AUTHOR=Zhang Jian , Qin Hao , Chang Mingyu , Yang Yang , Lin Jun 

TITLE=Gut Microbiota Dysbiosis in BK Polyomavirus-Infected Renal Transplant Recipients: A Case-Control Study

JOURNAL=Frontiers in Cellular and Infection Microbiology

VOLUME=12

YEAR=2022

URL=https://www.frontiersin.org/journals/cellular-and-infection-microbiology/articles/10.3389/fcimb.2022.860201

DOI=10.3389/fcimb.2022.860201

ISSN=2235-2988

ABSTRACT=<sec><title>Background</title><p>BK polyomavirus infection results in renal allograft dysfunction, and it is important to find methods of prediction and treatment. As a regulator of host immunity, changes in the gut microbiota are associated with a variety of infections. However, the correlation between microbiota dysbiosis and posttransplant BK polyomavirus infection was rarely studied. Thus, this study aimed to characterize the gut microbiota in BK polyomavirus-infected renal transplant recipients in order to explore the biomarkers that might be potential therapeutic targets and establish a prediction model for posttransplant BK polyomavirus infection based on the gut microbiota.</p></sec><sec><title>Methods</title><p>We compared the gut microbial communities of 25 BK polyomavirus-infected renal transplant recipients with 23 characteristic-matched controls, applying the 16S ribosomal RNA gene amplicon sequencing technique.</p></sec><sec><title>Results</title><p>At the phylum level, <italic>Firmicutes</italic>/<italic>Bacteroidetes</italic> ratio significantly increased in the BK polyomavirus group. <italic>Bacteroidetes</italic> was positively correlated with CD4/CD8 ratio. In the top 20 dominant genera, <italic>Romboutsia</italic> and <italic>Roseburia</italic> exhibited a significant difference between the two groups. No significant difference was observed in microbial alpha diversity. Beta diversity revealed a significant difference between the two groups. Nine distinguishing bacterial taxa were discovered between the two groups. We established a random forest model using genus taxa to predict BK polyomavirus infectious status, which achieved the best accuracy (80.71%) with an area under the curve of 0.82. Two genera were included in the best model, which were <italic>Romboutsia</italic> and <italic>Actinomyces</italic>.</p></sec><sec><title>Conclusions</title><p>BK polyomavirus-infected patients had gut microbiota dysbiosis in which the <italic>Firmicutes</italic>/<italic>Bacteroidetes</italic> ratio increased in the course of the viral infection. Nine distinguishing bacterial taxa might be potential biomarkers of BK polyomavirus infection. The random forest model achieved an accuracy of 80.71% in predicting the BKV infectious status, with <italic>Romboutsia</italic> and <italic>Actinomyces</italic> included.</p></sec>