Strain-specific association of cytotoxic activity and virulence of clinical Staphylococcus aureus isolates

Abstract

Staphylococcus aureus has been shown to invade and induce the death of various cell types. Here we investigate whether the cytotoxicity of intracellular S. aureus is a general feature or rather characteristic of individual S. aureus strains. The majority of 23 randomly collected clinical S. aureus isolates were killed inside keratinocytes and fibroblasts, indicating that the uptake of S. aureus represents an important mechanism of cell-autonomous host defense. However, seven independent S. aureus isolates survived intracellularly and induced significant cytotoxicity for their host cells. Subcloning analysis revealed that the ability or inability to kill host cells is a stable, apparently genetically determined trait of a given S. aureus isolate. We show that noncytotoxic strains but not cytotoxic strains colocalize with the lysosomal marker LAMP-1, suggesting that only cytotoxic strains escape degradation by the endolysosomal pathway. In a mouse septicemic model, cytotoxic S. aureus isolates produce significantly greater lethality (96%) compared to noncytotoxic strains (41%), which corresponds to 23-, 63-, and 30,000-fold increases of bacterial loads in the liver, spleen, and kidney, respectively. Finally, cytotoxic S. aureus strains produce clinically apparent arthritis in mice at a greater frequency than compared to noncytotoxic S. aureus strains. The results of our study unravel a previously unrecognized dichotomy of cytotoxic and noncytotoxic S. aureus isolates, which may play an important role in the dissemination of, and mortality induced by, S. aureus infection.

FIG. 1.

(A) S. aureus infection of mouse keratinocytes (PAM212) and fibroblasts (mKSA) cells. Cells were infected with clinical S. aureus isolate K42. Invasion of the cells by FITC-labeled staphylococci was measured 1 h after infection by flow cytometry. The marker indicates the cutoff used for the determination of the percentage of infected cells. (B) PAM212 cells were infected with nonlabeled S. aureus (ATCC 29213), fixed, and stained with Cy3-labeled anti-protein A antibody. The bright light differential interference contrast image (left) was superimposed with Cy3 fluorescence (right).

FIG. 2.

Cytotoxicity of individual S. aureus clinical isolates. mKSA cells were infected as described in Materials and Methods. (A) Percentage of FITC-positive cells was determined by flow cytometry 1 h after infection. (B) At 24 h after infection, host cell viability was assessed by trypan blue exclusion. Statistically significant differences are indicated (P < 0.01). (C) Strains K42 and K62 of cytotoxic or noncytotoxic phenotypes, respectively, were subcloned and analyzed for host cell cytotoxicity. (D) mKSA cells were infected with ATCC 29213 (cytotoxic strain) or K135 isolate (noncytotoxic strain) at different MOIs. Host cell death was measured 24 h after infection by trypan blue exclusion. Significant differences are denoted with asterisks (✽, P < 0.01). (E) Influence of bacterial viability or rifampin on S. aureus induced cytotoxicity. mKSA cells were infected with ATCC 29213 isolate or heat-killed bacteria. 10 μg of rifampin/ml was added to the culture medium 1 h after infection. The percentage of cell death was measured 24 h after infection. Statistically significant differences (P < 0.01) are indicated.

FIG. 2.

Cytotoxicity of individual S. aureus clinical isolates. mKSA cells were infected as described in Materials and Methods. (A) Percentage of FITC-positive cells was determined by flow cytometry 1 h after infection. (B) At 24 h after infection, host cell viability was assessed by trypan blue exclusion. Statistically significant differences are indicated (P < 0.01). (C) Strains K42 and K62 of cytotoxic or noncytotoxic phenotypes, respectively, were subcloned and analyzed for host cell cytotoxicity. (D) mKSA cells were infected with ATCC 29213 (cytotoxic strain) or K135 isolate (noncytotoxic strain) at different MOIs. Host cell death was measured 24 h after infection by trypan blue exclusion. Significant differences are denoted with asterisks (✽, P < 0.01). (E) Influence of bacterial viability or rifampin on S. aureus induced cytotoxicity. mKSA cells were infected with ATCC 29213 isolate or heat-killed bacteria. 10 μg of rifampin/ml was added to the culture medium 1 h after infection. The percentage of cell death was measured 24 h after infection. Statistically significant differences (P < 0.01) are indicated.

FIG. 3.

Detection of intracellular S. aureus 24 h after infection. mKSA cells were infected with FITC-labeled S. aureus and FITC fluorescence distribution of infected cells (in gray) was analyzed by flow cytometry. Host cell viability was determined by trypan blue exclusion. Forward- to side-scatter dot plots indicate changes in cell size and granularity associated with infection. (A) ATCC 29213 (cytotoxic); (B) K62 clinical isolate (noncytotoxic); (C) heat-killed ATCC 29213. (D) Cells infected with the noncytotoxic S. aureus isolate K62 were harvested at the indicated times, and the CFU numbers of intracellular S. aureus were determined. Means and standard deviations were determined by four independent experiments. The dashed line indicates the limit of detection.

FIG. 4.

Intracellular localization of noncytotoxic (A), cytotoxic (B), and heat-killed (C) S. aureus strains. PAM212 were infected with FITC-labeled bacteria. At 3 h after infection, cells were fixed and stained with anti-LAMP-1 antibody and visualized by using Cy3-labeled secondary antibody. Images captured by using an FITC filter (lower left), by using rhodamine (upper right), in bright light (lower right), or by superimposition of fluorescence images (upper right) are shown.

FIG. 5.

In vivo virulence of cytotoxic S. aureus. (A) Mice were infected i.v. with 10⁸ CFU of either cytotoxic (ATCC 29213; n = 7) or noncytotoxic (K135; n = 5) S. aureus isolates, respectively. The percentages of surviving animals in the cytotoxic (▪) and noncytotoxic (▵) groups are shown. (B) Two groups of three mice were infected i.v. with 10⁷ of either cytotoxic (ATCC 29213, black bars) or noncytotoxic (K135, white bars) S. aureus isolates. After 18 h, mice were sacrificed and livers, spleens, and kidneys were removed to determine the numbers of CFU per gram of organ. The results are representative for two independent experiments. Means and standard deviations are calculated from individual values of the two experiments.