Enterohemorrhagic Escherichia coli virulence regulation by two bacterial adrenergic kinases, QseC and QseE

Abstract

The human pathogen enterohemorrhagic Escherichia coli (EHEC) O157:H7 has two histidine sensor kinases, QseC and QseE, which respond to the mammalian adrenergic hormones epinephrine and norepinephrine by increasing their autophosphorylation. Although QseC and QseE are present in nonpathogenic strains of E. coli, EHEC exploits these kinases for virulence regulation. To further investigate the full extent of epinephrine and its sensors' impact on EHEC virulence, we performed transcriptomic and phenotypic analyses of single and double deletions of qseC and qseE genes in the absence or presence of epinephrine. We showed that in EHEC, epinephrine sensing seems to occur primarily through QseC and QseE. We also observed that QseC and QseE regulate expression of the locus of enterocyte effacement (LEE) genes positively and negatively, respectively. LEE activation, which is required for the formation of the characteristic attaching and effacing (A/E) lesions by EHEC on epithelial cells, is epinephrine dependent. Regulation of the LEE and the non-LEE-contained virulence factor gene nleA by QseE is indirect, through transcription inhibition of the RcsB response regulator. Finally, we show that coincubation of HeLa cells with epinephrine increases EHEC infectivity in a QseC- and QseE-dependent manner. These results genetically and phenotypically map the contributions of the two adrenergic sensors QseC and QseE to EHEC pathogenesis.

Fig 1

Confirmation of nonpolar deletion and rescue of expression of the adrenergic kinase-encoding genes qseC and qseE. (A) Summary of the QseC- and QseE-dependent signaling cascade involved in virulence regulation as reported prior to this work. Genes whose expression had been shown to be affected by epinephrine have Φ next to them. Asterisks indicate that the ler promoter is highly regulated by many transcription factors, including GrlA, Pch, GadE, QseA, and H-NS (2, 6, 25, 30, 67). epi, epinephrine; NE, norepinephrine; AE, attaching and effacing. (B) qRT-PCR analysis examining qseC and qseE expression in the wt and ΔqseC, ΔqseE, and ΔqseC ΔqseE mutant strains and the complemented double mutant strains grown to an OD₆₀₀ of 1.0 in low-glucose DMEM. The genes' transcript levels were quantified as fold differences normalized to wt gene transcription levels. The samples' rpoA transcript levels were used as internal controls to normalize the output C_T values. The data are from at least three independently grown replicates.

Fig 2

Global analysis of QseC's and QseE's effects on EHEC O157 gene transcription. Venn diagrams show the number of overlapping downregulated (A) and upregulated (B) genes between the ΔqseC, ΔqseE, and ΔqseC ΔqseE mutant strains compared to the wt. (C) Venn diagram indicating genes that are decreased in the ΔqseC strain and increased in the ΔqseE strain. (D) Venn diagram indicating genes with expression that is increased in the ΔqseC strain and decreased in the ΔqseE strain. Strains for the microarrays were grown to an OD₆₀₀ of 1.0 in low-glucose DMEM.

Fig 3

Both QseC and QseE regulate the LEE genes and nleA. qRT-PCR analyses of tir (LEE5) (A), eae (LEE5) (B), espA (LEE4) (C), and nleA (D) transcription. The mRNA levels for all of these genes were quantified and normalized to the mRNA levels of the endogenous internal control gene, rpoA. The mRNA levels were graphed as fold changes compared to wt transcript levels. The results are from at least three independent samples.

Fig 4

Effect of epinephrine on QseC- and QseE-dependent regulation of LEE and non-LEE genes. Expression of espA (LEE4) (A) and nleA (B) was evaluated by qRT-PCR in the wt strain and the mutant strains grown to the late exponential phase in the absence and presence of epinephrine (final concentration, 50 μM). The error bars indicate standard deviations of the ΔΔC_T values. The levels of endogenous rpoA mRNA were used to normalize the C_T values. (C) Representation of the converse regulation of the LEE genes and nleA transcription by the epinephrine-sensing kinases QseC and QseE. Although both kinases regulate the LEE genes and nleA, epinephrine-dependent regulation of the LEE genes is mostly via QseC (dotted arrow with α), while epinephrine-dependent regulation of nleA is mostly via QseE (dotted line with β).

