the forward analysis (FinnGen and GCST90473823). Significant SNPs associated with UC were selected using a genome-wide significance threshold of P\u0026thinsp;\u0026lt;\u0026thinsp;5e\u0026thinsp;\u0026minus;\u0026thinsp;8. To ensure independence, these instruments were clumped using a LD threshold of r2\u0026thinsp;\u0026lt;\u0026thinsp;0.001 within a 10,000 kb window, based on the 1000 Genomes Project European reference panel. Summary statistics for the outcome miRNAs were retrieved from the full genome-wide association statistics of the Nikpay et al. study. The causal effects were estimated using the IVW, Weighted Median, and Wald Ratio methods, consistent with the forward analysis. A P-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 in the reverse direction was considered indicative of potential reverse causality or bidirectional effects.\u003c/p\u003e\n\u003ch3\u003eBayesian Colocalization Analysis\u003c/h3\u003e\n\u003cp\u003eTo distinguish whether the identified associations between miRNAs and UC were driven by a shared causal variant (pleiotropy) or by distinct variants in linkage disequilibrium, we performed Bayesian colocalization analysis using the coloc R package. We focused on the genomic regions surrounding the lead SNPs identified in the MR analysis. For each candidate miRNA, we defined a 400 kb window (\u0026plusmn;\u0026thinsp;200 kb) centered on the top cis-eQTL. Since the miRNA summary statistics were based on the hg19 genome assembly while the UC data utilized hg38, we performed a genomic coordinate conversion using the LiftOver tool with the hg19-to-hg38 chain file to ensure precise alignment. We extracted the summary statistics (beta coefficients, standard errors, and minor allele frequencies) for all SNPs within these regions from both the miRNA eQTL dataset (Nikpay et al.) and the UC GWAS datasets (FinnGen and GCST90473823). The coloc.abf function was employed to calculate the posterior probabilities (PP) for five distinct hypotheses: H0 (no association with either trait), H1 (association with miRNA only), H2 (association with UC only), H3 (association with both traits but distinct causal variants), and H4 (association with both traits sharing a single causal variant). A high posterior probability for H4 (PP.H4\u0026thinsp;\u0026gt;\u0026thinsp;0.8) was considered strong evidence of colocalization, supporting a shared genetic mechanism underlying both miRNA expression and UC risk.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eTarget Prediction and Functional Enrichment Analysis\u003c/h2\u003e \u003cp\u003eTo elucidate the potential biological mechanisms linking the identified causal miRNAs to UC, we performed a downstream functional enrichment analysis. First, experimentally validated target genes of the significant miRNAs were retrieved from the multiMiR database (version 2.3.0), focusing exclusively on interactions supported by strong experimental evidence (\"validated\" table) to ensure reliability. The resulting gene list was then mapped to Entrez IDs using the org.Hs.eg.db package. Subsequently, we conducted Gene Ontology (GO) Biological Process and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses using the clusterProfiler package in R. Statistical significance was determined using the Benjamini-Hochberg (BH) method to adjust P-values for multiple testing, with a threshold of adjusted P\u0026thinsp;\u0026lt;\u0026thinsp;0.05. To specifically investigate the relevance of these targets to UC pathogenesis, we prioritized and visualized pathways containing keywords associated with intestinal inflammation and immune regulation, such as \"Inflammatory bowel disease\", \"Th17 cell differentiation\", \"Cytokine-cytokine receptor interaction\", and \"NF-kappa B signaling pathway\".\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eExternal Validation Using Transcriptomic Data\u003c/h3\u003e\n\u003cp\u003eTo corroborate the causal associations identified by Mendelian randomization, we performed an external validation using independent transcriptomic expression profiling data retrieved from the GEO database. We utilized the dataset GSE122618, which contains plasma miRNA expression profiles from patients with UC and healthy controls. The dataset comprised 39 UC samples and 7 healthy control samples. Raw count data were imported into R and subjected to rigorous preprocessing. To account for differences in sequencing depth across samples, raw counts were normalized to Counts Per Million (CPM) and subsequently log2-transformed (log2 (CPM\u0026thinsp;+\u0026thinsp;1)) to stabilize variance and ensure comparability. We specifically extracted the expression levels of the candidate miRNAs identified in the MR analysis. The differential expression of these miRNAs between the UC and control groups was evaluated using the non-parametric Wilcoxon rank-sum test (Mann-Whitney U test), as the data did not necessarily follow a normal distribution. Visualizations were generated using the ggpubr package, and a P-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered statistically significant for validation.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eIdentification of Causal miRNAs in the Discovery Phase\u003c/h2\u003e \u003cp\u003eIn the discovery phase, we utilized cis-acting eQTLs from the Nikpay dataset to screen for miRNAs causally associated with UC (Supplementary materials: Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). All selected IVs exhibited strong predictive power for miRNA expression, with F-statistics ranging from 22.13 to 200.46 (mean F\u0026thinsp;=\u0026thinsp;50.8), well above the conventional threshold of 10, indicating a low risk of weak instrument bias. Analysis of the FinnGen dataset identified seven miRNAs significantly associated with UC risk (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05, Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Specifically, genetically predicted higher levels of miR-101-3p (OR\u0026thinsp;=\u0026thinsp;4.03, 95% CI: 1.96\u0026ndash;8.27, P\u0026thinsp;=\u0026thinsp;1.43e\u0026thinsp;\u0026minus;\u0026thinsp;4) and miR-1908-5p (OR\u0026thinsp;=\u0026thinsp;1.61, 95% CI: 1.29\u0026ndash;2.01, P\u0026thinsp;=\u0026thinsp;2.83e\u0026thinsp;\u0026minus;\u0026thinsp;5) were