Search in the Abstract Database

Abstracts Search 2019

P834 Proteus is a key candidate in the pathogenesis of Crohn’s disease: mucosa, stool genomics and functional analysis: the ENIGMA study

J. Zhang*1,2, E. Berendsen3, E. Hoedt3, Q. Liu1,2, F. Zhang1,2, Z. Xu1,2, A. Hamilton4,5, A. W. O’ Brien4,5, J. Ching1, J. J. Sung1, M. A. Kamm4,5, M. Morrison3, J. Yu1,2, S. C. Ng1

1The Chinese University of Hong Kong, Medicine & Therapeutics, Hong Kong, Hong Kong, 2The Chinese University of Hong Kong, Institute of Digestive Disease, Hong Kong, Hong Kong, 3The University of Queensland Diamantina Institute, Faculty of Medicine, Brisbane, Australia, 4St Vincent's Hospital, Department of Gastroenterology, Melbourne, Australia, 5The University of Melbourne, Department of Medicine, Melbourne, Australia


Proteus, Gram-negative facultative anaerobic bacilli, has recently been identified as a key genus in Crohn’s disease (CD) recurrence after intestinal resection. In this study, we investigated the role of Proteus as a gut pathogen in mediating inflammation in Crohn’s disease.


54 pairs of faecal samples and 80 colonic samples (61 CD patients; 19 healthy controls) were collected. The abundance of Proteus were determined by quantitative PCR. Proteus was isolated from faeces and biopsies of CD patients by selection culture in conditional agar and confirmed by Proteus specific target sequencing. To study effects of isolated Proteus, we established an in vitro microbe-enterocyte co-culture system by using two normal epithelial cells INT407, NCM460 and two CRC cell lines CaCo2 and HT29. Pathogenic function of Proteus was determined by in vivo mouse models and in vitro cell assays. Bacterial invasion ability was measured by fluorescence staining and confocal microscopy. Intracellular gene expression profiles and regulated pathways in normal cells treated with or without Proteus were analysed by RNA seq and KEGG analysis.


We confirmed the presence of Proteus in the gut and stool. The prevalence of Proteus in faecal samples was higher in CD patients compared with healthy controls (p < 0.05). Levels of Proteus were significantly increased in CD biopsies compared with control tissue. Amongst 24 Proteus-monoclones isolated from faeces and biopsy of CD patients, all of them belonged to members of P. mirabilis lineages. Proteus gavaged mice showed a shortened colon length compared with mice treated with E. coli 1655 (5.97 cm vs. 7.15 cm; p < 0.05). Mice depleted of bacteria and exposed to Proteus and DSS showed significantly higher severity of inflammation on HE staining. Compared with the cells co-cultured with E. coli 1655 or cultured in medium only (showed normal phenomenon), 70–80% of cells co-cultured with Proteus were unhealthy (rounded and detached) or dead. Increased necrotic cells were found in four cell lines co-cultured with Proteus due to bacterial invasion. Moreover, Proteus stimulated the production of pro-inflammatory cytokines including IL-18 (p < 0.001) and IL-1α (p < 0.01) in co-cultured cells (INT407 and NCM460). In keeping with this, Proteus induced key pro-inflammatory pathways including NOD-like receptor signalling (p < 0.001), Jak-STAT signalling (p < 0.01) and MAPK signalling pathways (p < 0.01) identified by RNA sequencing.


Our results show P. mirabilis, a urease producing organism in the gut, is associated with CD and can induce inflammation in cell lines and animal models of colitis. We contend that P. mirabilis and related species may act as a pathobiont and thereby play a critical role in the pathogenesis of CD.