Attenuation of bacterial virulence by quorum sensing-regulated lysis

Abstract

Genetically attenuated pathogenic bacteria have been extensively considered as vaccine candidates. However, insufficient attenuation has been a frequent limitation of this approach. Many pathogens use quorum sensing to escape host defense mechanism. Here, we hypothesized that quorum sensing can be manipulated to diminish pathogenesis. To test this hypothesis, we modified the quorum sensing circuitry of a live cholera vaccine strain to add a second layer of attenuation. Attenuation resulted from the expression of phage PhiX174 lysis gene E on a balanced lethal plasmid from the quorum sensing-regulated luxC promoter. For conditional expression of quorum sensing and positive selection in vivo, the host strain was deleted of its cqsA and thyA genes encoding cholera autoinducer 1 (CAI-1) synthase and thymidylate synthase, respectively. A recombinant cqsA gene expressed from the cholera toxin (CT) promoter and an active thyA gene was provided in trans. The resulting strain expressed CAI-1 in AKI cultures (CT permissive condition) but not in LB medium. Additionally, it expressed elevated biofilm in LB medium compared to AKI conditions where CAI-1 is synthesized to repress biofilm formation. Induction of lysis gene E by quorum sensing restricted growth to a lower cell density in AKI medium, the suckling mouse intestine or LB supplemented with exogenous CAI-1. Microscopic examination revealed the presence of Vibrio cholerae ghost cells at high cell density. Lysis was accompanied by the release of intracellular β-galactosidase to the culture medium. We conclude that it is possible to manipulate quorum sensing to attenuate a live vaccine vector and restrict its shedding to the environment and diminish its subsequent dissemination.

Fig. 1. Structure of balanced lethal plasmid for quorum sensing-regulated lysis

Abbreviations: Amp^R, ampicillin resistance; bla, β-lactamase gene; ORI, colE1 origin of replication.

Fig. 2. A. Production of CAI-1 activity in wild type and recombinant V. cholerae strains

Strains 1333 (WT) and mutants were grown to stationary phase in LB medium at 37°C with agitation (□) or AKI medium at 3O°C (■). The CAI-1 activity present in the clear supernatants was measured using a bioassay as described in methods. Error bars indicate the standard deviation (n = 3). B. Biofilm formation by wild type and recombinant V. cholerae strains. Strains 1333 (WT) and mutants were grown in LB medium at 30°C for 16 h and diluted 1:100 in fresh LB medium (□) or AKI medium (■). The cultures were transferred to 96-well microtiter plates. Biofilms were allowed to develop at 30°C for 24 h and measured by crystal violet staining as indicated in methods. Error bars indicate the standard deviation (n = 3).

Fig. 3. Growth of wild type and recombinant strains under different conditions

A. Strain 1333 (WT) and mutants were grown to saturation in LB (□) or AKI medium (■) (n = 3). B. Strain 1333 (WT) and mutants were orally fed (10⁵ cells) to 2–5 days-old suckling mice (n = 5). Mice were incubated 16 h, euthanized by cervical dislocation, the small intestine dissected and homogenized in PBS. The final cell density (or titer of colonizing Vibrios) was determined by dilution plating on LB agar. Error bars indicate the standard deviation.

Fig. 4. Cross feeding of CAI-1. Panel A. Growth arrest

Strains were grown to mid-exponential phase, centrifuged and resuspended in fresh LB or LB containing 60 % conditioned media as described in methods. Symbols: (□), AJB203 in fresh LB; (■), AJB203 in conditioned media; (▵), AJB203 containing pBRS3 in fresh LB; (▴), AJB203 containing pBRS3 in conditioned media. Panel B. Leakage of intracellular activity. β-Galactosidase activity of cell suspensions was determined with and without permeabilization with chloroform and SDS. Strain AJB203 containing pBRS3 was incubated 90 min in fresh LB or LB plus conditioned media. Unfilled bar, fresh LB without permeabilization (no treatment); grey bar, conditioned media without permeabilization; black bar conditioned media with permeabilization. Each bar is the mean of three independent experiments. Error bars indicate the standard deviation.

Fig. 5. Transmission electron microscopy

Samples of AJB203 containing pBRS3 incubated 90 min in fresh LB media or LB containing conditioned media were examined for the presence of VCG. VCG are indicated with arrows.