strongly associated with an increased susceptibility to UC based on the Wald ratio method. Conversely, miR-130b-3p (OR\u0026thinsp;=\u0026thinsp;0.81, 95% CI: 0.68\u0026ndash;0.97, P\u0026thinsp;=\u0026thinsp;0.019) and miR-130b-5p (OR\u0026thinsp;=\u0026thinsp;0.67, 95% CI: 0.48\u0026ndash;0.94, P\u0026thinsp;=\u0026thinsp;0.020) demonstrated significant protective effects. In the GCST dataset, six miRNAs were identified as potential causal determinants. Notably, miR-27a-3p (OR\u0026thinsp;=\u0026thinsp;9.47, P\u0026thinsp;=\u0026thinsp;8.68e\u0026thinsp;\u0026minus;\u0026thinsp;4) and let-7a-5p (OR\u0026thinsp;=\u0026thinsp;9.54, P\u0026thinsp;=\u0026thinsp;0.002) exhibited substantial risk effects. Most importantly, the protective effect of miR-130b-3p was consistently replicated in the GCST cohort using the IVW method (OR\u0026thinsp;=\u0026thinsp;0.82, 95% CI: 0.67\u0026ndash;0.99, P\u0026thinsp;=\u0026thinsp;0.044), reinforcing its potential role as a robust protective factor against UC. The Steiger directionality test confirmed the validity of the causal direction for all significant associations (all Steiger P\u0026thinsp;\u0026lt;\u0026thinsp;0.05), ruling out reverse causality in this phase.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eValidation of Causal Associations Using Alternative Instruments\u003c/h2\u003e \u003cp\u003eTo further substantiate the robustness of our findings, we conducted a validation analysis using alternative sets of IVs derived from trans-eQTLs (Nikpay et al.) and cis-eQTLs from the FHS cohort (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, Supplementary materials: Table \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e). The strength of these instruments remained robust across all analyses (F-statistics\u0026thinsp;\u0026gt;\u0026thinsp;22). In the validation using trans-eQTLs from the Nikpay dataset, miR-4326 showed a significant risk effect on UC in the FinnGen cohort (OR\u0026thinsp;=\u0026thinsp;1.22, 95% CI: 1.07\u0026ndash;1.39, P\u0026thinsp;=\u0026thinsp;0.003), although this association was not statistically significant in the GCST cohort (P\u0026thinsp;=\u0026thinsp;0.563). Other candidates, including miR-130b-3p, did not reach statistical significance using trans-IVs in either outcome dataset, likely due to the inherent complexity and potential pleiotropy of trans-acting variants. However, validation using cis-eQTLs from the FHS dataset yielded compelling results. miR-130b-5p was significantly associated with a reduced risk of UC in both the FinnGen cohort (OR\u0026thinsp;=\u0026thinsp;0.92, 95% CI: 0.86\u0026ndash;1.00, P\u0026thinsp;=\u0026thinsp;0.045) and the GCST cohort (OR\u0026thinsp;=\u0026thinsp;0.91, 95% CI: 0.84\u0026ndash;0.99, P\u0026thinsp;=\u0026thinsp;0.038). This finding aligns with the protective effect observed for the miR-130b family in the discovery phase. Additionally, miR-130b-3p showed a trend towards protection in the FinnGen cohort (OR\u0026thinsp;=\u0026thinsp;0.94, 95% CI: 0.89\u0026ndash;0.99, P\u0026thinsp;=\u0026thinsp;0.031), further supporting the potential therapeutic relevance of the miR-130b cluster. The Steiger directionality test confirmed the correct causal direction for all significant estimates (all P\u0026thinsp;\u0026lt;\u0026thinsp;0.05).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eAssessment of Reverse Causality\u003c/h2\u003e \u003cp\u003eTo investigate the potential reverse causal effects of UC on the identified candidate miRNAs, we performed a reverse Mendelian Randomization analysis using genetic instruments associated with UC from two independent large-scale GWAS datasets (FinnGen and GCST). In the analysis using the FinnGen UC dataset as the exposure, the IVW method showed no significant causal association between genetic liability to UC and the expression levels of any tested miRNAs (all P\u0026thinsp;\u0026gt;\u0026thinsp;0.05, Supplementary materials: Table \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e). Sensitivity analysis using the weighted median method yielded consistent non-significant estimates, further suggesting that UC liability does not influence the levels of these miRNAs in this cohort. Validation analysis using the GCST UC dataset largely corroborated these findings. The majority of miRNAs showed no significant association with UC liability. Although nominal associations were observed for miR-6810-3p (β\u0026thinsp;=\u0026thinsp;0.070, SE\u0026thinsp;=\u0026thinsp;0.033, P\u0026thinsp;=\u0026thinsp;0.032) and miR-130b-5p (β\u0026thinsp;=\u0026thinsp;0.074, SE\u0026thinsp;=\u0026thinsp;0.037, P\u0026thinsp;=\u0026thinsp;0.048) under the IVW method in the GCST cohort, these associations were not statistically significant in the weighted median sensitivity analysis (P\u0026thinsp;=\u0026thinsp;0.051 and P\u0026thinsp;=\u0026thinsp;0.099, respectively). Furthermore, given the lack of replication in the FinnGen dataset (β\u0026thinsp;=\u0026thinsp;0.027, P\u0026thinsp;=\u0026thinsp;0.318 for miR-6810-3p; β = -0.046, P\u0026thinsp;=\u0026thinsp;0.070 for miR-130b-5p), the evidence for a reverse causal effect remains weak and inconsistent. Collectively, the reverse MR analyses across two independent cohorts provide no robust evidence to support a reverse causal pathway from UC to the identified miRNAs, thereby strengthening the validity of the causal effects observed in the forward MR analysis.