Inactivation of the dnaK gene in Clostridium difficile 630 Δerm yields a temperature-sensitive phenotype and increases biofilm-forming ability

Abstract

Clostridium difficile infection is a growing problem in healthcare settings worldwide and results in a considerable socioeconomic impact. New hypervirulent strains and acquisition of antibiotic resistance exacerbates pathogenesis; however, the survival strategy of C. difficile in the challenging gut environment still remains incompletely understood. We previously reported that clinically relevant heat-stress (37-41 °C) resulted in a classical heat-stress response with up-regulation of cellular chaperones. We used ClosTron to construct an insertional mutation in the dnaK gene of C. difficile 630 Δerm. The dnaK mutant exhibited temperature sensitivity, grew more slowly than C. difficile 630 Δerm and was less thermotolerant. Furthermore, the mutant was non-motile, had 4-fold lower expression of the fliC gene and lacked flagella on the cell surface. Mutant cells were some 50% longer than parental strain cells, and at optimal growth temperatures, they exhibited a 4-fold increase in the expression of class I chaperone genes including GroEL and GroES. Increased chaperone expression, in addition to the non-flagellated phenotype of the mutant, may account for the increased biofilm formation observed. Overall, the phenotype resulting from dnaK disruption is more akin to that observed in Escherichia coli dnaK mutants, rather than those in the Gram-positive model organism Bacillus subtilis.

Figure 1

Validation of C. difficile 630 Δerm::dnaK 723a mutant by PCR screening and Southern blotting. Lanes for each gel/experiment were loaded with PCR products as follows: M, 1 kb Plus DNA ladder (Invitrogen); Lane 1, C. difficile 630 Δerm; Lane 2, dnaK mutant; Lane 3, pMTL007C-E2 plasmid DNA; Lane 4; negative control (water). (a) PCR across the intron-exon junction using EBS universal and Cdi-dnaK-R primers generated a 428 bp product from dnaK mutant (lane 2) showing presence of the intron; (b) Southern blot analysis to confirm single genomic insertion of the intron: An intron-specific probe for the ErmRAM was hybridised to HindIII-digested: genomic DNA extracted from C. difficile 630 Δerm (Lane 1), pMTL007C-E2 plasmid DNA (Lane 2, positive control), and genomic DNA from the dnaK mutant (Lane 3). (c) Additional confirmatory PCR: (i) PCR using Cdi-dnaK-F and Cdi-dnaK-R primers generated a 210 bp product from C. difficile 630 Δerm (lane 1), whereas the dnaK mutant produced a 2059 bp product, indicating the insertion of the group II intron (lane 2); (ii) PCR using ErmRAM-F and ErmRAM-R primers generated a 900 bp product from the dnaK mutant (lane 2) indicative of splicing out of the td group I intron, whereas unmodified pMTL007C-E2 template generated a 1300 bp product (lane 3), (iii) PCR across the other intron-exon junction using ErmRAM-R and Cdi-dnaK-F primers generated a 1300 bp product from the dnaK mutant only (lane 2). These experiments confirm insertion of the group II intron into the C. difficile 630 Δerm chromosome at the desired site and in the correct orientation, resulting in dnaK inactivation.

Figure 2

Growth of C. difficile 630Δerm (◆) and C. difficile 630 Δerm::dnaK 723a mutant (□) in BHIS broth at different temperatures. Temperature shifts were induced at early exponential phase, 4 h. (a) When grown at 37 °C, the dnaK mutant exhibited a temperature-sensitive phenotype, growing more slowly than C. difficile 630 Δerm. (b) Cells grown to early exponential phase at 37 °C and then transferred to 30 °C grew in a comparable manner. Cells grown to early exponential phase at 37 °C were challenged by transfer to temperatures of (c) 41 °C and (d) 45 °C, respectively, where temperature sensitivity of the dnaK mutant was more pronounced. D_650nm values are plotted on a logarithmic scale and are averages of D_650nm measurements from biological triplicate cultures; error bars represent the standard error of mean.

Figure 3

Motility of C. difficile strains in BHIS agar (0.175%). (a) C. difficile 630 Δerm, (b) dnaK mutant. Motility was visualised as a diffuse spreading pattern from the point of stab inoculation.

Figure 4

Electron microscopic analysis of C. difficile 630 Δerm and C. difficile 630 Δerm::dnaK 723a mutant. (a) Transmission electron microscopy image of C. difficile 630 Δerm. (b) Transmission electron microscopy image of dnaK mutant. Arrows indicate flagellar filaments. (c) Scanning electron microscopy image of C. difficile 630 Δerm. (d) Scanning electron microscopy image of dnaK mutant. The images depict the filamentous phenotype of the dnaK mutant in comparison to the wild-type.

Figure 5

Expressional changes in class 1 chaperone genes and the flagellar filament gene, fliC, in the C. difficile 630 Δerm::dnaK 723a mutant. RNA was extracted and reverse transcribed from biological duplicate cultures and cDNA was quantified in technical triplicate qPCR reactions. The ‘calibrator normalised relative quantification including efficiency correction’ experimental mode assessed gene expression using the tpi gene, whose expression did not change, as a reference. Bars represent average fold-changes in gene expression in the dnaK mutant compared with the Δerm parental strain. Error bars represent standard deviation of the mean.

Figure 6

Biofilm-forming ability of C. difficile 630 Δerm and C. difficile 630 Δerm::dnaK 723a mutant. Biofilm assays were performed in biological triplicates, each with 6 independent technical replicates. Strains were classified as strong- (A ₅₇₀ > 1), moderate- (A ₅₇₀ = 0.5−1), or weak- (A ₅₇₀ < 0.5) biofilm producers. P values represent statistical comparison (Student’s t-test, 2 tailed) between BHIS broth and BHIS broth with 0.9% (w/v) additional glucose.