Introduction Immune thrombocytopenia (ITP) is an acquired autoimmune hemorrhagic disease characterized by isolated thrombocytopenia. In a previous study, we observed dysbiosis in the phylogenetic composition and function of the gut microbiome of patients with ITP [Sci China Life Sci, 2020]. It is suggested that impaired intestinal barrier as well as the gut microbiota-related metabolites may provide signaling molecules and facilitate communication between the intestinal tract and the immune system. For example, bile acids (BAs), which are synthesized in the liver and transformed into secondary BAs in the gut, have recently been considered to be important signaling metabolites for the immune system. The bile acid receptor FXR can also inhibit the TLR4/MAPK pathway and participate in the regulation of the immune response. Here, we provide evidence that the gut-bile acid-immune axis plays a role in the pathogenesis of ITP and that obeticholic acid (OCA) treatment can repair the gut barrier and restore the gut-bile acid balance to reduce the destruction of platelets. Methods After establishing an active ITP mouse model, pathological analyses of biopsies, 16S amplicon sequencing, targeted bile acid metabolomics and real-time fluorescent quantitative PCR were used to explore the changes in intestinal barrier function, gut microbiota composition, and bile acid metabolites in feces and plasma. OCA treatment of the ITP mouse model and bone marrow-derived dendritic cells was used to investigate the effect of OCA on gut bile acid metabolism and T cell differentiation. In a clinical sample study, 65 treatment-naïve ITP patients and 62 healthy controls were enrolled. Metagenomic sequencing, bile acid-targeted metabolomics and mass cytometry were used to explore the characteristics of the "gut-bile acid-immune” axis in the ITP patients. Results In the mouse model of ITP, the gut barrier was damaged, as shown by exfoliation of IECs according to H&E staining and greater FITC-dextran leakage from the intestinal mucosa into the peripheral circulation. RNA sequencing of the IEC transcriptome revealed that the expression of claudin-4 and claudin-8 was significantly diminished in ITP mice. As a result of the impaired gut barrier, plasma levels of LPS were elevated in ITP mice. Moreover, the Firmicutes:Bacteroidetes ratio was elevated in the gut microbiota according to 16S amplicon sequencing, and increased fecal levels of BAs were observed, which gave priority to primary BAs like β-MCA. The accumulation of β-MCA, which is an FXR antagonist, inhibited the activity of FXR in IECs, hepatocytes and immune cells such as DCs. After stimulation with LPS in vitro, murine bone marrow-derived DCs were observed to induce greater Th1 differentiation, which could be reversed by an FXR agonist (GW4064 or OCA). The secretion of IL-12 by DCs induced Th1 differentiation and was regulated by the TLR4/MAPK/p38 pathway, and this effect was inhibited by OCA through the activation of FXR. In vivo, OCA increased the expression of claudin-4 and claudin-8, and restored the balance of bile acid metabolism in the mice with ITP. After OCA treatment, the Th1/Th2 imbalance in the bone marrow of the mice with ITP was corrected, and reduced platelet destruction in the spleen and liver was observed with an in vivo imaging system. Clinical studies have shown that the gut microbiota in ITP patients is dysregulated, which showed an increased abundance of Firmicutes. Scanning electron microscopy showed that the gaps between IECs were increased in the ITP patients. The bile acid pool of the ITP patients was shown to have an increased proportion of primary BAs in the bone marrow. After in vitro treatment with OCA for 72 h, the proportion of human Th1 cells among human bone marrow mononuclear cells was significantly reduced, which suggested the therapeutic effect of OCA in ITP patients. Conclusions This study preliminarily revealed the characteristics of the "gut-bile acid-immune” axis in ITP patients and a mouse model. In vitro and in vivo studies confirmed that OCA could repair the intestinal barrier, correct gut microbiota dysbiosis and bile acid metabolism abnormalities, and decrease the proportion of peripheral Th1 cells via MAPK signaling in DCs, resulting in a decrease in the excessive destruction of platelets. This study provides a theoretical basis for targeting the gut microbiota in the clinical treatment of ITP.