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eBayesian Colocalization Analysis\u003c/h2\u003e \u003cp\u003eTo distinguish between shared causal variants (pleiotropy) and distinct variants in linkage disequilibrium, we performed Bayesian colocalization analysis for the identified miRNA-UC associations (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, Supplementary materials: Table \u003cspan refid=\"MOESM4\" class=\"InternalRef\"\u003eS4\u003c/span\u003e). A high PP.H4 indicates a shared causal variant. In the analysis of the GCST cohort, miR-27a-3p exhibited the strongest evidence for colocalization, with a PP.H4 of 0.608, suggesting a potential shared genetic mechanism at the 19p13.11 locus (lead SNP rs2594718). Similarly, let-7a-5p and let-7d-5p showed moderate support for colocalization in the GCST dataset (PP.H4\u0026thinsp;=\u0026thinsp;0.409 and 0.430, respectively). In the FinnGen cohort, miR-1908-5p displayed suggestive evidence of colocalization (PP.H4\u0026thinsp;=\u0026thinsp;0.583) at the 11q12.1 locus, whereas its colocalization probability in the GCST cohort was low. However, for miR-130b-3p and miR-130b-5p, which demonstrated consistent protective effects in the MR analysis, the evidence for colocalization was limited in both the FinnGen (PP.H4\u0026thinsp;\u0026asymp;\u0026thinsp;0.11) and GCST (PP.H4\u0026thinsp;\u0026asymp;\u0026thinsp;0.04) cohorts. This relatively low PP.H4, combined with low PP.H3 values (probability of distinct causal variants), suggests that while the MR assumptions hold, the current GWAS summary statistics may lack sufficient resolution to definitively pinpoint a single shared variant, or the association may be driven by multiple variants in the region. Other candidates, such as miR-101-3p in FinnGen, showed high support for distinct causal variants (PP.H3\u0026thinsp;=\u0026thinsp;0.95), indicating that the observed association might be partly driven by linkage disequilibrium rather than direct pleiotropy.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eIntegrative Functional Enrichment Analysis of miR-130b-3p/5p Targets\u003c/h2\u003e \u003cp\u003eTo comprehensively characterize the regulatory network of miR-130b, we predicted the downstream targets for both miR-130b-3p and miR-130b-5p, identifying 5,149 and 346 potential target genes (Supplementary materials: Table \u003cspan refid=\"MOESM5\" class=\"InternalRef\"\u003eS5\u003c/span\u003e), respectively. Given the overlapping biological contexts of these distinct miRNA strands, we performed a combined functional enrichment analysis to delineate their collective biological impact. Hierarchical clustering of GO terms highlighted that the predicted targets are functionally interconnected, clustering into five distinct biological modules (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). A substantial proportion of targets were enriched in clusters related to structural organization and adhesion (e.g., \"actin filament organization,\" \"cell-substrate adhesion\"), suggesting a critical role in regulating cellular morphology and motility. Furthermore, a distinct cluster enriched in cell cycle progression terms (e.g., \"Wnt signaling pathway,\" \"chromosome segregation\") indicated that miR-130b may function as a proliferative regulator. Notably, we also observed specific enrichment in neurodevelopmental processes (e.g., \"axonogenesis,\" \"forebrain development\"), hinting at a potential tissue-specific function in the nervous system. Consistent with the GO findings, KEGG pathway analysis demonstrated that the targets were predominantly enriched in pathways governing cell-matrix interactions and signal transduction (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). \"Focal adhesion\" was identified as the most significantly enriched pathway (lowest adjusted P-value), reinforcing the GO results regarding cell adhesion. Additionally, the analysis highlighted broad regulatory potential in multiple signaling cascades, including the MAPK, FoxO, and Insulin signaling pathways. It is noteworthy that several cancer-related pathways (e.g., \"Proteoglycans in cancer,\" \"Prostate cancer,\" \"Renal cell carcinoma\") were highly represented, implying that miR-130b-3p/5p dysregulation may be a pivotal event in tumorigenesis and cancer progression.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eExternal Validation in Plasma Samples from UC Patients\u003c/h2\u003e \u003cp\u003eTo evaluate whether the identified causal miRNAs exhibit detectable alterations in circulating biofluids, we further examined their expression levels in plasma samples from an independent cohort of UC patients and healthy controls. In this plasma-based validation, none of the candidate miRNAs reached the conventional threshold for statistical significance (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05, Supplementary materials: Table \u003cspan refid=\"MOESM6\" class=\"InternalRef\"\u003eS6\u003c/span\u003e, Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). However, notable trends were observed for the miR-130b family. Specifically, miR-130b-3p showed a trend towards upregulation in UC patients compared to controls (log2 FC\u0026thinsp;=\u0026thinsp;0.37, P\u0026thinsp;=\u0026thinsp;0.075), as did miR-130b-5p (log2 FC\u0026thinsp;=\u0026thinsp;0.58, P\u0026thinsp;=\u0026thinsp;0.080). Other candidates, such as miR-671-3p and miR-27a-3p, also displayed elevated expression levels in the UC group but lacked statistical support (P\u0026thinsp;\u0026gt;\u0026thinsp;0.10). The absence of significant differences in circulating miRNAs, in contrast to tissue-specific associations, may reflect the dilution effect in peripheral blood or distinct regulatory dynamics between local intestinal tissue and systemic circulation.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis comprehensive two-sample Mendelian randomization study provides genetic evidence that multiple circulating miRNAs are involved in UC susceptibility. Across discovery and validation phases, we observed a pattern in which members of the miR-130b family (miR-130b-3p and miR-130b-5p) showed consistent protective associations, whereas several other miRNAs\u0026mdash;most prominently miR-101-3p, miR-1908-5p, miR-27a-3p, let-7a-5p, miR-103a-3p, and miR-4326\u0026mdash;were associated with increased UC risk in at least one dataset. Additional candidates, including miR-6891-3p, miR-6810-3p, miR-671-3p, and let-7d-5p, showed suggestive associations and biologically plausible links to mucosal immunity or inflammatory signaling. Together with colocalization, enrichment, reverse-MR and expression analyses, these findings extend previous observational work on miRNAs in inflammatory bowel disease (IBD) and highlight a mechanistically diverse set of miRNAs that may contribute to UC pathogenesis.