Transcriptional inhibitor of virulence factors in enteropathogenic Escherichia coli

Abstract

The type III secretion system (TTSS) is a key virulence mechanism of many important gram-negative bacterial pathogens. The TTSS is conserved among different bacterial pathogens, and mutations and deletions to the system significantly decrease virulence, making the TTSS an important potential therapeutic target. We have developed a high-throughput assay to search for inhibitors of the TTSS. We screened a commercial library of 20,000 small molecules for their ability to inhibit type III secretion by enteropathogenic Escherichia coli (EPEC). After discarding compounds that had no effect on secretion, inhibited bacterial growth, and/or caused degradation of EPEC-secreted proteins, the search was focused on a class of compounds that, while not direct inhibitors of type III secretion, inhibit expression of TTSS-related genes and other genes involved in virulence. This class of compounds does not affect bacterial viability or motility, indicating that it is not significantly affecting the expression of essential genes and is specific to virulence-associated genes. Transcriptional fusion assays confirmed that virulence-associated promoters were more sensitive to inhibition by this class of compounds. Overall, we have identified a class of compounds that can be used as a tool to probe the mechanism(s) that regulates virulence gene expression in EPEC.

FIG. 1.

Flowchart of assays to screen libraries of small molecules for their ability to inhibit type III secretion by EPEC.

FIG. 2.

Structure of compounds from the Maybridge library of small molecules discussed in this study. Compound 1 decreased production of type III system-associated and other virulence-associated proteins, compound 2 had no observed effect on type III secretion or viability, compound 3 affected bacterial viability, and compounds 4 to 6 have the salicylideneaniline substructure and act like compound 1.

FIG. 3.

Screen for inhibitors of type III secretion by EPEC with no change in bacterial viability. (A) Graph of ELISA results monitoring percent EspB secretion compared to that in untreated EPEC. Bacteria were treated with the compounds at 40 μM for 3 h. The right panel graphs the viability of EPEC as measured by OD₆₀₀. The results are shown ± standard deviation and are the average of three different experiments performed in triplicate. (B) Western blot for secreted proteins Tir and EspB in wt EPEC, ΔescN (type III apparatus mutant that does not secrete Tir or EspB), and wt EPEC treated with the compounds as described above. (C) Total secreted protein stained by Sypro-Ruby (Bio-Rad).

FIG. 4.

Compound 1 does not induce the breakdown of type III secreted proteins. To determine if compound 1 affects the stability of the type III secreted proteins or the autotransporter EspC, bacterial supernatant containing secreted proteins from wild-type EPEC was incubated with 40 μM compound. (A) The amount of EspB was determined by ELISA. The results are shown ± standard deviation and are the average of three different experiments performed in triplicate. (B) The amounts of Tir and EspB were determined by Western blot analysis, and the amount of EspC was monitored by Sypro-Ruby protein stain (Bio-Rad).

FIG. 5.

Compound 1 reduced production of both LEE-encoded EPEC-secreted proteins and type III apparatus components but not the production of non-virulence-associated proteins. Wild-type EPEC was grown in DMEM for 5 h with 40 μM compound 1 (wt + 1) or DMSO. The ΔescN, ΔescJ, and ΔescC type III secretion apparatus mutants were grown in DMEM for 5 h with the corresponding volume of DMSO. ΔC/C-HSV (short for ΔescC/pescC-HSV) is the escC type III apparatus mutant complemented in trans with a plasmid containing HSV-tagged escC under the control of a tetracycline resistance gene promoter and was grown in DMEM for 5 h with 40 μM compound 1 or DMSO. The bacteria were harvested by centrifugation and lysed in SDS sample buffer. The levels of Tir, EspB, DnaK, DnaJ, MBP, EscJ, EscC, and EscC-HSV were determined by Western blot analysis.

FIG. 6.

Virulence-associated promoters are more sensitive to transcriptional inhibition by compound 1. Wild-type EPEC cells carrying the indicated transcriptional fusions (Table 1) were grown in DMEM at 37°C in 5% CO₂ for 8 h with the indicated amount of compound 1 (C1) or DMSO. Chloramphenicol acetyltransferase (CAT) microplate assays and protein determinations to calculate specific activity were performed and graphed as a percentage of CAT specific activity compared to that in untreated EPEC for each transcriptional fusion. The results are shown ± standard deviation and are the average of at least three different experiments performed in duplicate.

FIG. 7.

Compound 1 does not affect the motility or the level of flagellin of EPEC. (A) Strains of bacteria were stabbed into motility agar containing 40 μM compound 1 (+ 1) or compound 2 (+ 2) and grown for 18 h at 37°C. The positive control is a motile strain of EAEC, and the negative control is a nonmotile strain of EAEC. (B) Wild-type EPEC, the ΔescN type III mutant, or the ΔfliC flagellin filament subunit mutant were grown in DMEM for 5 h with 40 μM compound or DMSO as indicated. The bacteria were harvested by centrifugation and lysed in SDS sample buffer to yield the bacterial pellet, and the culture supernatant yielded the secreted proteins. The level of flagellin in the samples was determined by Western blot analysis.