The present study uses protected specimen brushings of the colonic mucosa as the sampling method, reducing potential contamination from luminal contents [40]. This unique approach to studying the innate microbiota yields findings more likely represent the true burden of these species within the outer layer of the MGL than faecal samples or whole mucosal biopsies, traditionally used for study of the microbiota, each of which have been shown to harbour distinct microbial communities [40, 41].
The primary aim of this study was to perform quantitative analysis of R. hominis, a species known to produce substantial amounts of butyrate in vitro [47]. This species widely colonised the MGL in health and UC, indicating that this is an ubiquitous member of the innate microflora. Its’ abundance was reduced in patients with acute UC compared to health and QUC. This reduction was observed in all four colonic regions examined, supporting the hypothesis that UC is associated with a global dysbiosis and the concept of spatial homogeneity along the longitudinal axis of the colon [43, 48]. An inverse correlation between R. hominis abundance and inflammatory cell infiltrates was demonstrated, indicating that depletion of this species is a feature of inflammation in the colitic colon.
In keeping with these data, reductions in the abundance of this species, as well as an inverse correlation with disease severity, have been reported in UC in previous studies based on faecal sampling [28, 32]. However, studies of mucosal biopsies yielded conflicting results [49, 50]. Willing et al reported a reduction in R. hominis in patients with ileal CD, but not UC [49], while Lepage et al reported no reduction in the species in mucosal biopsies of twins with UC [50]. These findings, combined with the observation by Machiels et al that a stable reduction occurred in the faeces of patients with quiescent disease, led to the hypothesis that alterations in Roseburia abundance may play a role in the study of the transient microbiota, rather than the adherent microbiota [28]. The current study challenges this theory, suggesting that the reduction in R. hominis contributes to the dysbiosis of the MGL adherent microbiota in the inflamed colon, with abundances returning towards healthy levels when disease is in remission. R. hominis has been shown to have anti-inflammatory mechanisms in vitro [51]. It is tempting to hypothesise that this is related to the production of butyrate however, the authors acknowledge that this cannot be inferred from compositional data alone.
This study also aimed to determine colonisation patterns of H2S producing species. Desulfovibrio remains the most studied of these bacteria and to date a consensus has not been reached as to their precise role in UC [25, 46, 52, 53]. Few studies analysed other low-abundance microbes such as D. desulfuricans and D. propionicus, both of which are capable of foraging sulfate and producing H2S [24]. No data currently exist pertaining to the involvement of these species in UC. This study demonstrated low colonisation rates of D. desulfuricans and D. propionicus in both health and UC. 40% of healthy subjects were positive for D. curvatus in the current study, in keeping with data from Gibson et al., who reported hydrogenotropic species in 50% of faecal samples [54]. On the other hand, it is at odds Nava et al who detected both species in mucosal biopsies of 25 healthy subjects [55]. It should also be noted that in the aforementioned study, the species Desulfovibrio accounted for the majority of hydrogenotropic present. Previous work from our laboratory demonstrated a high rate of colonisation with Desulfovibrio [46] in these patient samples. Desulfovibrio may be the dominant species of sulphate reducing bacteria colonising this patient cohort also. Given the low colonisation rates of D. desulfuricans and D. proprionicus in UC, there is no compelling evidence of an association between these species and UC in the present study.
B. wadsworthia was present at detectable levels in the majority of individuals in this study. These data demonstrate a significant reduction in B. wadsworthia in patients with AC compared to healthy individuals and those with QUC. The reduced abundance in patients with AC compared to healthy individuals remained significant after normalisation against total bacterial counts. A positive association between B. wadsworthia abundance and total bacterial counts in AC was also observed. The inverse correlation between B. wadsworthia abundance and inflammatory score confirm that a reduction in abundance of this species is a feature of the inflamed colon. Although these findings are not in keeping with the hypothesis of increased activity of hydrogenotropic bacteria and reduced butyrate in UC, the finding is nonetheless significant. Taken together, these data suggest a tentative link between this species and dysbiosis in AC.
In this study, normalisation of data against total bacterial copy number was performed to reduce potential reporting errors by minimising the effect of between-sample variation and taking the efficiency of the quantification procedure into account for both species investigated. These data are more representative of the actual burden of the target in the MGL. However, normalisation negated the significant difference in abundance observed between health and acute UC when analysis was based on raw copy numbers at each of the four colonic regions examined. This may be due to the fact that individually, these species account for a small proportion of the overall bacterial load at any given site in the colon.
This study investigated the microbiological basis of UC as an energy deficiency disease, with specific focus on butyrogenic and hydrogenotropic bacteria at the level of the MGL in the colon. In keeping with the hypothesis that UC is associated with reduced butyrate concentrations, reductions in the abundance of R. hominis were noted. However reduced abundance of the hydrogenotropic B. wadsworthia was also observed in the colitic colon in this study. These data suggest that the reported lack of butyrate in the colitic colon may be a result of a lack of butyrogenic bacteria, rather than bacterial inhibition by microbial by-products such as H2S. Furthermore, altered abundances of these species are associated with active inflammation in UC. Overall, these data suggest that reductions in these species are a feature of the altered microbial signature of the MGL in UC.
Limitations
The authors acknowledge that the current study has some limitations. It is possible that the use of bowel preparation may have resulted in a loss of some loosely adherent microbes in the patients undergoing colonoscopy (healthy controls and quiescent cohorts), however, this would not account for the low colonisation rates in patients with acute UC, as this cohort did not receive bowel preparation prior to surgery. This would be difficult to control for, as adequate bowel preparation is a prerequisite for successful colonoscopy. This study did not determine whether the observed alterations in microbes equated to alterations in butyrate levels in the colon. Further functional studies are warranted to validate these findings. Finally, as with many studies of the microbiota, the authors acknowledge that it is not possible to determine from these data whether the observed changes are a potential driving force behind the pathogenesis of inflammation, or arising as a result of the inflammatory process. The association between these microbial changes and UC warrants further investigation, with a view to establishing whether a cause-effect relationship indeed exists, and whether manipulation of the dysbiosis in UC will have clinical applications in the future.