\u003c/p\u003e \u003cp\u003eThe most robust and internally consistent finding in our study is the protective association of the miR-130b family. In the discovery phase, genetically predicted higher levels of miR-130b-3p and miR-130b-5p were associated with reduced UC risk in FinnGen (OR\u0026thinsp;~\u0026thinsp;0.67\u0026ndash;0.81). This protective pattern was replicated for miR-130b-3p in the GCST dataset and for miR-130b-5p using independent cis-eQTLs from the Framingham Heart Study in both UC GWAS cohorts. The consistency across instruments (cis-eQTLs in two cohorts) and outcomes, as well as Steiger tests supporting the miRNA\u0026rarr;UC direction, make reverse causation unlikely. Functionally, miR-130b has been implicated in several inflammatory and tissue-remodeling contexts. In adipose tissue, miR-130b promotes obesity-associated inflammation and insulin resistance by repressing PPAR-γ and skewing macrophages toward a pro-inflammatory M1 phenotype, linking it to metabolic inflammation [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. In contrast, in diabetic nephropathy, miR-130b mimics attenuated renal tubulointerstitial fibrosis by inhibiting Snail-driven epithelial\u0026ndash;mesenchymal transition (EMT) and restoring E-cadherin expression, thereby limiting fibrotic remodeling [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. More recently, miR-130b-3p has been shown to dampen Th2-type airway inflammation in asthma by targeting HMGB1 and suppressing the HMGB1\u0026ndash;TLR4\u0026ndash;DRP1 axis [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. These apparently divergent effects suggest that miR-130b is a context-dependent regulator of inflammatory and EMT-related pathways, with potential to be either detrimental (e.g. in metabolic inflammation or colorectal cancer) or protective (e.g. in fibrotic kidney disease and allergic airway inflammation) depending on cellular context and target engagement. Our MR results, which integrate genetically proxied lifelong higher circulating miR-130b levels, align more closely with its anti-inflammatory/fibrosis-limiting roles. In chronic intestinal inflammation, where barrier injury, EMT-like changes, and innate immune activation coexist, up-regulation of miR-130b may preferentially restrain profibrotic and HMGB1/TLR-driven inflammatory circuits, outweighing any adverse effects seen in tumorigenic settings. This interpretation is supported by our enrichment analysis, which highlighted miR-130b targets in pathways such as autophagy, FoxO signaling pathway, and MAPK signaling pathway\u0026mdash;pathways centrally implicated in UC pathophysiology. Although plasma expression of miR-130b-3p/5p trended higher rather than lower in UC patients in our external cohort, this discrepancy is biologically plausible: genetically protective miRNAs can be secondarily up-regulated as a compensatory response in active disease, and cross-sectional plasma levels may not faithfully recapitulate lifetime genetically determined exposure.\u003c/p\u003e \u003cp\u003eIn contrast to the miR-130b cluster, several miRNAs showed risk-increasing effects, with variable support across datasets and instruments. In the FinnGen discovery analysis, miR-101-3p and miR-1908-5p emerged as strong risk factors, while in the GCST replication, miR-27a-3p and let-7a-5p displayed large risk-enhancing ORs. In the trans-eQTL\u0026ndash;based validation using the Nikpay dataset, miR-4326 showed a significant risk effect in FinnGen. These signals, although not uniformly replicated across all analyses, are biologically coherent with known roles of these miRNAs in epithelial homeostasis, cytokine signaling, and autoimmunity. miR-101-3p is widely recognized as a tumor-suppressor miRNA in several cancers, including colorectal cancer (CRC), where it directly targets the PGE₂ receptor EP4 and represses PI3K/AKT and other oncogenic pathways [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. In colon cancer specimens, down-regulation of miR-101 is associated with increased EP4 expression, enhanced proliferation and migration, and poor prognosis [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. Moreover, miR-101 is part of a diagnostic panel (miR-19a, miR-21, miR-31, miR-146a, miR-375) that differentiates UC from Crohn\u0026rsquo;s disease across colon tissue, blood, and saliva, implicating it in IBD pathogenesis [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. Our MR finding that genetically higher plasma miR-101-3p levels increase UC risk is therefore counterintuitive relative to its tumor-suppressive role, but may reflect context-specific, isoform-specific (-3p vs -5p) or compartment-specific effects. It is conceivable that systemic elevation of miR-101-3p modulates immune cell or stromal targets differently from local epithelial miR-101 in the colon, or that the MR signal partially captures linkage with nearby inflammatory regulators.\u003c/p\u003e \u003cp\u003emiR-1908-5p represents another notable risk-associated miRNA in our analysis. Integrative eQTL work has previously shown that Crohn\u0026rsquo;s disease and rheumatoid arthritis risk SNPs can significantly alter miR-1908-5p expression in lymphoblastoid cell lines, with enrichment of predicted targets among autoimmune GWAS loci, suggesting a potential role in systemic autoimmunity [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. A recent review summarized its broad involvement in cancer, fibrosis, obesity, and rheumatoid arthritis, often converging on NF-κB activation and TGF-β1/Smad signaling [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. Our MR results, together with this prior evidence, support a model in which miR-1908-5p promotes pro-inflammatory and profibrotic circuits relevant to mucosal immunity and UC.\u003c/p\u003e \u003cp\u003eThe large risk effects for miR-27a-3p and let-7a-5p in the GCST cohort are in line with their known roles in inflammation and intestinal pathology. miR-27a-3p is consistently up-regulated in CRC tissue and cell lines and promotes proliferation and survival partly by targeting BTG1 and modulating ERK/MEK signaling [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. Experimental work in osteoblasts also shows that miR-27a-3p represses GLP1R and AMPK signaling, reducing autophagy and differentiation while increasing inflammatory cytokine production, linking it to metabolic-inflammatory regulation [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. Our colocalization analysis indicated the strongest evidence for a shared causal variant at the miR-27a-3p locus in the GCST dataset (PP.H4\u0026thinsp;\u0026asymp;\u0026thinsp;0.61), supporting a direct genetic mechanism at 19p13.11.