Long-term changes of bacterial and viral compositions in the intestine of a recovered Clostridium difficile patient after fecal microbiota transplantation

Abstract

Fecal microbiota transplantation (FMT) is an effective treatment for recurrent Clostridium difficile infections (RCDIs). However, long-term effects on the patients' gut microbiota and the role of viruses remain to be elucidated. Here, we characterized bacterial and viral microbiota in the feces of a cured RCDI patient at various time points until 4.5 yr post-FMT compared with the stool donor. Feces were subjected to DNA sequencing to characterize bacteria and double-stranded DNA (dsDNA) viruses including phages. The patient's microbial communities varied over time and showed little overall similarity to the donor until 7 mo post-FMT, indicating ongoing gut microbiota adaption in this time period. After 4.5 yr, the patient's bacteria attained donor-like compositions at phylum, class, and order levels with similar bacterial diversity. Differences in the bacterial communities between donor and patient after 4.5 yr were seen at lower taxonomic levels. C. difficile remained undetectable throughout the entire timespan. This demonstrated sustainable donor feces engraftment and verified long-term therapeutic success of FMT on the molecular level. Full engraftment apparently required longer than previously acknowledged, suggesting the implementation of year-long patient follow-up periods into clinical practice. The identified dsDNA viruses were mainly Caudovirales phages. Unexpectedly, sequences related to giant algae-infecting Chlorella viruses were also detected. Our findings indicate that intestinal viruses may be implicated in the establishment of gut microbiota. Therefore, virome analyses should be included in gut microbiota studies to determine the roles of phages and other viruses-such as Chlorella viruses-in human health and disease, particularly during RCDI.

Keywords: recurrent infection of the gastrointestinal tract.

Figure 1.

Analysis of fecal microbiota. (A) Sample description. Fecal samples of the donor and the patient were collected at the indicated time points and subjected to metagenomic and/or 16S rRNA gene sequencing. FMT, fecal microbiota therapy. (B) Bacterial compositions at the phylum level are shown as pie charts. The inlay graph shows bacterial diversities inferred by Shannon indices as bars. (C) Bacterial compositions at the class, order, and family levels (from top to bottom) are shown as stacked bar graphs. Only taxa supported by at least 1% of total reads at each level are shown. (D) Relative abundances of the five most dominant bacterial genera in samples D0, D4, and P4 are shown as bar graphs. Dialister and Faecalibacterium genera were solely represented by the indicated species in each sample. The respective family names are given in parentheses. (E) Viral compositions of Caudovirales families are shown as pie charts.