Autophagy Defends Cells Against Invading Group A Streptococcus

See allHide authors and affiliations

Science  05 Nov 2004:
Vol. 306, Issue 5698, pp. 1037-1040
DOI: 10.1126/science.1103966


We found that the autophagic machinery could effectively eliminate pathogenic group A Streptococcus (GAS) within nonphagocytic cells. After escaping from endosomes into the cytoplasm, GAS became enveloped by autophagosome-like compartments and were killed upon fusion of these compartments with lysosomes. In autophagy-deficient Atg5–/– cells, GAS survived, multiplied, and were released from the cells. Thus, the autophagic machinery can act as an innate defense system against invading pathogens.

Autophagy mediates the bulk degradation of cytoplasmic components in eukaryotic cells in which a portion of the cytoplasm is sequestered in an autophagosome and eventually degraded upon fusion with lysosomes (13). Streptococcus pyogenes (also known as group A Streptococcus, GAS) is the etiological agent for a diverse collection of human diseases (4). GAS invades nonphagocytic cells (5, 6), but the destination of GAS after internalization is not well understood. To clarify the intracellular fate of GAS, especially any possible involvement of autophagy, we first investigated whether intracellular GAS colocalizes with LC3, an autophagosome-specific membrane marker, following invasion of HeLa cells (79). After infection, GAS strain JRS4 cells colocalized with LC3-positive vacuole-like structures in HeLa cells (Fig. 1A). The size (5 to 10 μm) and morphology of the structures were distinct from standard starvation-induced autophagosomes with a diameter of about 1 μm (fig. S1A), so we designated these structures GAS-containing LC3-positive autophagosome-like vacuoles (GcAVs). The number of cells bearing GcAVs, the area of GcAVs, and ratio of GAS trapped in GcAVs to total intracellular GAS increased in a time-dependent manner, reaching a maximum at 3 hours after infection (Fig. 1, B and C; figs. S1B and S2A). A similar result was obtained in mouse embryonic stem (ES) cells (figs. S2B and S3A). About 80% of intracellular GAS were eventually trapped by the compartments (Fig. 1C; fig. S1B). LC3 frequently surrounded GAS, fitting closely around a GAS chain (Fig. 1, D and E; movie S1).

Fig. 1.

Intracellular GAS is acquired by LC3-positive compartments. (A) LC3-positive compartments (green) sequestered intracellular GAS in HeLa cells expressing enhanced green fluorescent protein (EGFP)–LC3 at 3 hours after infection. After incubating with GAS for 1 hour, infected cells were cultured for 3 hours with antibiotics to kill extracellular GAS. Cellular and bacterial DNA were stained with propidium iodide (PI, magenta). Bar, 10 μm. (B) The number of cells bearing GcAVs (gray bars; means ± SE, n = 20) was counted and the area of GcAVs was measured by Image-J software (black circles; means ± SE, n = 20). The micrographs at each point are shown in fig. S1. (C) The total area of GAS within GcAVs was calculated as the percentage of total area of the invaded GAS. (D) High-resolution microscope image of a GcAV (green) and GAS (magenta) at 1 hour after infection. Bar, 2 μm. (E) Confocal microscopic image of a GcAV at 3 hours after infection (LC3, green; DNA, red). The three-dimensional image is available as movie S1. Bar, 2 μm. (F) Immunoblot analysis of LC3-II in GAS-infected HeLa cells. In (B) and (C), data are representative of at least three independent experiments.

LC3 exists in two molecular forms. LC3-I (18 kD) is cytosolic, whereas LC3-II (16 kD) binds to autophagosomes (7, 8). The amount of LC3-II, which directly correlates with the number of autophagosomes (8), increased after infection (Fig. 1F). Thus, GAS invasion appears to induce autophagy, specifically trapping intracellular GAS.

