Search

Citrobacter rodentium infection is a mouse model for the important human diarrheal infection caused by enteropathogenic E. coli (EPEC). The pathogenesis of both species is very similar and depends on their unique ability to form intimately epithelium-adherent microcolonies, also known as “attachment/effacement” (A/E) lesions. These microcolonies must be dynamic and able to self-renew by continuous re-infection of the rapidly regenerating epithelium. It is unknown whether sustained epithelial A/E lesion pathogenesis is achieved through re-infection by planktonic bacteria from the luminal compartment or local spread of sessile bacteria without a planktonic phase. Focusing on the earliest events as C. rodentium becomes established, we show here that all colonic epithelial A/E microcolonies are clonal bacterial populations, and thus depend on local clonal growth to persist. In wild-type mice, microcolonies are established exclusively within the first 18 hours of infection. These early events shape the ongoing intestinal geography and severity of infection despite the continuous presence of phenotypically virulent luminal bacteria. Mechanistically, induced resistance to A/E lesion de-novo formation is mediated by TLR-MyD88/Trif-dependent signaling and is induced specifically by virulent C. rodentium in a virulence gene-dependent manner. Our data demonstrate that the establishment phase of C. rodentium pathogenesis in vivo is restricted to a very short window of opportunity that determines both disease geography and severity. [ABSTRACT FROM AUTHOR]

Soon after birth the mammalian gut microbiota forms a permanent and collectively highly resilient consortium. There is currently no robust method for re-deriving an already microbially colonized individual again-germ-free. We previously developed the in vivo growth-incompetent E. coli K-12 strain HA107 that is auxotrophic for the peptidoglycan components D-alanine (D-Ala) and meso-diaminopimelic acid (Dap) and can be used to transiently associate germ-free animals with live bacteria, without permanent loss of germ-free status. Here we describe the translation of this experimental model from the laboratory-adapted E. coli K-12 prototype to the better gut-adapted commensal strain E. coli HS. In this genetic background it was necessary to complete the D-Ala auxotrophy phenotype by additional knockout of the hypothetical third alanine racemase metC. Cells of the resulting fully auxotrophic strain assembled a peptidoglycan cell wall of normal composition, as long as provided with D-Ala and Dap in the medium, but could not proliferate a single time after D-Ala/Dap removal. Yet, unsupplemented bacteria remained active and were able to complete their cell cycle with fully sustained motility until immediately before autolytic death. Also in vivo, the transiently colonizing bacteria retained their ability to stimulate a live-bacteria-specific intestinal Immunoglobulin (Ig)A response. Full D-Ala auxotrophy enabled rapid recovery to again-germ-free status. E. coli HS has emerged from human studies and genomic analyses as a paradigm of benign intestinal commensal E. coli strains. Its reversibly colonizing derivative may provide a versatile research tool for mucosal bacterial conditioning or compound delivery without permanent colonization. [ABSTRACT FROM AUTHOR]

The intestinal ecosystem is formed by a complex, yet highly characteristic microbial community. The parameters defining whether this community permits invasion of a new bacterial species are unclear. In particular, inhibition of enteropathogen infection by the gut microbiota ( = colonization resistance) is poorly understood. To analyze the mechanisms of microbiotamediated protection from Salmonella enterica induced enterocolitis, we used a mouse infection model and large scale highthroughput pyrosequencing. In contrast to conventional mice (CON), mice with a gut microbiota of low complexity (LCM) were highly susceptible to S. enterica induced colonization and enterocolitis. Colonization resistance was partially restored in LCM-animals by co-housing with conventional mice for 21 days (LCMcon21). 16S rRNA sequence analysis comparing LCM, LCMcon21 and CON gut microbiota revealed that gut microbiota complexity increased upon conventionalization and correlated with increased resistance to S. enterica infection. Comparative microbiota analysis of mice with varying degrees of colonization resistance allowed us to identify intestinal ecosystem characteristics associated with susceptibility to S. enterica infection. Moreover, this system enabled us to gain further insights into the general principles of gut ecosystem invasion by non-pathogenic, commensal bacteria. Mice harboring high commensal E. coli densities were more susceptible to S. enterica induced gut inflammation. Similarly, mice with high titers of Lactobacilli were more efficiently colonized by a commensal Lactobacillus reuteriRR strain after oral inoculation. Upon examination of 16S rRNA sequence data from 9 CON mice we found that closely related phylotypes generally display significantly correlated abundances (co-occurrence), more so than distantly related phylotypes. Thus, in essence, the presence of closely related species can increase the chance of invasion of newly incoming species into the gut ecosystem. We provide evidence that this principle might be of general validity for invasion of bacteria in preformed gut ecosystems. This might be of relevance for human enteropathogen infections as well as therapeutic use of probiotic commensal bacteria. [ABSTRACT FROM AUTHOR]