Function of the Borrelia burgdorferi FtsH Homolog Is Essential for Viability both In Vitro and In Vivo and Independent of HflK/C

Abstract

In many bacteria, the FtsH protease and its modulators, HflK and HflC, form a large protein complex that contributes to both membrane protein quality control and regulation of the cellular response to environmental stress. Both activities are crucial to the Lyme disease pathogen Borrelia burgdorferi, which depends on membrane functions, such as motility, protein transport, and cell signaling, to respond to rapid changes in its environment. Using an inducible system, we demonstrate that FtsH production is essential for both mouse and tick infectivity and for in vitro growth of B. burgdorferi FtsH depletion in B. burgdorferi cells resulted in membrane deformation and cell death. Overproduction of the protease did not have any detectable adverse effects on B. burgdorferi growth in vitro, suggesting that excess FtsH does not proteolytically overwhelm its substrates. In contrast, we did not observe any phenotype for cells lacking the protease modulators HflK and HflC (ΔHflK/C), although we examined morphology, growth rate, growth under stress conditions, and the complete mouse-tick infectious cycle. Our results demonstrate that FtsH provides an essential function in the life cycle of the obligate pathogen B. burgdorferi but that HflK and HflC do not detectably affect FtsH function.

Importance: Lyme disease is caused by Borrelia burgdorferi, which is maintained in nature in an infectious cycle alternating between small mammals and Ixodes ticks. B. burgdorferi produces specific membrane proteins to successfully infect and persist in these diverse organisms. We hypothesized that B. burgdorferi has a specific mechanism to ensure that membrane proteins are properly folded and biologically active when needed and removed if improperly folded or dysfunctional. Our experiments demonstrate that FtsH, a protease that fulfills this role in other microorganisms, is essential to B. burgdorferi viability. Cells depleted of FtsH do not survive in laboratory culture medium and cannot colonize mice or ticks, revealing an absolute requirement for this protease. However, the loss of two potential modulators of FtsH activity, HflK and HflC, does not detectably affect B. burgdorferi physiology. Our results provide the groundwork for the identification of FtsH substrates that are critical for the bacterium's viability.

FIG 1

Phenotypic characterization of the ΔhflK/C mutant. (A) Genetic organization of the WT, mutant, and complemented strains at the hflK and hflC loci. (B) Immunoblot analysis of HflK synthesis by different B. burgdorferi strains. Antiserum raised against B. burgdorferi HflK reacts with a protein of the appropriate size (36 kDa [arrowhead]) from whole-cell lysates of WT and complemented strains, but not with a lysate of the ΔhflK/C mutant. (C) In vitro growth rate of the ΔhflK/C mutant compared to the WT and hflK/C-comp strains. The mean and standard deviation (SD) are displayed. (D) Immunoblot analysis of cell lysates from WT B. burgdorferi exposed to a 40°C heat shock treatment for 1 h or maintained at the standard growth temperature (35°C). FlaB levels are shown to demonstrate equivalent protein loads. (E) Cell viability of B. burgdorferi strains subjected to 1 N NaCl osmotic stress compared to untreated spirochetes. After treatment, cells were plated in solid BSK medium, and the viability ratio was determined by counting CFU of treated compared to untreated spirochetes. SD bars are shown; no significant difference in cell viability was detected among B. burgdorferi strains at either the log phase (one-way analysis of variance [ANOVA], P = 0.2850) or stationary phase (one-way ANOVA, P = 0.1478).

FIG 2

The ΔhflK/C strain colonizes ticks as efficiently as WT B. burgdorferi. Fed larval I. scapularis ticks (A) and nymphs (B) were equally well colonized by all three B. burgdorferi strains. Larval ticks were allowed to feed on infected mice, and 8 to 10 days postfeeding, spirochete density was determined by macerating and plating individual ticks and enumerating CFU. Some fed larvae were allowed to molt to the nymphal stage and fed on naive mice, and nymphs were mechanically disrupted 10 days postfeeding and plated to determine spirochete CFU. Each point represents an individual tick, and the mean and upper standard deviation bars are shown (lower SD bars fall below the x axis). No significant difference among strains was detected using the Kruskal-Wallis test. The ratios below the graphs denote the number of ticks that had acquired B. burgdorferi relative to the number of ticks assessed for each strain.

FIG 3

FtsH is required for B. burgdorferi cell growth. (A) Genetic organization of an ftsH-inducible strain and the corresponding WT strain. The lacI repressor was inserted into the bbe02 locus on linear plasmid lp25 along with the streptomycin-resistance gene (aadA). The IPTG-inducible promoter (flacp) was derived from the B. burgdorferi flgB promoter with the 20-nucleotide lac operator sequence inserted (27). flaBp, B. burgdorferi flaB promoter. (B) Growth curve of the B. burgdorferi WT (B31-68-LS) and ftsH(in) strains. The IPTG-inducible ftsH(in) strain demonstrated a growth-dependent response to the concentration of IPTG in the medium. Cell numbers were monitored daily by dark-field microscopy using a Petroff-Hauser counting chamber. (C) FtsH protein levels reflect the IPTG concentration in the ftsH(in) strain. An arrow indicates the FtsH protein signal as determined by immunoblot analysis. The IPTG concentration is shown above the immunoblots along with the time post-IPTG depletion. LacI levels are shown to demonstrate relative protein loads. The number of cells loaded in each lane is indicated below the corresponding immunoblots. Note that the immunoblot corresponding to the WT strain contained approximately 10-fold more cell lysate relative to the other immunoblots. The faint band below the FtsH protein is due to nonspecific binding of the antibody to an unrelated protein, as it is present at the same level at all IPTG concentrations.

FIG 4

Morphology of FtsH-depleted B. burgdorferi cells compared to FtsH⁺ cells. Shown are representative scanning electron micrographs of the B. burgdorferi ftsH(in) strain grown for 3 days either in the presence of 2 mM IPTG and producing FtsH (left panel) or in the absence of IPTG, resulting in FtsH-depleted cells (right panel). Large membrane blebs are evident in FtsH-depleted cells.

FIG 5

A mutation in the lac operator sequence may permit resumption of FtsH production. The ideal lac operator sequence (lacOid) used for construction of the ftsH(in) strain is symmetrical (49). A transversion mutation (boxed) in the lac operator isolated from ftsH(res) might allow production of FtsH to resume, independent of any inducer, as demonstrated in Fig. 3B and C.