Fig 5

Deletion of the two adrenergic kinases QseC and QseE impairs epinephrine-dependent regulation of multiple EHEC virulence factors. Shown are heat maps from microarray analysis representing the effects of epinephrine (Epi) on the wt and ΔqseC, ΔqseE, and ΔqseC ΔqseE mutant strains. The strains treated with epinephrine were compared to the same strains with no treatment. Red indicates upregulation, green indicates downregulation, and black indicates no change. (A) Heat map representing differential regulation of the LEE genes. (B) Heat map showing the differential expression of non-LEE genes.

Fig 6

QseE regulates nleA and the LEE genes through its inhibition of rcsB transcription. (A) Transcription (qRT-PCR) of rcsB in the wt and ΔqseC, ΔqseE, and ΔqseC ΔqseE mutant strains. (B) Confirmation by qRT-PCR of the deletion and rescue in expression of rcsBi. (C) Transcriptional LEE gene expression for the wt strain, ΔrcsB mutant, and its complement. (D) qRT-PCR evaluating the transcription of nleA in the wt and ΔrcsB mutant. Error bars indicate the standard deviations of the ΔΔC_T values. The mRNA levels of endogenous rpoA were used to normalize the C_T values. (E) Representation of how the inhibition of the expression of the LEE genes and nleA by QseE is indirect via RcsB. RcsB, whose transcription is inhibited by QseE, is a transcriptional activator of the LEE genes and nleA.

Fig 7

Fluorescent actin staining (FAS) assays. HeLa cells were infected for 6 h in the absence or presence of epinephrine (final concentration, 50 μM). HeLa cell actin was stained green with FITC-phalloidin, while HeLa cell nuclei and bacteria were stained red with propidium iodide. Formation of pedestals was visualized as bright green (actin) cups holding red bacterial cells. The experiments were performed in duplicate at least three times. For every slide, at least 100 cells were evaluated. (A) Visualization of pedestals formed by bacteria on HeLa cells. (B) Representation of the percentage of infected HeLa cells.

Fig 8

Motility regulation is QseC dependent but QseE independent. (A) Tryptone motility plates with the wt strain and the ΔqseC, ΔqseE, ΔqseC ΔqseE mutants and their complemented strains. (B) Representation of the diameter of the bacterial halos. (C) β-Galactosidase assays were performed using plasmid pVS177 with an fliC::lacZ promoter fusion in the wt and ΔqseC, ΔqseE, and ΔqseC ΔqseE mutant strains. (D) Representation indicating the QseC-dependent and QseE-independent activation of motility genes. QseC phosphorylates QseB, which directly binds to the regulatory region of flhDC, encoding the master regulators of flagella, leading to increase fliC expression, production of flagella, and motility (10).

Fig 9

Model of the QseC and QseE regulatory cascade. Solid lines with arrows indicate confirmed positive interactions, while dotted lines indicate indirect or unconfirmed direct interactions. QseC phosphorylates QseB, which directly activates transcription of flhDC to promote expression of flagella. Through phosphorylation of KdpE, QseC activates expression of the LEE genes. QseF is phosphorylated by both QseC and QseE. QseF indirectly activates expression of espFu and Shiga toxin. QseE inhibits rcsB transcription in an as yet undetermined manner. Given that RcsB activates expression of the LEE and nleA, QseE inhibition of rcsB inhibits LEE and nleA expression. How QseC influences nleA expression is unknown. epi, epinephrine; NE, norepinephrine; AE, attaching and effacing; HUS, hemolytic-uremic syndrome.