Xenogeneic modulation of the ClpCP protease of Bacillus subtilis by a phage-encoded adaptor-like protein

Abstract

Like eukaryotic and archaeal viruses, which coopt the host's cellular pathways for their replication, bacteriophages have evolved strategies to alter the metabolism of their bacterial host. SPO1 bacteriophage infection of Bacillus subtilis results in comprehensive remodeling of cellular processes, leading to conversion of the bacterial cell into a factory for phage progeny production. A cluster of 26 genes in the SPO1 genome, called the host takeover module, encodes for potentially cytotoxic proteins that specifically shut down various processes in the bacterial host, including transcription, DNA synthesis, and cell division. However, the properties and bacterial targets of many genes of the SPO1 host takeover module remain elusive. Through a systematic analysis of gene products encoded by the SPO1 host takeover module, here we identified eight gene products that attenuated B. subtilis growth. Of the eight phage gene products that attenuated bacterial growth, a 25-kDa protein called Gp53 was shown to interact with the AAA+ chaperone protein ClpC of the ClpCP protease of B. subtilis Our results further reveal that Gp53 is a phage-encoded adaptor-like protein that modulates the activity of the ClpCP protease to enable efficient SPO1 phage progeny development. In summary, our findings indicate that the bacterial ClpCP protease is the target of xenogeneic (dys)regulation by a SPO1 phage-derived factor and add Gp53 to the list of antibacterial products that target bacterial protein degradation and therefore may have utility for the development of novel antibacterial agents.

Keywords: ATP-dependent protease; Bacillus; bacteria; bacteriophage; chaperone.

Figure 1.

SPO1 host takeover module genes that attenuate B. subtilis growth. A, schematic of the SPO1 host takeover module. The molecular masses (kilodaltons) of the individual gene products are shown above each gene in bold, and operons are indicated by dotted lines. The predicted positions of promoters are shown as arrows, indicating the direction of transcription. B, schematic of the experimental procedure used to identify SPO1 host takeover module gene products that attenuate the growth of B. subtilis. O/N, overnight. C, graph showing the A₆₀₀ values of B. subtilis cultures at 5 h of growth in the presence of IPTG, which induces expression of the individual host takeover module genes. Gene products shown in red displayed 50% or more attenuation compared with control cells expressing GFP. D, graphs showing growth curves (red) of B. subtilis cultures expressing SPO1 host takeover module genes that attenuated growth 50% or more and that of control cultures (see key). E, graphs showing growth curves (red) of B. subtilis cultures expressing the individual operons of the SPO1 host takeover module and that of control cultures (see key). The lag time preceding growth and growth rate (μ) of B. subtilis cultures expressing SPO1 host takeover module gene product(s) is shown in the bottom panels in D and E. Error bars in C–E represent S.E. (n = 3). Statistical analyses were performed by one-way ANOVA (**, p < 0.01; ***, p < 0.001).

Figure 2.

Gp53 interacts with the ClpC ATPase of the ClpCP protease in B. subtilis. A, bar chart comparing the efficacy of growth attenuation of a culture of B. subtilis either expressing N-terminal His₆-tagged Gp53 (red) or untagged Gp53 (gray). B, schematic of the pulldown assay used to identify the bacterial target(s) of Gp53. C, a representative image of an SDS-PAGE gel showing results of the pulldown assay with Gp53 and whole-cell extracts (WCL) of B. subtilis. The band specifically enriched in reactions containing immobilized Gp53 is indicated by an arrowhead in lane 3. D, a representative image of an SDS-PAGE gel showing results of the pulldown assay with purified Gp53 and N-terminal FLAG-tagged ClpC. The migration positions of Gp53 and ClpC are indicated. E, bar chart showing the results from the bacterial two-hybrid interaction assay with ClpC and mutant variants of Gp53. The ClpC-binding activity of the Gp53 mutants as a percentage of WT Gp53 activity is indicated. Error bars in A and E represent S.E. (n = 3). Statistical analyses were performed by one-way ANOVA (ns, not significant; ***, p < 0.001).

