Fig 1.
Model structures of ClpB, DnaK and the molecular dynamic simulation.
(A) The overall ClpB monomer structure, comprised of an N-domain (magenta), nucleotide binding domain-1 (NBD-1) (light blue), NBD-2 (yellow) and M-domain shown (blue). The image was generated using the UCSF Chimera program based on the T. thermophilus ClpB. (B) The model structure of DnaK is comprised of an N-terminal NBD (red) containing four subdomains, IA, IB, IIA and IIB, and a substrate-binding domain (SBD) (orange). The image was generated using the UCSF Chimera program based on the E. coli DnaK. (C) The best-docked complex of ClpBFt-DnaKFt. The DnaK subdomains IB and IIB are in contact with the ClpB M domain. (D) An average complex structure of ClpBFt-DnaKFt during 100 ns molecular dynamic (MD) simulations. Important hydrogen bonding interactions between ClpB residues (black) and DnaK residues (light blue) observed during the entire MD simulation are highlighted.
Fig 2.
Characterization of M-domain variants of ClpB with regard to disaggregation and ATPase activity.
(A) MDH disaggregation activities of M-domain variants of ClpB were monitored by loss of turbidity in the presence of ATP and the co-chaperones DnaK, DnaJ and GrpE (KJE) of F. novicida, as described in Materials and Methods. The initial MDH turbidity was set as 100% and data were calculated compared to the denatured MDH and shown as percentage of disaggregation. At least three independent experiments were performed and data with mean ± SD are shown. (B) ClpB-mediated refolding activities of urea-denatured luciferase were determined in the presence of ATP and the co-chaperones KJE of F. novicida, as described in Materials and Methods. Refolding results in an increase of fluorescence over time. The KJE control refers to co-chaperones only (no ClpB). The initial fluorescence was set as zero and data of at least three independent experiments with mean ± SD are shown. (C) Heat shock survival of indicated F: novicida strains upon heat shock. Bacteria were exposed to 50°C for 30 min and the mean ± SD CFU are indicated. The wild-type strain U112 did not exhibit any significant killing during the treatment, and the value was set as 100%. Sign + indicates trans complemented strains. A significant difference in the bacterial numbers of mutant strains vs. U112 is indicated as follows: *** P < 0.001; ** P < 0.01; * P < 0.05; NS (not significant) P > 0.05. (D) ATPase activity of wild-type and indicated M-domain variants of ClpB were determined in the absence or presence of α-casein (10 mM). Basal ATPase activity of wild-type ClpB was set at 1.0. At least three independent experiments were performed and data with mean ± SD are shown.
Fig 3.
Characterization of M-domain variants of ClpB with regard to intracellular replication, substrate secretion and heat shock survival.
(A) The wild-type strain U112, ΔclpB, or ΔclpB expressing M-domain variants of clpB variants in trans or inserted in cis on the chromosome, were used to infect J774A.1 cells. Infected cells were lysed at 0 h and 24 h and the number of CFU were determined. The net growth mean values ± SEM of at least three independent experiments are shown. A significant difference in the bacterial numbers of mutant strains vs. U112 is indicated as follows: ** P < 0.01; NS (not significant) P > 0.05. (B) Analysis of T6S by bacterial strains. Indicated strains were grown at 37°C to an OD of 1.5 in TSB medium supplemented with 5% KCl. Precipitated supernatants or pellets of the same strain were separated by SDS-PAGE and analyzed using Western blot analysis and anti-IglC antiserum. The signal intensity of each band on scanned images of supernatant samples was measured using the Image J program (http://rsbweb.nih.gov/ij/) and the signal of each strain is presented as a percentage of the band-intensity of the U112 strain, the latter set as 100%. The in-frame deleted ΔiglE mutant of F. novicida U112 was used as negative control. At least three independent experiments were performed and a representative image with the average band intensity percentage is shown. (C) Survival of indicated strains upon heat shock. Bacteria were exposed to 50°C for 30 min and the (mean ± SD) CFU are indicated. The wild-type strain U112 did not exhibit any significant killing during the treatment, and the value was set as 100%. A significant difference in the bacterial numbers of mutant strains vs. U112 is indicated as follows: *** P < 0.001; ** P < 0.01; NS (not significant) P > 0.05 for cis-complemented strains vs. U112. (D) After subcutaneous inoculation with 1 × 103 CFU of the indicated F. novicida strains, mice were sacrificed on day 3, and bacterial burdens (log10 CFU/ml) in liver were determined. The mean ± SEM for six mice per group is indicated. A significant difference in the bacterial numbers of mutant strains vs. U112 is indicated as follows: *** P < 0.001; NS (not significant) P > 0.05.
