SalY of the Streptococcus pyogenes lantibiotic locus is required for full virulence and intracellular survival in macrophages

Abstract

Streptococcus pyogenes utilizes numerous mechanisms for evading the host immune response but has only recently been found to survive in the intracellular environment. In this study, we demonstrate the requirement of a putative ABC transporter permease for intracellular survival in macrophages. The highly attenuated S. pyogenes mutant, SalY, was identified from a transposon mutagenesis screen, with over 200-fold attenuation in virulence in a zebrafish invasive-disease model. Sequencing of the region surrounding the insertion identified a locus that is highly conserved in other S. pyogenes genomes and is homologous to an operon involved in lantibiotic production. In vitro analysis demonstrated that the SalY mutant is deficient in intracellular survival in murine macrophages, a phenotype also observed in zebrafish macrophages in vivo. Macrophage crude cell lysates added to bacterial cultures resulted in the death of the SalY mutant but only growth inhibition of the wild-type strain. Specific depletion of zebrafish macrophages in vivo restored the ability of the SalY mutant to cause disease to wild-type levels. The SalY-infected, macrophage-depleted zebrafish exhibit large lesions and invasive dissemination at a rate and level similar to those of the wild type. In contrast, an M protein mutant with a degree of attenuation similar to that of the SalY mutant did not regain full virulence by in vivo depletion of macrophages. The putative SalY ABC transporter may be an example of the ability of S. pyogenes to adapt and evolve new survival strategies that allow dissemination and growth in previously uninhabitable sites.

FIG. 1.

sal locus structure of S. pyogenes, strain HSC5, serotype M14. Horizontal arrows depict genes within the locus and the direction of transcription. Vertical arrows identify sites of transposon insertions. Open arrows illustrate changes in open reading frames in HSC5 compared to S. salivarius (see text).

FIG. 2.

Zebrafish infections. The histologies of zebrafish dorsal muscle tissue sections infected i.m. are shown 20 h postinjection with 1 × 10⁵ CFU wild-type bacteria (A) and with 1 × 10⁵ CFU SalY mutant bacteria (B). Magnification, ×1,000. Samples were stained using a standard hemotoxylin and eosin staining protocol. Arrows in panel A point to aggregates of bacteria. Arrowheads in panel B point to inflammatory cells at the site of infection. (C) Counts of bacterial CFU recovered from zebrafish spleens. Each indicated strain was injected i.m. into zebrafish dorsal muscles at 1 × 10⁵ CFU. The fish were euthanized, and spleens were removed for bacterial isolation at the indicated time points. Zebrafish were injected i.p. with sterile PBS to act as a negative control for later experiments (see Fig. 5C and text). All assays were done twice with six fish per time point (n = 12). Error bars represent ±SEMs. WT, wild type; *, P < 0.05.

FIG. 3.

(A) Bacteria recovered from in vitro macrophage infection. Each indicated strain was added to RAW 264.7 macrophages at an MOI of 1. At the indicated time points after a 1-hour gentamicin treatment to kill extracellular bacteria (0 time point), macrophages were lysed and viable bacteria were enumerated from supernatants and lysates. Assays were done twice in quadruplicate (n = 8). Error bars represent ±SEMs. (B) TEMs of RAW 264.7 cells infected with the S. pyogenes wild type (top) or the SalY mutant (bottom). Wild-type-infected cells at 24 h postinfection show multiple cocci in cytoplasm or released from dead cells. SalY mutant-infected RAW 264.7 cells at 24 h postinfection show a large empty vacuole and a single cocci in a vacuole (black arrows). WT, wild type.

FIG. 4.

Crude whole-cell lysate was added to bacterial cultures grown in DMEM with 10% FBS. RAW 264.7 cells were lysed in sterile double-distilled water. The graph shows bacterial survival of the wild type (WT) and the SalY mutant strain at 7 h after the addition of cell lysates or water alone. The cell lysate groups include RAW 264.7 cells that had not been activated or RAW 264.7 cells activated for 24 h with gentamicin-killed wild-type bacteria. The graph shows the results of assays done in triplicate on three different days. Error bars represent ±SEMs.

FIG. 5.

Zebrafish infections. Panels A and B show zebrafish spleen cytospin preparations 16 h after i.m. injection. (A) Images of splenic cells from wild-type-infected zebrafish. Multiple intracellular cocci can be visualized. (B) Images of splenic macrophages from SalY-infected zebrafish illustrating the actively changing morphology and the absence of cocci. Magnification, ×1,000. (C) Bacterial CFU recovered from spleens of CAR-treated zebrafish. Prior to injection of bacterial strain, zebrafish were injected i.p. with 100 μg of CAR. Each indicated strain was injected i.m. into zebrafish dorsal muscles at 1 × 10⁵ CFU. The fish were euthanized, and spleens were removed for bacterial isolation at the indicated time points. All assays were done twice with six fish per time point (n = 12). Error bars represent ±SEMs. WT, wild type; *, P < 0.05. For controls, zebrafish were also treated with PBS prior to infection, the results for which are shown in Fig. 2C.

FIG. 6.

Images of zebrafish injected i.m. with the wild type (WT), the SalY mutant, or the Emm mutant at 20 h postinjection. (A) Zebrafish treated i.p. with PBS; (B) zebrafish treated i.p. with CAR.