Figure 3.

Gp53 stimulates the ATPase activity of ClpC and competes with the B. subtilis adaptor protein MecA for binding to ClpC. A, schematic showing how the ATP hydrolysis and adaptor protein mediated formation of the functional ClpCP protease in B. subtilis (adapted from Molière et al. (32)). B, graph showing the amount of ATP hydrolyzed (P_i release, micromolar) as a function of time by ClpC (0.2 μm) alone and in the presence of different amounts of Gp53 (0.2, 0.4, and 1 μm). Numerical ATPase rates are shown on the right. C, bar chart showing results from the ATPase assay (as in B) in which ClpC (50 nm) was incubated with equimolar amounts of MecA (reaction I), Gp53 (reaction II), or MecA and Gp53 (added to the reaction in different orders, reactions III and IV). The amount of P_i released (micromolar) is expressed as -fold change with respect to the reaction with ClpC alone, i.e. its basal ATPase activity. D, bottom panel, bar chart showing the results from the modified bacterial two-hybrid interaction assay to demonstrate that Gp53 competes with MecA for binding ClpC. Top panel, the assay setup (see text for details). E, bottom panel, bar chart showing the results from the bacterial two-hybrid assay demonstrating binding of Gp53 or MecA to different domains of ClpC (as shown in the top panel). In B–E, error bars represent S.E. (n = 3). Statistical analyses were performed by one-way ANOVA (ns, not significant; **, p < 0.01; *** p < 0.001).

Figure 4.

Gp53 modulates the specificity of the ClpCP protease in B. subtilis. A, representative images of SDS-PAGE gels of in vitro degradation of MecA and Gp53 by ClpCP protease. The intensities of the bands corresponding to MecA or Gp53 are shown in the graph relative to the intensity of the ClpP band in the corresponding lanes. The migration positions of ClpC (1 μm), MecA (1 μm), Gp53 (1 μm), and ClpP (1 μm) are indicated. Pyruvate kinase (PK, 20 ng/ml) and phosphoenolpyruvate (4 mm) were used as an ATP generation system. B, as in A, but equimolar amounts of MecA and Gp53 were added together. C, as in A, but the in vitro degradation assays were conducted in the presence of 3 μm β-casein and in the absence of MecA or Gp53. D, as in C, but the in vitro degradation assays were conducted in the presence of MecA. E, as in C, but the in vitro degradation assays were conducted in the presence of Gp53. F, as in C, but the in vitro degradation assays were conducted in the presence of MecA and Gp53. G, as in C, but the in vitro degradation assays were conducted with McsA/B (1 μm each) in the absence and presence of Gp53. In A–G, the same color coding is used in the schematics, gels, and graphs to aid data interpretation. H, left panel, a log/log plot comparing the accurate mass and retention time of peptides in whole-cell extracts of B. subtilis containing pHT08-Gp53 expressing Gp53 upon induction by IPTG and control whole-cell extracts of B. subtilis containing pHT08-Gp53 to which no IPTG was added. A paired t test (p < 0.05) was carried out to identify peptides that had a -fold change in abundance of 2 or more and that lie on or outside of the diagonal outer green lines. Right panel, as in the left graph, but whole cell-extracts of B. subtilis with and without IPTG added were compared, which demonstrated that the change in peptide abundance was specific to the presence of Gp53.

Figure 5.

Compromised ClpCP protease activity affects the efficacy of SPO1 development in B. subtilis. A, graph showing the growth curves of WT, ΔclpC (IH25), clpC DWB (IH140), and clpC-loop (IH217) B. subtilis cultures. B, graph showing the optical density as a function of time of a culture of exponentially growing WT and ΔclpC B. subtilis cells following infection with SPO1 at A₆₀₀ 0.2. C, as in B, but with WT, clpC DWB, and clpC-loop B. subtilis cells. Error bars in A–C represent S.E. (n = 3).