The MapZ-Mediated Methylation of Chemoreceptors Contributes to Pathogenicity of Pseudomonas aeruginosa

Abstract

The pathogenic bacterium Pseudomonas aeruginosa is notorious for causing acute and chronic infections in humans. The ability to infect host by P. aeruginosa is dependent on a complex cellular signaling network, which includes a large number of chemosensory signaling pathways that rely on the methyl-accepting chemotaxis proteins (MCPs). We previously found that the second messenger c-di-GMP-binding adaptor MapZ modulates the methylation of an amino acid-detecting MCP by directly interacting with a chemotaxis methyltransferase CheR1. The current study further expands our understanding of the role of MapZ in regulating chemosensory pathways by demonstrating that MapZ suppresses the methylation of multiple MCPs in P. aeruginosa PAO1. The MCPs under the control of MapZ include five MCPs (Aer, CtpH, CptM, PctA, and PctB) for detecting oxygen/energy, inorganic phosphate, malate and amino acids, and three MCPs (PA1251, PA1608, and PA2867) for detecting unknown chemoattractant or chemorepellent. Chemotaxis assays showed that overexpression of MapZ hampered the taxis of P. aeruginosa toward chemoattractants and scratch-wounded human cells. Mouse infection experiments demonstrated that a dysfunction in MapZ regulation had a profound negative impact on the dissemination of P. aeruginosa and resulted in attenuated bacterial virulence. Together, the results imply that by controlling the methylation of various MCPs via the adaptor protein MapZ, c-di-GMP exerts a profound influence on chemotactic responses and bacterial pathogenesis.

Keywords: PilZ adaptor; Pseudomonas aeruginosa; bacterial pathogenesis; chemotaxis; cyclic di-GMP; methyl-accepting chemotaxis protein; methyltransferase cheR1.

Figure 1

CheR1-catalyzed methylation of eight MCPs. (A) Methylation of eight MCPs by CheR1 demonstrated by in vitro methyltransferase assay with the methylated MCPs visualized by radioautography. The membrane fraction-containing chemoreceptor and [³H] Ado-Met were used as the substrate and co-substrates for CheR1, respectively. The amount of membrane fraction was normalized according to the corresponding protein expression level. Data are representative of two independent experiments. (B) Schematic representation of the architecture of PctA with the ligand-binding domain located in the periplasm and the cytoplasmic portion comprised of a HAMP domain, methylated helices and a signaling domain. The MapZ-mediated inhibition of CheR1 by c-di-GMP is depicted along with the predicted methylation sites (E/Q) in the methylated helices. (C) Partial sequence alignment of the eight MCP substrates of CheR1 with the predicted methylation sites are shown in the frame. (D) In vitro methyltransferase assay shows the methylation of PctA, but not PctA_E576AE577A, by CheR1. The membrane fraction-containing MCP and [³H] Ado-Met were incubated with CheR1 before the methylated MCP was visualized by radioautography. SDS-PAGE gel (upper panel) shows that the amount of MCP proteins used in the assay were comparable. The methylated MCPs were visualized by radioautography (lower panel). Data are representative of two independent experiments.

Figure 2

Inhibition of CheR1 by c-di-GMP in methylating MCPs requires MapZ. Methylation of eight MCPs by CheR1 demonstrated by in vitro methyltransferase assay with the methylated MCPs visualized by radio-autography. Shown radio-autography images (A–C) are representative of two independent in vitro methyltransferase assay. The membrane fraction-containing chemoreceptor and [³H]-Ado-Met were used as the substrate and co-substrates for CheR1, respectively. The amount of membrane fraction was normalized according to the corresponding protein expression level.

Figure 3

Chemotaxis and energy taxis assays demonstrating that MapZ affects the taxis toward inorganic phosphate, malate, and energy source. (A) Capillary chemotaxis assay of P. aeruginosa strains attracted toward 10 mM inorganic phosphate (Pi). The data were the means of three replicates and were normalized with the number of bacteria that swam into buffer-containing capillaries (Two-tailed t-test, ^***P < 0.001). (B) Capillary chemotaxis assays of P. aeruginosa strains attracted toward 5 mM malate. The data were the means of three replicates and were normalized with the number of bacteria that swam into buffer-containing capillaries (Two-tailed t-test, ^****P < 0.0001). (C,D) Energy taxis of P. aeruginosa strains on the minimal medium plates that contained 50 mM glucose as the sole carbon and energy source. The area of the swimming zone was quantified and compared in (D). The data were the means of three replicates (Two-tailed t-test, ^*P < 0.05).

Figure 4

Chemotaxis-guided migration of P. aeruginosa strains toward scratch-wounded A549 human cells. (A) Representative microscopic images showing the accumulation of P. aeruginosa cells around wounded A549 human cells. The P. aeruginosa cells (in green) and wounded A549 cells (in white) are shown near the edge of the wound as indicated by the red dotted lines. The movies (Movies S1–S3) can be found in the supplementary material. Scale bar = 10 μm. (B) Quantitative comparison of the accumulation of P. aeruginosa cells around the injured A549 cells as indicated by fluorescence intensity. Three independent experiments were performed on each strain and at least 10 cells from each strain were used for quantitative analysis [Data are mean ± SD (n > 10)]. Two-tailed t-test, ^****P < 0.0001, ^*P < 0.05.

Figure 5

Overexpression of MapZ attenuated P. aeruginosa virulence. Survival rate of mice following intraperitoneal injection of 3.3 × 10⁶ CFU of PAO1/p and PBS (A), 3.3 × 10⁶ CFU of PAO1/p and PAO1/pMapZ strains (B), and 2.4 × 10⁶ CFU of PAO1/p and cheR1^D144AY222A strain (C). Data are representative of three independent experiments with five mice used for each group. Histology of liver after intraperitoneal infections. Liver was harvested at the end of 24 h infection and processed for paraffin inclusion. Sections were stained with haematoxylin and eosin staining of liver tissue from the Babl/C mice. All cultures of the strains have the same OD₆₀₀ at 0.1. (D) Representative photos for haematoxylin and eosin staining of bacteria-infected liver tissue from Babl/C mice after inoculation with PBS, PAO1/p, and PAO1/pMapZ for 24 h, respectively, PBS was used to be a negative control. Tissue infected with PBS can be seen a small amount of liver cells are mild edema around the central vein and edge with cell swelling and cytoplasm loose light dye. Tissue infected with PAO1/p is visible liver cells widely moderate edema with cell swelling and cytoplasm loose light dye, while tissue infected with mapZ_R13A/p, and PAO1/pMapZ only can be seen mild edema liver cells at part of the tissue and edge around the central vein with cell swelling and cytoplasm loose light dye. Data are representative of two independent experiments. Scale bar = 50 μm.