\u003c/p\u003e \u003cp\u003elet-7a-5p belongs to the broadly expressed let-7 family, which generally acts as a tumor suppressor and differentiation-promoting factor in the intestine. In mouse models, global intestinal loss of let-7 miRNAs leads to adenocarcinoma formation and stem-like transcriptional programs, accompanied by up-regulation of Hmga1/2 and other oncogenic targets [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. However, several studies have demonstrated anti-inflammatory roles for let-7 family members: let-7a ameliorates ConA-induced hepatitis by inhibiting IL-6\u0026ndash;dependent Th17 differentiation, and let-7a-5p reduces TNF-α, IL-1β and IL-6 expression and Ras/MAPK activation in chronic rhinosinusitis fibroblasts [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. In pediatric UC, tissue levels of let-7 are reduced along with miR-124, contributing to enhanced STAT3 phosphorylation [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. Thus, our MR evidence that genetically increased plasma let-7a-5p heightens UC risk may again reflect systemic vs mucosal compartmentalization, family-member specificity, or complex feedback loops: higher circulating let-7a-5p could be a marker of a broader regulatory state that, while compensatory in some tissues, correlates with greater UC susceptibility overall.\u003c/p\u003e \u003cp\u003emiR-103a-3p and let-7d-5p, which showed nominal associations in our discovery analyses and/or colocalization, also have compelling links to intestinal inflammation and fibrosis. In Crohn\u0026rsquo;s disease, exosomal miR-103a-3p derived from creeping fat adipose-derived stem cells promotes intestinal fibrosis by targeting TGFBR3, activating Smad2/3, and driving fibroblast activation; its expression in diseased intestine correlates with the severity of fibrosis [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. Beyond the gut, miR-103a-3p promotes fibrosis and inflammation in non-alcoholic fatty liver disease by repressing HBP1 [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. let-7d-5p, in a neonatal necrotizing enterocolitis (NEC) model, is down-regulated in inflamed intestinal tissue, and its over-expression suppresses TLR4/NF-κB signaling and pro-inflammatory cytokines via targeting LGALS3, thereby reducing epithelial apoptosis and mucosal injury [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. These data suggest that miR-103a-3p may be broadly pro-fibrotic and pro-inflammatory in the intestine, whereas let-7d-5p appears anti-inflammatory. The direction and strength of their MR associations with UC in our study were more modest, but their inclusion among suggestive candidates and their enrichment in UC-relevant pathways warrant further follow-up.\u003c/p\u003e \u003cp\u003emiR-4326 and miR-671-3p, although less well studied in IBD, both converge on signaling axes that intersect with mucosal homeostasis. miR-4326 is up-regulated in lung cancer and promotes proliferation by targeting APC2, a negative regulator of the Wnt/β-catenin pathway [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. Given that aberrant Wnt signaling is pivotal in intestinal epithelial renewal and colorectal carcinogenesis, genetically elevated miR-4326 could theoretically perturb epithelial barrier dynamics and regeneration, making its risk association with UC in our trans-eQTL MR (FinnGen) mechanistically plausible. miR-671-3p has been described both as an oncogenic factor in glioma and as a regulator of inflammation and matrix metabolism in osteoarthritis, where it targets TRAF3 and modulates chondrocyte apoptosis and pro-inflammatory cytokine production [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. TRAF3 is a key adaptor in TNF receptor and Toll-like receptor pathways, suggesting that miR-671-3p could influence innate immune activation in the gut as well.\u003c/p\u003e \u003cp\u003eAmong the discovery-phase miRNAs, miR-6891-3p and miR-6810-3p stand out as largely unexplored in the IBD literature, yet their genomic context and predicted functions are suggestive. miR-6891 is encoded within intron 4 of HLA-B in the MHC class I region; prior functional work has shown that its 5p strand (miR-6891-5p) regulates IgA heavy chain transcripts (IGHA1/2), and that elevated miR-6891-5p expression in lymphoblastoid cells from IgA-deficient patients suppresses IgA production [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. Secretory IgA is central to mucosal barrier immunity, and selective IgA deficiency is epidemiologically linked to IBD and celiac disease. Although our MR implicates the 3p strand, the shared precursor locus suggests that MIR6891-derived miRNAs could mechanistically couple HLA variation, IgA homeostasis, and UC risk.\u003c/p\u003e \u003cp\u003emiR-6810-3p, while not previously connected to intestinal inflammation, has been detected in human plasma and is highly enriched in brain tissue and appendix according to transcriptomic atlases. A recent population-based profiling of plasma miRNAs and persistent organic pollutants suggested that miR-6810-3p targets autophagy-related pathways [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. Given the central role of autophagy genes (e.g., ATG16L1, IRGM) in Crohn\u0026rsquo;s disease and their contributions to Paneth cell function and bacterial handling, a causal link between genetically determined miR-6810-3p levels and UC could point toward an underappreciated autophagy-modulating miRNA in mucosal immunity. At present, however, the evidence is preliminary and driven primarily by MR statistics and in-silico functional prediction rather than direct experimental data.