There are approximately 10^14 bacteria belonging to nearly 1000 different species in the human gastrointestinal (GI) tract that co-exist with host cells. Within the GI tract, signaling molecules secreted by both eukaryotic and prokaryotic cells are abundant. Recent studies have shown that both bacteria and human cells recognize and respond to the signals from each other, presumably to gain a competitive advantage. The cross-recognition of signals is known as Inter-kingdom (IK) signaling and this phenomenon is considered to be important in the onset of infections in the GI tract. Of the eukaryotic signaling molecules present in the GI tract, the neuroendocrine hormone norepinephrine (NE) is considered to be important in the context of infections as NE is produced at very high concentration in the intestine under post traumatic stress, is known to increase bacterial virulence and infection, and has also been shown to be a potent chemoattractant for GI tract pathogens such as enterohemorrhagic E. coli (EHEC). The focus of this study is on elucidating the mechanisms underlying the recognition and chemotaxis of bacteria towards NE. While chemotaxis has been typically investigated in the context of bacteria moving towards a metabolizable source (e.g., amino acids), chemotaxis is potentially important in the onset of infections in the human GI tract. In this study we use a microfluidic plug assay to investigate the receptor and mechanism utilized by a model bacterium Escherichia coli in its chemotactic response to NE. A series mutant of E. coli RP437 strains of knockouts for four MCP-encoding genes was used in this study. The results from the microfluidic plug assay were then confirmed quantitatively by capillary assay. We have shown that Tsr receptor is necessary for chemotaxis of NE for E. coli RP437, and attraction of E. coli towards NE may require an additional receptor. Results from the priming experiments suggest that exposure to NE may result in the de novo expression of co-receptor(s) that are crucial to chemotaxis towards NE. The requirement for high cell density also suggests the possibility that NE per se may not be an attractant for E. coli, but could be a precursor that is modified into a chemoattractant by cells. These results are expected to further our understanding of bacterial chemotaxis and its role in bacterial colonization and infection of the human GI tract.

The overall goal of this research was to understand the role of inter-kingdom signaling in enterohemorrhagic Escherichia coli (EHEC) infections of the human gastro-intestinal (GI) tract from the perspective of both the invading pathogen and the human intestinal epithelial cells, which they colonize. Differential gene expression of EHEC was studied upon exposure to the human neuroendocrine hormones epinephrine and norepinephrine. We determined that these hormones increase EHEC chemotaxis, motility, biofilm formation, colonization of host cells, and virulence gene expression. We also studied the EHEC response to the GI tract commensal bacterial signaling molecules indole and autoinducer-2 (AI-2). We observed that indole decreases all the EHEC phenotypes that are increased by the human hormones and represses EHEC virulence. However, the effect of AI-2 was similar to that observed with hormones and opposite to that observed with indole, i.e. AI-2 increases EHEC virulence phenotypes. We studied changes in host cell transcriptome in the presence of the commensal bacterial signal indole. Indole increases expression of genes involved in tight junction and gap junction formation, and production of mucins and actin cytoskeleton genes. Indole also down-regulates genes encoding for pro-inflammatory cytokines, chemokines, and Toll-like receptors. The gene expression results were confirmed with phenotypic assays where we observed an increase in trans-epithelial resistance, increase in the anti-inflammatory cytokine IL-10, decrease in the pro-inflammatory cytokine IL-8, decrease in the activity of the pro-inflammatory transcription factor NF-κB, and decrease in colonization by EHEC of the indole-pre-treated HCT-8 cells. We established that factors secreted by epithelial cells are important determinants of EHEC virulence. Gene expression studies showed that 34 out of 41 LEE virulence genes were induced when EHEC was cultured in conditioned medium. In addition, the data showed increased expression of the shiga toxin-2 prophage 933W. These changes in gene expression were corroborated by a 5-fold increase in HCT-8 cell colonization and increased intracellular Stx2 phage titers. We determined that the HCT-8-secreted factor(s) was protein-based and that it was greater than 3 kDa in size. In conclusion, we have characterized the pathogen response to various eukaryotic and prokaryotic GI tract signals. We have established, for the first time, that the commensal bacterial signal indole is an inter-kingdom signal for the host epithelial cells. Overall, our studies provide a greater understanding of host-pathogen interactions.