Fig 4.
Characterization of Walker and Arginine finger motif mutants of F. novicida.
(A) The wild-type strain U112, ΔclpB or ΔclpB expressing various ClpB variants of NBD-1 (WA1, WB1 or Arg1), NBD-2 (WA2, WB2 or Arg2) or both NBDs (WA1-2, WB1-2 or Arg1-2) in cis were used to infect BMDM. The specific mutations were as follows WA1: K212A, WB1: E279A, Arg1: R332A, WA2: K613A, WB2: E680A, and Arg2: R757A. Infected cells were lysed at 0 h and 24 h and the number of CFU were determined. The net growth mean values ± SEM of at least three independent experiments are shown. A significant difference in the bacterial numbers of mutant strains vs. U112 is indicated as follows: ** P < 0.01; * P < 0.05; NS (not significant) P > 0.05 (B) Survival of indicated strains upon heat shock. Bacteria were exposed to 50°C for 30 min and mean ± SD CFU are shown. The wild-type strain U112 did not exhibit any significant killing during the treatment, and this value was set as 100%. A significant difference in the bacterial numbers of mutant strains vs. U112 is indicated as follows: *** P < 0.001; NS (not significant) P > 0.05. (C) Analysis of T6S of bacterial strains. Indicated strains were grown at 37°C to an OD of 1.5 in TSB medium supplemented with 5% KCl. Supernatants were collected, filter sterilized and TCA precipitated. Precipitated supernatants or pellets of the same strain were separated by SDS-PAGE and analyzed using Western blot analysis with an anti-IglC antiserum. The signal intensity of each band was measured as described in Fig 3B and percentage of the band-intensity vs U112 (set as 100%) is presented. At least three independent experiments were performed and a representative image is shown. (D) After subcutaneous inoculation with 1 × 103 CFU of the indicated F. novicida strains, mice were sacrificed on day 3, and bacterial burdens (log10 CFU/ml) in liver were determined. The mean ± SEM for six mice per group is indicated. A significant difference in the bacterial numbers of mutant strains vs. U112 is indicated as follows: *** P < 0.001. (E) ATPase activity of wild-type ClpB, NBDs variants, and the N-terminal truncated (ΔNClpB) variant of ClpB was determined in the absence or presence of α-casein (10 mM). Basal ATPase activity of wild-type ClpB was set at 1.0. Experiments were conducted in triplicate and mean ± SD are shown.
Fig 5.
Sequence alignment and structural comparison of the N-terminals of V. cholerae ClpV and VipB with F. tularensis ClpB and IglB.
(A). ClpV and ClpB sequences were retrieved from NCBI (https://www.ncbi.nlm.nih.gov/) and sequence alignments were performed using MAFFT (https://mafft.cbrc.jp/alignment/server/), and the corresponding image was generated using the web server ESPript 3 (http://espript.ibcp.fr). The first 174 amino acids (156 for Francisella ClpB) of the N-terminal domain of the ClpV-ClpB alignment is shown. Secondary structures as predicted for V. cholerae ClpV are displayed above the alignment. (B). The VipB and IglB sequences were retrieved from NCBI (https://www.ncbi.nlm.nih.gov/), and essentially the same procedure for alignment was performed as aforementioned. The first 110 amino acids, including the N-terminal domain (1–95 aa), of the VipB-IglB alignment is shown. Secondary structure elements as predicted for V. cholerae VipB are displayed above the alignment. The VipB α-helix known to interact with the ClpV N-terminal is boxed and highlighted in yellow. Conserved consensus sequence residues that contribute to the interaction with the ClpV-N-terminal are boxed and colored in yellow in helix 1. Identical amino acids are highlighted with red color. (C). Ribbon view of the F. tularensis IglB (IglB in pink; PDB: 3j9o) superimposed on the V. cholerae VipB (VipB in blue; PDB: 5mxn).