\u003c/p\u003e \u003cp\u003eOur Bayesian colocalization analysis helped distinguish likely shared causal variants from associations potentially driven by linkage disequilibrium. The strongest support for colocalization (PP.H4\u0026thinsp;\u0026gt;\u0026thinsp;0.6) was seen for miR-27a-3p in the GCST dataset and for miR-1908-5p in FinnGen, whereas the miR-130b locus showed relatively low PP.H4 but also low PP.H3, suggesting either multiple causal variants or limited resolution of current GWAS summary statistics. This pattern is not unexpected in regions with complex LD structure and modest effect sizes; it argues for cautious interpretation and motivates fine-mapping using denser genotyping or whole-genome sequencing data in UC cohorts. Although the reverse-direction MR results were not the primary focus of this work, the absence of strong and consistent UC\u0026rarr;miRNA effects for most candidates (as suggested by Steiger directionality and the lack of robust reverse IVW estimates) supports our main inference that the identified miRNAs are more likely upstream determinants rather than downstream consequences of genetic UC liability. Nonetheless, disease activity, treatment, and environmental exposures can modulate miRNA levels independently of germline predisposition, which likely explains why our external plasma validation (GSE122618) did not show statistically significant differences despite suggestive trends for miR-130b-3p/5p, miR-671-3p, and miR-27a-3p. The apparent disconnect between MR and cross-sectional expression findings underscores a broader point: MR estimates reflect the effect of lifelong genetic perturbation of miRNA levels, averaged across tissues and disease stages, whereas transcriptomic snapshots capture short-term, tissue- and context-specific regulation. For example, protective miRNAs may be up-regulated as a counter-regulatory response in established disease, while genetically harmful miRNAs could be partially suppressed by therapy or feedback inhibition. In addition, many of our instruments derive from whole-blood or plasma eQTLs; their relationship to miRNA abundance in the colonic epithelium, lamina propria lymphocytes and mesenchymal cells is imperfect, and some genetically determined variation may be cell-type specific.\u003c/p\u003e \u003cp\u003eKey strengths of this study include the two-stage design with independent discovery and validation cohorts, the use of multiple instrument sets (cis and trans eQTLs across Nikpay and FHS), careful LD clumping and weak-instrument filtering, extensive sensitivity analyses (Steiger tests, pleiotropy screening via LDtrait, reverse MR), and Bayesian colocalization. The integration of experimentally validated miRNA\u0026ndash;target data and pathway enrichment (highlighting IBD, Th17, NF-κB and cytokine signaling pathways) provides mechanistic plausibility that is often lacking in purely statistical MR screens. Our work also extends miRNA-focused MR beyond cardiometabolic traits to complex immune-mediated disease. Several limitations should be acknowledged. First, all GWAS and eQTL data were derived from individuals of European ancestry, limiting generalizability to other populations where both miRNA eQTL architecture and UC genetics may differ. Second, some miRNAs were instrumented by a small number of SNPs, and while all F-statistics exceeded 10, power to detect modest causal effects and to robustly assess pleiotropy remained limited. Third, the reliance on blood/plasma eQTLs may not capture tissue-specific miRNA regulation in the gut; future work using intestinal eQTL datasets, once available at scale, will be critical. Fourth, colocalization results for several key miRNAs (e.g., miR-130b-3p/5p) were inconclusive, reflecting LD complexity and limited SNP density near short miRNA genes. Finally, our external expression validation cohort was relatively small and heterogeneous with respect to disease activity, which reduces power to detect modest differences in circulating miRNA levels.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn conclusion, this study provides compelling genetic evidence that specific circulating miRNAs causally influence UC susceptibility. The identification of protective (miR-130b family) and risk-increasing (miR-101-3p, miR-1908-5p, miR-27a-3p, let-7a-5p) miRNAs advances our understanding of UC pathogenesis and highlights potential biomarkers and therapeutic targets. These findings support the emerging paradigm of miRNAs as key regulators in inflammatory bowel disease and provide a foundation for developing precision medicine approaches in UC management.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics Approval and Consent to Participate\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study is a secondary analysis of publicly available summary-level data. The original studies (FinnGen, Framingham Heart Study, and other GWAS cohorts) received ethical approval from their respective institutional review boards, and informed consent was obtained from all participants. No separate ethical approval or informed consent was required for the present study, as it utilized anonymized, aggregate data.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for Publication\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of Data and Materials\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets presented in this study can be found in online repositories. The summary statistics for Ulcerative Colitis were obtained from the FinnGen consortium (Release 12, https://www.finngen.fi/en) and the EBI GWAS Catalog (Accession ID: GCST90473823). The miRNA eQTL data were sourced from the publications by Nikpay et al. and the Framingham Heart Study. The external validation transcriptomic dataset is available in the Gene Expression Omnibus (GEO) under accession number GSE122618. All other data generated or analyzed during this study are included in this published article and its supplementary information files.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; Contributions\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eConceptualization and Methodology: W. Chu; Investigation and Formal Analysis: J. He and H. Jin; Resources: W. Chu; Writing \u0026ndash; Original Draft: J. He; Writing \u0026ndash; Review \u0026amp; Editing: H. Jin and W. Chu. All authors have read and agreed to the published version of the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe gratefully acknowledge the participants and investigators of the FinnGen study, the Framingham Heart Study, and the authors of the original GWAS and eQTL studies for making their summary statistics publicly available. We also thank the contributors to the GEO database for providing the transcriptomic data used for validation.