To substantiate this idea, we examined GcAV formation in Atg5-deficient (Atg5–/–) cells lacking autophagosome formation (7). In contrast to the wild-type cells (fig. S2, B and C), no GcAVs were observed in Atg5–/– ES cells (J1-2) (Fig. 2A) or in Atg5–/– mouse embryonic fibroblasts (MEFs) (fig. S2C). Thus, GcAV formation requires an Atg5-mediated mechanism. We also examined LC3-II formation. During infection with GAS, Atg5–/– cells showed no induction of LC3-II (Fig. 2B). By electron microscopy, in wild-type MEF cells infected with GAS, we observed characteristic cisternae surrounding GAS in the cytoplasm (Fig. 2C). No GAS were found surrounded by the membranes in Atg5–/– cells (Fig. 2C). The autophagosome-like multiple membrane–bound compartment containing GAS was also found in HeLa cells (Fig. 2D).

Fig. 2.

Atg5 deficiency allows GAS survival within host cells. (A) Intracellular GAS were not acquired by LC3-positive compartments in Atg5-deficient ES cells at 3 hours after infection. Yellow fluorescent protein (YFP)–LC3, green; PI-stained DNA, magenta. Bar, 10 μm. (B) LC3-II was not formed after GAS infection of Atg5–/– ES cells (A11) but formed in Atg5+/+ (R1) and Atg5 cDNA transformant of A11 (WT13). (C) Ultrastructural observations of GAS-infected Atg5+/+ ES cells and Atg5–/– ES cells. The lower left panel is the magnified image of the area indicated by the box in the upper panel. Arrows indicate the cisternae surrounding naked GAS. Bars, 1 μm. (D) Ultrastructural observations of GAS-infected HeLa cells. We observed the multiple-membrane–bound compartment containing intact cytosol and GAS at 1 hour after infection (upper panel) and the single membrane–bound compartment with degraded cytosol and GAS (arrowhead) at 4 hours after infection (middle and lower panels). Arrows indicate the membranes of the compartments. Bars, 1 μm. (E) Viability of intracellular GAS in Atg5+/+ (closed symbols) and Atg5–/– (open symbols) cells was measured in the presence (squares) or absence (circles) of tannic acid (TA; final concentration, 0.5%) (9). Data are representative of at least three independent experiments. *P < 0.01.

Next, we asked whether the bacteria are killed or survive after entering the compartments. To address this question, we directly scored bacterial viability by counting colonyforming units (CFU viability assay) in wild-type and Atg5–/– MEFs (Fig. 2E). In wild-type MEFs at 4 hours after infection, intracellular GAS had been killed (Fig. 2E), whereas the decrease of GAS viability was suppressed in the Atg5–/– MEFs. Tannic acid is a cellimpermeable fixative that prevents fusion between secretory vesicles and the plasma membrane but does not affect intracellular membrane trafficking (10). In Atg5–/– cells treated with tannic acid to prevent external escape of GAS, the viable bacteria increased by 2 hours after infection and maintained this level at 4 hours after infection (Fig. 2E). In contrast, the numbers of intracellular GAS decreased rapidly in tannic acid–treated wild-type cells as well as in untreated wild-type cells (Fig. 2E). GAS were not killed at all in the Atg5–/– cells, and some of the GAS were released from the cells, suggesting that the autophagic machinery can kill intracellular GAS and helps prevent the expansion of GAS infection. This idea was also supported by the uptake of [35S]methionine and [35S]cysteine by GAS into Atg5–/– cells but not into the wild-type cells (fig. S3B).

At 4 hours after infection, we observed GcAVs with features characteristic of autophagosomes fused with lysosomes, as observed by electron microscopy: a single membrane–bound compartment and containing degraded cytosol (Fig. 2D) and partially degraded GAS (arrowhead). LC3 and LAMP-1, a lysosomal membrane protein, also colocalized at 2 to 3 hours after infection (Fig. 3, A and B), suggesting fusion with lysosomes after formation of GcAVs, similar to what occurs in the standard autophagic pathway. To examine whether the viability of GAS was impaired by lysosomal enzymes, we performed a bacterial viability assay in the presence of the lysosomal protease inhibitors (Fig. 3C). The decrease of intracellular GAS in wild-type cells was suppressed significantly by treatment with protease inhibitors. In Atg5–/– cells, however, the protease inhibitors did not affect the number of viable intracellular GAS, implying that the decrease in GAS viability requires autophagosome formation and fusion with lysosomes.

Fig. 3.