Fig 6.
The role of the N-terminal for the ability of F. novicida ClpB to support T6S and virulence in mice.
(A) Domain organization of F. novicida ClpB. The protein consists of an N-terminal (N) domain (magenta), two NDB domains (NDB-1, turquoise and NBD-2, yellow), and an inserted middle (M) domain (blue). The M domain contains four alpha-helices that are numbered accordingly. At the domain boundaries, the amino acid positions are indicated. (B) Heat shock survival of indicated strains. Bacteria were exposed to 50°C for 30 min and the mean ± SD CFU are indicated. The wild-type strain U112 did not exhibit any significant killing during the treatment, and was set as 100%. A significant difference in the bacterial numbers of mutant strains vs. U112 is indicated as follows: *** P < 0.001; * P < 0.05; NS (not significant) P > 0.05. (C) Analysis of T6S by bacterial strains. Indicated strains were grown at 37°C to an OD of 1.5 in TSB medium supplemented with 5% KCl. Precipitated supernatants or pellets of the same strain were separated by SDS-PAGE and analyzed using Western blot analysis using anti-IglC antiserum. At least three independent experiments were performed and a representative image is shown. The signal intensity of each band was measured as described in Fig 3B and the percentage of the band-intensity vs U112 (set as 100%) is presented. (D) Indicated strains were used to infect J774A.1 cells. Infected cells were lysed at 0 h and 24 h and the number of CFU were determined. The net growth mean values ± SEM of at least three independent experiments are shown. A significant difference in the bacterial numbers of mutant strains vs. U112 is indicated as follows: * P < 0.05; NS (not significant) P > 0.05. (E) After subcutaneous inoculation with 1 × 103 CFU of the indicated F. novicida strains, mice were sacrificed on day 3, and bacterial burdens (log10 CFU/ml) in liver were determined. The mean ± SEM for six mice per group is indicated. A significant difference in the bacterial numbers of mutant strains vs. U112 is indicated as follows: *** P < 0.001; NS (not significant) P > 0.05.
Fig 7.
E. coli ClpB phenotypically complements F. novicida ClpB.
(A) Heat shock survival of indicated strains. Bacteria were exposed to 50°C for 30 min and the mean ± SD CFU is shown. The wild-type strain U112 did not exhibit any significant killing during the treatment, and the value was set as 100%. A significant difference in the bacterial numbers of mutant strains vs. U112 is indicated as follows: *** P < 0.001; NS (not significant) P > 0.05. (B) Analysis of T6S by bacterial strains. Indicated strains were grown at 37°C to an OD of 1.5 in TSB medium supplemented with 5% KCl. Precipitated supernatants, or pellets of the same strain were separated by SDS-PAGE and analyzed using Western blot analysis with an anti-IglC antiserum. At least three independent experiments were performed and a representative image is shown. The signal intensity of each band was measured as described in Fig 3B and the percentage of the band-intensity vs U112 (set as 100%) is presented. (C) After subcutaneous inoculation with 1 × 103 CFU of the indicated F. novicida strains, mice were sacrificed on day 3, and bacterial burdens (log10 CFU/ml) in liver were determined. The mean ± SEM for six mice per group is indicated. A significant difference in the bacterial numbers of mutant strains vs. U112 is indicated as follows: *** P < 0.001; NS (not significant) P > 0.05.