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eNg SC et al (2017) Worldwide incidence and prevalence of inflammatory bowel disease in the 21st century: a systematic review of population-based studies. Lancet 390(10114):2769\u0026ndash;2778\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLewis JD et al (2023) Incidence, Prevalence, and Racial and Ethnic Distribution of Inflammatory Bowel Disease in the United States. Gastroenterology 165(5):1197\u0026ndash;1205e2\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu F et al (2018) Health-Risk Behaviors and Chronic Conditions Among Adults with Inflammatory Bowel Disease - United States, 2015 and 2016. MMWR Morb Mortal Wkly Rep 67(6):190\u0026ndash;195\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDahlhamer JM et al (2016) Prevalence of Inflammatory Bowel Disease Among Adults Aged\u0026thinsp;\u0026ge;\u0026thinsp;18 Years - United States, 2015. MMWR Morb Mortal Wkly Rep 65(42):1166\u0026ndash;1169\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVermeire S et al (2025) Obefazimod in patients with moderate-to-severely active ulcerative colitis: efficacy and safety analysis from the 96-week open-label maintenance phase 2b study. J Crohns Colitis, 19(5)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSokal-Dembowska A et al (2025) The Role of microRNAs in Inflammatory Bowel Disease. Int J Mol Sci, 26(10)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu F et al (2008) MicroRNAs are differentially expressed in ulcerative colitis and alter expression of macrophage inflammatory peptide-2 alpha. Gastroenterology 135(5):1624\u0026ndash;1635e24\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTakagi T et al (2010) Increased expression of microRNA in the inflamed colonic mucosa of patients with active ulcerative colitis. J Gastroenterol Hepatol 25(Suppl 1):S129\u0026ndash;S133\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKoukos G et al (2015) A microRNA signature in pediatric ulcerative colitis: deregulation of the miR-4284/CXCL5 pathway in the intestinal epithelium. Inflamm Bowel Dis 21(5):996\u0026ndash;1005\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eParaskevi A et al (2012) Circulating MicroRNA in inflammatory bowel disease. J Crohns Colitis 6(9):900\u0026ndash;904\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDuttagupta R et al (2012) Genome-wide maps of circulating miRNA biomarkers for ulcerative colitis. PLoS ONE 7(2):e31241\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSch\u0026ouml;nauen K et al (2018) Circulating and Fecal microRNAs as Biomarkers for Inflammatory Bowel Diseases. Inflamm Bowel Dis 24(7):1547\u0026ndash;1557\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDavey Smith G, Hemani G (2014) Mendelian randomization: genetic anchors for causal inference in epidemiological studies. Hum Mol Genet 23(R1):R89\u0026ndash;98\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu B et al (2022) Two-Sample Mendelian Randomization Analysis Investigates Causal Associations Between Gut Microbial Genera and Inflammatory Bowel Disease, and Specificity Causal Associations in Ulcerative Colitis or Crohn's Disease. Front Immunol 13:921546\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNikpay M et al (2019) Genome-wide identification of circulating-miRNA expression quantitative trait loci reveals the role of several miRNAs in the regulation of cardiometabolic phenotypes. Cardiovasc Res 115(11):1629\u0026ndash;1645\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuan T et al (2015) Genome-wide identification of microRNA expression quantitative trait loci. Nat Commun 6:6601\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang M et al (2016) MiR-130b promotes obesity associated adipose tissue inflammation and insulin resistance in diabetes mice through alleviating M2 macrophage polarization via repression of PPAR-γ. Immunol Lett 180:1\u0026ndash;8\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBai X et al (2016) MicroRNA-130b improves renal tubulointerstitial fibrosis via repression of Snail-induced epithelial-mesenchymal transition in diabetic nephropathy. Sci Rep 6:20475\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHan X et al (2024) Mechanism of miR-130b-3p in relieving airway inflammation in asthma through HMGB1-TLR4-DRP1 axis. Cell Mol Life Sci 82(1):9\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChandramouli A et al (2012) MicroRNA-101 (miR-101) post-transcriptionally regulates the expression of EP4 receptor in colon cancers. Cancer Biol Ther 13(3):175\u0026ndash;183\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchaefer JS et al (2015) MicroRNA signatures differentiate Crohn's disease from ulcerative colitis. BMC Immunol 16:5\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWohlers I, Bertram L, Lill CM (2018) Evidence for a potential role of miR-1908-5p and miR-3614-5p in autoimmune disease risk using integrative bioinformatics. J Autoimmun 94:83\u0026ndash;89\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGhafouri-Fard S et al (2022) miR-1908: a microRNA with diverse functions in cancers and non-malignant conditions. Cancer Cell Int 22(1):281\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSu C et al (2019) miR-27a-3p regulates proliferation and apoptosis of colon cancer cells by potentially targeting BTG1. Oncol Lett 18(3):2825\u0026ndash;2834\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZeng Z et al (2021) MiR-27a-3p Targets GLP1R to Regulate Differentiation, Autophagy, and Release of Inflammatory Factors in Pre-Osteoblasts via the AMPK Signaling Pathway. Front Genet 12:783352\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMadison BB et al (2015) Let-7 Represses Carcinogenesis and a Stem Cell Phenotype in the Intestine via Regulation of Hmga2. PLoS Genet 11(8):e1005408\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang Y et al (2013) MicroRNA let-7a ameliorates con A-induced hepatitis by inhibiting IL-6-dependent Th17 cell differentiation. J Clin Immunol 33(3):630\u0026ndash;639\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKoukos G et al (2013) MicroRNA-124 regulates STAT3 expression and is down-regulated in colon tissues of pediatric patients with ulcerative colitis. Gastroenterology, 145(4): p. 842\u0026thinsp;\u0026ndash;\u0026thinsp;52.e2.