Immune thrombocytopenia (ITP) is an autoimmune disorder characterized by abnormal immune response. Though many therapies have been used, corticosteroid-resistance remains to be a challenge clinically. Extensive research has improved our understanding of ITP, showing that environmental factors affect the disease profile, such as Helicobacter pylori being proven to be associated with thrombocytopenia. Though evidence that the gut microbiota contributes to the development of auto-immune disorders is accumulating, there is no information available on relationship between gut microbiome and ITP. Berberine(BBR), a traditional compound isolated from a Chinese herb, has been widely used as a nonprescription drug to treat diarrhea. Recently, BBR has been reported to modulate microbiota structures, which contributes to improving metabolic disorders. Here, we hypothesized that BBR might modulate gut microbiota to treat ITP.


In order to investigate the relationship between gut microbiome and ITP, we performed deep shotgun sequencing on 253 fecal samples totally from consecutive ITP patients and healthy controls. Metagenome-wide association study (MWAS) was conducted, and clinical characteristics of patients were collected to analyze the correlation with gut microbiome (Nan Qin, et al. Nature. 2012). Certainly, a clinical cohort study was performed to assess the efficacy of BBR in corticosteroid-resistant ITP patients. To better characterize the role of gut microbiota in the development of ITP and to verify the modulating effect of BBR on gut microbiota structure, we performed colonization of mice with specific bacterium and established active murine models (immunized-splenocyte engraftment) of BBR treatment.


We integrated the sequencing data into an existing gut microbial reference-gene catalog and identified 35275 genes that differ in abundance between ITP patients and healthy controls. We then clustered the genes into metagenomic species (MGS) and finally identified 15 MGS which were significantly different in both discovery cohort and validation cohort. Dysbiosis was detected in the gut microbiome of ITP patients, as both phylogenetic analysis and MGS annotation indicated that Lachnospiraceae bacterium, Clostridium asparagiforme were over-represented while Bacteroides spp was depleted in ITP patients comparing with healthy controls. Functionally, KEGG annotation showed that the most enriched orthologs in ITP patients were related to membrane transport. Moreover, the biosynthesis of microbe-associated molecular patterns (MAMPs) such as lipopolysaccharides (LPS), peptidoglycan biosynthesis and flagellin were highly abundant in patients. Gene biomarkers and cluster markers based on gut microbiome were established to identify ITP patients and were validated in an independent cohort. The alterations of gut microbiome in ITP patients are partly reversed after treatment. Furthermore, L. bacterium shows more abundant in corticosteroid-resistant ITP comparing with newly diagnosed ITP. Specifically, BBR treatment could improve the microbial dysbiosis of corticosteroid-resistant ITP patients, the complete response (CR), response(R), and overall response (OR) rates being 26.3%, 47.4% and 73.7%, respectively. Targeted QPCR assay determined that L. bacterium accumulated in corticosteroid-resistant ITP, consistent with the result of shotgun sequencing data. Gavage with L. bacterium results in significantly alterations of gut microbiota structure in mice comparing with mice without bacterial administration or those receiving Clostridium asparagiforme administration. Moreover, colonization of L. bacterium caused more severe thrombocytopenia and impaired the response to corticosteroid therapy in active ITP model. Intriguingly, BBR treatment, but not any other antibiotics, could reverse the effect of L. bacterium colonization on gut microbiota structure and enhance the response to corticosteroid therapy.


Our findings demonstrate that the gut microbiome alters in ITP and is partly normalized after treatment. Gut microbiota dysbiosis may contribute to the development of corticosteroid-resistant ITP. BBR may correct corticosteroid-resistance by modulating the gut microbiota structure, thus being a novel potential second-line candidate to treat ITP.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.