Article Text
Statistics from Altmetric.com
We read with interest the recent review by Tilg et al,1 which summarised the role of microbiota in liver diseases and pointed out that a causal link with systemic inflammation has still not been established. This letter fills in this gap and provides an analysis of the circulating microbiota in portal vein as the link between gut and liver. The access to portal circulation is possible during the implantation of a transjugular intrahepatic portosystemic shunt (TIPS). Therefore, we characterised the circulating microbiome in portal vein (first venous outflow in gut–liver axis), liver outflow, central venous blood and peripheral venous blood from seven patients with decompensated liver cirrhosis receiving TIPS for either variceal bleeding (n=3) or refractory ascites (n=4) (mean Model for End-stage Liver Disease (MELD) 8.4 (range 6–13), Child-Pugh-Score (CHILD) A: n=4, CHILD B: n=3) (figure 1A). We performed 16S ribosomal RNA (rRNA) gene sequencing of buffy coat samples and identified 65 genera belonging to four phyla (predominantly Proteobacteria) in this cohort (online supplementary figure 1 and figure 1B). Blood microbiome phylum compositions identified in our study agreed with previous findings investigating the peripheral blood microbiome in buffy coat samples from patients with liver fibrosis2 as well as healthy individuals3 but differed from the gut microbiome measured in faecal samples, where Bacteroidetes and Firmicutes are predominant.2
Supplemental material
Although overall bacterial community structure did not show a compartment-specific clustering as assessed by Bray-Curtis dissimilarity measures (figure 1C), the abundance of several genera varied strongly in circulation forming compartment-specific patterns (figure 1D, using DESeq2 for differential abundance analysis4). To verify whether the bacteria observed in 16S rRNA gene sequencing were viable in blood, we performed aerobic and anaerobic cultivation of blood from the four sites of five additional TIPS patients. Three out of five patients showed positive cultivation with Staphylococcus and one that showed bacterial growth of Acinetobacter, both abundant in the sequencing data (online supplementary figure 1). These results demonstrate that circulating microbiota are indeed viable.
Bacterial infections are frequent in cirrhotics, often trigger acute-on-chronic liver failure and are associated with high mortality.5 A number of cytokines, which might show immune dysfunction, systemic inflammation and oxidative stress, have been linked to decompensation, acute-on-chronic liver failure and mortality.6–10
To elaborate on the relationship with systemic inflammation, we measured cytokine levels in serum of the same patients and compartments except central venous blood. Inflammatory cytokines formed patient-specific clusters (figure 2A), and their individual measurements showed robust associations with the abundance of blood microbiome genera measured by Spearman correlation (figure 2B), which establish the association of circulating microbiota with systemic inflammation.
To the best of our knowledge, this study is the first to track the major part of microbiome of portal venous blood through liver into central venous blood and circulating into peripheral blood. With this study, we provide snapshots of the circulating blood microbiome, its compartment-specific patterns, viability of the microbial members and their association with inflammation. Further studies with larger cohorts are required to better understand the role of circulating microbiome and expand the knowledge about gut microbiome contribution to liver diseases reviewed recently.1
Acknowledgments
The authors would like to thank Gudrun Hack and Silke Bellinghausen for excellent technical assistance.
Footnotes
30 RS and CA-S share first authorship.
JT and AM share last authorship.
Contributors Conceptualisation: RS, CA-S, JT and MA; methodology: RS, CA-S, GTH, EM, JT and MA; formal analysis: RS, CA-S, JT and MA; investigation, RS, CA-S, MSAM, CCK, CM, DT, FEU, FM, CJ, AP, MP and GTH; resources: CCK, CM, DT, FEU, FM, CJ, AP, MP, GTH and EL; data curation: BL; writing – original draft: RS, CA-S, EM, EL, BL, JT and MA; visualisation: RS, CA-S, JT and MA; supervision: JT and MA.
Funding The authors were supported by grants from the Deutsche Forschungsgemeinschaft (SFB TRR57), Cellex Foundation, Novo Nordisk Foundation (NNF10CC1016515 and NNF16CC0020896) and European Union’s Horizon 2020 research and innovation program (No 668031).
Disclaimer The funders had no influence on study design, data collection and analysis, decision to publish or preparation of the manuscript.
Patient consent Obtained.
Ethics approval The ethics committee of the University Clinic Bonn in agreement with the Declaration of Helsinki permitted the study (No. 295/16).
Provenance and peer review Not commissioned; internally peer reviewed.
Author note RS and CA-S share first authorship. JT and MA share last authorship.
Correction notice This article has been corrected since it published Online First. The first authorship statement has been added and funding statement updated.