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQian W et al (2023) Exosomal miR-103a-3p from Crohn's Creeping Fat-Derived Adipose-Derived Stem Cells Contributes to Intestinal Fibrosis by Targeting TGFBR3 and Activating Fibroblasts. J Crohns Colitis 17(8):1291\u0026ndash;1308\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChu K, Gu J (2022) microRNA-103a-3p promotes inflammation and fibrosis in nonalcoholic fatty liver disease by targeting HBP1. Immunopharmacol Immunotoxicol 44(6):993\u0026ndash;1003\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun L et al (2020) Let-7d-5p suppresses inflammatory response in neonatal rats with necrotizing enterocolitis via LGALS3-mediated TLR4/NF-κB signaling pathway. Am J Physiol Cell Physiol 319(6):C967\u0026ndash;C979\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu G et al (2018) miR-4326 promotes lung cancer cell proliferation through targeting tumor suppressor APC2. Mol Cell Biochem 443(1\u0026ndash;2):151\u0026ndash;157\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLu G et al (2018) MicroRNA-671-3p promotes proliferation and migration of glioma cells via targeting CKAP4. Onco Targets Ther 11:6217\u0026ndash;6226\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChitnis N et al (2017) An Expanded Role for HLA Genes: HLA-B Encodes a microRNA that Regulates IgA and Other Immune Response Transcripts. Front Immunol 8:583\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQu J et al (2025) Persistent organic pollutants and plasma microRNAs: A community-based profiling analysis. Environ Int 197:109328\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"internal-and-emergency-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"iaem","sideBox":"Learn more about [Internal and Emergency Medicine](http://link.springer.com/journal/11739)","snPcode":"11739","submissionUrl":"https://www.editorialmanager.com/iaem/default.aspx","title":"Internal and Emergency Medicine","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"miRNAs, ulcerative colitis, Mendelian randomization, Bayesian colocalization, pleiotropy","lastPublishedDoi":"10.21203/rs.3.rs-8634046/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8634046/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eCirculating microRNAs (miRNAs) are frequently dysregulated in ulcerative colitis (UC), yet whether these alterations represent causes or consequences of the disease remains unclear. This study aimed to systematically investigate the causal effect of plasma miRNAs on UC risk using a comprehensive multi-stage Mendelian randomization (MR) framework.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eWe employed a two-sample MR design utilizing genetic instruments for miRNAs derived from independent cis- and trans-eQTL datasets (Nikpay et al. and Framingham Heart Study). Outcome summary statistics were obtained from the FinnGen (n\u0026thinsp;=\u0026thinsp;499,380) and EBI GWAS Catalog (n\u0026thinsp;=\u0026thinsp;458,440) cohorts. Analyses included inverse variance weighted estimation, rigorous sensitivity checks, Bayesian colocalization to assess pleiotropy, and reverse MR. Findings were further biologically interpreted through target gene functional enrichment and triangulated with external transcriptomic validation (GSE122618).\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eThe miR-130b family (miR-130b-3p and miR-130b-5p) demonstrated consistent protective effects against UC across discovery and validation phases (e.g., miR-130b-5p: OR\u0026thinsp;=\u0026thinsp;0.91\u0026ndash;0.92, P\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Conversely, miR-101-3p, miR-1908-5p, miR-27a-3p, and let-7a-5p were identified as significant risk factors. Reverse MR analyses provided no robust evidence that UC liability influences these miRNA levels. Bayesian colocalization strongly supported shared causal variants for miR-27a-3p (PP.H4\u0026thinsp;=\u0026thinsp;0.61) and miR-1908-5p. Functional enrichment linked miR-130b targets to focal adhesion and MAPK/Wnt signaling pathways critical for mucosal homeostasis.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eThis study provides compelling genetic evidence that specific circulating miRNAs causally influence UC susceptibility. The identification of protective (miR-130b family) and risk-increasing miRNAs advances our understanding of UC pathogenesis, highlighting these molecules as promising biomarkers and potential therapeutic targets.\u003c/p\u003e","manuscriptTitle":"Unraveling the Causal Association Between microRNAs and Ulcerative Colitis: A Multi-Stage Mendelian Randomization Study with External Validation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-05 04:55:39","doi":"10.21203/rs.3.rs-8634046/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewersInvited","content":"","date":"2026-02-03T10:59:53+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-28T05:46:51+00:00","index":"","fulltext":""},{"type":"submitted","content":"Internal and Emergency Medicine","date":"2026-01-27T20:47:07+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"internal-and-emergency-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"iaem","sideBox":"Learn more about [Internal and Emergency Medicine](http://link.springer.com/journal/11739)","snPcode":"11739","submissionUrl":"https://www.editorialmanager.com/iaem/default.aspx","title":"Internal and Emergency Medicine","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"8bde31d0-edf1-4939-b191-1fac81c49339","owner":[],"postedDate":"February 5th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-02-05T04:55:39+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-05 04:55:39","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8634046","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8634046","identity":"rs-8634046","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}