Fusion of GcAVs with lysosomes. (A) Confocal microscopic images of GAS (red)–containing LC3-positive compartments (green) fused with lysosomes (blue). After fixation, lysosomes were stained with an antibody to LAMP-1 (blue). Bars, 2 μm. (B) Increase of GcAVs fused with lysosomes with time. After fixation, cells were stained with an antibody to LAMP-1. The numbers of cells, including LAMP-1–positive or –negative GcAVs, was counted (means ± SE, n = 20). (C) Decrease of intracellular GAS in Atg5+/+ cells was inhibited by addition of the lysosomal enzyme inhibitors (+inhibitors; 1 mg/ml of leupeptin and 10 μM E64d). The numbers of intracellular GAS were determined by bacterial viability assay (mean CFU ± SE). Data are representative of at least three independent experiments. *P < 0.01.

GAS is known to secrete streptolysin O (SLO), a member of a conserved family of cholesterol-dependent pore-forming cytolysins (11). Although the role of SLO is not clear, we found that the intracellular fate of JRS4ΔSLO, an isogenic SLO-deficient mutant of strain JRS4, differed from that of the wild type. At early stages (–0.5 and 0 hours) after infection, GAS often colocalized with the early endosome marker, the FYVE domain of EEA-1 (Fig. 4, A and C; fig. S4) (12), demonstrating that GAS first enter into endosomes. Then, at 1 hour after infection, endosomes containing GAS gradually disappeared (Fig. 4C). In contrast to the wild-type strain, most JRS4ΔSLO cells remained within FYVE-positive compartments even at 2 hours after infection (Fig. 4, A and C), suggesting that JRS4ΔSLO failed to escape from endosomes. Furthermore, only a few GcAVs were observed in the JRS4ΔSLO-infected cells (Fig. 4, B and D). Taken together with the ultrastructural observation (Fig. 2C), we suggest that GAS escapes from endosomes via a SLO-dependent mechanism and that its entry into the cytoplasm induces autophagy and entrapment of GAS in autophagosome-like compartments.

Fig. 4.

The hemolytic toxin (Streptolysin O; SLO)–deficient GAS cannot escape from endosomes and are not acquired by LC3-positive compartments. (A) Confocal microscopic images of GAS (magenta) and FYVE domain of EEA-1 (green)–positive endosomes in FYVE-EGFP–transfected HeLa cells with a SLO-deficient mutant of JRS4 (JRS4ΔSLO). Yellow arrowheads indicate the location of GAS in the endosomes. (B) JRS4ΔSLO (magenta) was not acquired by LC3-positive compartments. HeLa cells expressing EGFP-LC3 (green) were infected with JRS4ΔSLO for 1 hour. After fixation, cells were stained with PI (magenta). Bars, 10 μm. (C) JRS4ΔSLO failed to escape from early endosomes. FYVE-EGFP–transfected HeLa cells were infected with GAS. The area of GAS colocalized with the markers was measured by Image-J software (means ± SE, n = 20). (D) GcAVs were formed by infection of GAS JRS4 but not by JRS4ΔSLO (means ± SE, n = 20).

In keratinocytes, more than 80% of the internalized GAS are killed by 4 hours after infection, and the organisms continue to die over the next 18 hours until they reach ∼1% of their original numbers (13). Here, killing of GAS during the early phase (by 4 hours of after infection) was solely due to autophagic activity. At 24 hours after GAS infection, ∼50% of the infected cells induced apoptosis (13), suggesting that the autophagic killing of GAS is not protective toward the cells. However, autophagy is likely to contribute to suppression of GAS virulence, because killing of GAS inside cells results in a reduction of extracellular GAS that is hazardous and cytotoxic for host tissues and cells (14). Indeed, decreased invasion rates of GAS in fibronectin-deficient mice results in an increased mortality rate (15). Severe and invasive diseases caused by GAS might thus be induced by the attenuation of autophagic activity.

Several bacterial species, including Rickettsia conorii (16), Listeria monocytogenes (17), Porphyromonas gingivalis (18), Brucella abortus (19), and Legionella pneumophila (20), reside within double-membrane–bound compartments resembling autophagosomes after the invasion of host cells (21). However, the significance of this localization has not been clear. Here we have demonstrated that autophagy can play a role in bacterial invasion of host cells.

Supporting Online Material

Materials and Methods

Figs. S1 to S4


Movie S1

References and Notes

View Abstract

Navigate This Article