The bile salt glycocholate induces global changes in gene and protein expression and activates virulence in enterotoxigenic Escherichia coli

Abstract

Pathogenic bacteria use specific host factors to modulate virulence and stress responses during infection. We found previously that the host factor bile and the bile component glyco-conjugated cholate (NaGCH, sodium glycocholate) upregulate the colonization factor CS5 in enterotoxigenic Escherichia coli (ETEC). To further understand the global regulatory effects of bile and NaGCH, we performed Illumina RNA-Seq and found that crude bile and NaGCH altered the expression of 61 genes in CS5 + CS6 ETEC isolates. The most striking finding was high induction of the CS5 operon (csfA-F), its putative transcription factor csvR, and the putative ETEC virulence factor cexE. iTRAQ-coupled LC-MS/MS proteomic analyses verified induction of the plasmid-borne virulence proteins CS5 and CexE and also showed that NaGCH affected the expression of bacterial membrane proteins. Furthermore, NaGCH induced bacteria to aggregate, increased their adherence to epithelial cells, and reduced their motility. Our results indicate that CS5 + CS6 ETEC use NaGCH present in the small intestine as a signal to initiate colonization of the epithelium.

Figure 1

Modulation of ETEC transcriptome in the presence of bile or NaGCH. (a) Heat map and hierarchical clustering of 61 genes differentially expressed (>3-fold change) in the presence of bile or NaGCH in the CS5 + CS6 ETEC strains E1777 and E2265. Color scale represents relative expression levels: green indicates low level (downregulated genes), red indicates high level (upregulated genes), and black indicates no change. (b) The 10 genes most significantly upregulated (red) or downregulated (green) by bile or NaGCH, including KEGG pathways of the corresponding genes. Values are the average fold change for E1777 and E2265.

Figure 2

Quantitative proteomic analysis of ETEC isolates (E2265 and E1777) treated with 0.2% NaGCH. Graph shows fold change in protein abundance after NaGCH exposure; fold change <0.8 indicates a significant decrease, fold change >1.2 indicates a significant increase. Proteins were organized with GraphPad Prism 7.00 for Mac OS X based on their respective pathways.

Figure 3

Validation of RNA-Seq and proteomic data by quantitative real-time RT-PCR. mRNA from ETEC isolates E1777 and E2265 was extracted after 3 hours or 24 hours (ON) of growth in LB, LB + 0.15% bile, or LB + 0.2% NaGCH. Relative mRNA expression values are the mean plus standard deviation (error bars) of at least three independent experiments. Asterisks indicate significant difference by one-way ANOVA (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001) using GraphPad Prism 7.00 for Mac OS X.

Figure 4

Expression of virulence-associated genes in globally distributed CS5 + CS6 ETEC strains. mRNA from 10 CS5 + CS6 ETEC isolates was extracted after 3 hours or 24 hours (ON) of growth in LB, LB + 0.15% bile, or LB + 0.2% NaGCH. Quantitative real-time RT-PCR was used to analyze differential gene expression. Relative mRNA expression values are the mean plus standard deviation (error bars) of three independent experiments. Asterisks indicate significant difference by one-way ANOVA (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001) using GraphPad Prism 7.00 for Mac OS X.

Figure 5

Linear representation of plasmid p1 from ETEC E2265. PacBio sequencing and analysis of the largest plasmid in E2265 (142359 bp, GenBank Accession number CP02334) with virulence-associated genes that were induced by bile and NaGCH (red arrows). Magenta arrows, ORFs; gray arrows, insertion sequences; blue arrows, virulence genes not induced by bile and NaGCH; and black arrows, Tra (transfer) genes.

Figure 6

Bile and NaGCH induce bacterial co-aggregation mediated by the CF CS5. (a) Cell–cell aggregation and (b) settling patterns of ETEC E2265 and its isogenic mutants (E2265 ΔCS5 and E2265 ΔCS6) after growth in the presence of 0.15% bile or 0.2% NaGCH. Liquid suspensions of bacterial cells were left standing for 24 hours, and the OD₆₀₀ of planktonic bacteria was measured at the times indicated and the mean and standard deviation (error bars) were plotted. Comparisons between control (E2265 in LB) with E2265 + 0.15% bile, E2265 + 0.2% NaGCH, E2265 ΔCS6 + 0.15% bile, and E2265 ΔCS6 + 0.2% showed significant differences (P < 0.001).

Figure 7

Role of NaGCH and CFs in epithelial cell adherence. Adherence assays were performed for 3 hours in DMEM with 10% of FBS (“medium”) with or without 0.2% NaGCH. Percent bacteria adhered to Caco-2 cells for (a) 5 CS5 + CS6 ETEC strains (E88, E1111, E1724, E1777, and E1779) and (b) E2265 and its isogenic mutants (E2265 ΔCS5 and E2265 ΔCS6). Assays were performed in triplicate, and graphs show the mean and standard deviation (error bars). The percentage of adhered bacteria was calculated as described in Materials and Methods. Asterisks indicate significant difference by one-way ANOVA (*P < 0.05; ****P < 0.0001) using GraphPad Prism 7.00 for Mac OS X.

Figure 8

Schematic overview of NaGCH signaling during ETEC infection. (a) In the proximal part of the small intestine, ETEC meets a high concentration of NaGCH. This downregulates bacterial motility but activates virulence, inducing bacteria–bacteria adherence and favoring bacteria–host interactions and attachment. (b) We speculate that when bacteria progress through the gut and NaGCH is reabsorbed and its concentration decreases, activating bacterial motility to disperse the cells occurs.