P. McLellan1, A. Lavelle1, L. Brot1, M. Straube1, L. de Sordi1, J.P. Grill1, A. Bourrier2, J. Kirchgesner2, J.M. Chatel3, L. Beaugerie2, H. Sokol1, P. Seksik1
1INSERM, Pierre and Marie Curie University, CRSA UMRS_938 Team Microbiota Intestine and Inflammation, Paris, France, 2Saint Antoine Hospital, Department of Gastroenterology, Paris, France, 3INRA, Micalis, Jouy-enJosas, France
Background
The intestinal microbiota of patients with Crohn’s disease (CD) is characterised by a specific dysbiosis, including the loss of Faecalibacterium prausnitzii (F. prau), a major bacterium with anti-inflammatory properties. MAM protein (microbiotia anti-inflammatory molecule) produced by F. prau has recently been identified as partially carrying the anti-inflammatory activity of this bacterium. The description of microbiota for diagnostic purposes in CD is currently difficult to achieve. MAM could be a biomarker of dysbiosis. Our goal was to study the expression of MAM in the intestinal ecosystem and to evaluate its interest as a biomarker.
Methods
An in silico approach was used to determine primers of MAM gene sequences specific to two currently known phylogroups (I and II) of F. prau. Faecal samples were collected from a monocentric cohort of CD patients in flare (n = 24) and in remission (n = 24) and from healthy controls (n = 12). To measure MAM expression in vivo in the microbiota, we extracted total RNA and DNA from each sample and performed quantitative PCR (qPCR) with MAM primers (I and II), F. prau primers and universal bacterial primers. We compared the relative expression (ΔCt) of MAM mRNA and DNA with qPCR to the overall amount of bacteria in faecal samples.
Results
MAM analyses of F. prau phylogroup II were the most informative. We observed that the relative expression of MAM mRNA was statistically lower in active patients 5.84 × 10−5 [95% CI 3.8 × 10−6–2.1 × 10−3] and in remission 5.36 × 10−5 [95% CI 2.5 × 10−8–1.9 × 10−2] compared with healthy controls 5.42 × 10−4 [95% CI 8.5 × 10−5–9.1 × 10−2], (p = 0.004). There was no difference in ΔCt between flare and remission in CD patients. The most relevant relative expression threshold was 1.6 × 10−4. At this threshold, sensitivity was 81%, specificity 76%, positive predictive value 94% and negative predictive value 47%. In multivariate analysis, the relative expression of MAM was an independently associated with the diagnosis of CD; OR 2.961 [95% CI 2.026–4.831] (p = 0.003). Using relative expression of MAM DNA, we found a significant decrease in the flare group compared with remission patients and controls. We found a negative correlation of MAM DNA expression with CRP (p = 0.0053) and HBI (p = 0.025).
Conclusion
Our results showed that the loss of MAM RNA expression is observed in CD patients. Moreover, MAM DNA detection was associated with CD activity parameters. This opens up the prospect of an ecological biomarker of CD based on the expression of molecules from the intestinal ecosystem. These results will need to be validated in larger independent cohorts and their specificities compared with other pathological situations.
