Autophagy is a key tolerance mechanism during Staphylococcus aureus infection

Abstract

Defense strategies against infectious threats can be divided into resistance and tolerance mechanisms. Resistance mechanisms involve reduction of pathogen burden and include many established examples, one of them being the destruction of intracellular pathogens through autophagy (xenophagy). Tolerance mechanisms protect the host from damage caused by the pathogen or the immune response independent of pathogen load. The role of autophagy in maintaining homeostasis in response to environmental stress suggests that this pathway is involved in tolerance to a variety of infectious agents. However, demonstrating that autophagy promotes tolerance independent of its role in resistance has been a challenge, especially during infection by clinically relevant pathogens. We have found that autophagy protects against Staphylococcus aureus infection by maintaining tolerance toward a pore forming toxin secreted by the bacteria, α-toxin.

Staphylococcus aureus is a Gram-positive bacterium responsible for considerable morbidity and mortality in humans, causing diseases including pneumonia, endocarditis, osteomyelitis, sepsis, and wound infection. The inability to treat infections caused by antibiotic-resistant strains, such as methicillin-resistant S. aureus (MRSA), represents one of the greatest modern challenges in microbiology. Based on previous work linking autophagy and in vitro Staphylococcus aureus infection, we investigated the importance of autophagy during in vivo infection by the most prevalent community-acquired MRSA clone in the United States, USA300. We found that ATG16L1 hypomorph (ATG16L1^HM) and lc3b^−/− mice, both of which display decreased autophagy in all tissues, are significantly more susceptible to lethality and other signs of disease following both blood and lung infection with USA300. Despite this increase in disease susceptibility, autophagy inhibition was not associated with an increase in bacterial burden. Comparison of blood and 10 different organs at 3 time points each indicated that ATG16L1^HM mice harbor the same or fewer numbers of S. aureus than control wild-type (WT) mice following bloodstream infection. Therefore, we hypothesize that autophagy inhibition leads to impaired tolerance to damage caused by S. aureus.

S. aureus is notorious for producing an arsenal of virulence factors that mediate pathogenesis. We found that ATG16L1^HM mice are not susceptible to a mutant USA300 strain lacking the global virulence regulator agr, suggesting that autophagy is protecting against a virulence factor produced by the bacterium. One major virulence factor under the control of agr is α-toxin. Upon binding ADAM10 (a disintegrin and metallopeptidase domain 10) on endothelial and epithelial cells, α-toxin disrupts tissue integrity and promotes bacterial dissemination by inducing pore-formation and cleavage of cadherins. We found that ATG16L1^HM mice are not susceptible to infection by an α-toxin deletion mutant of USA300. Moreover, α-toxin administered intranasally results in greater lethality in ATG16L1^HM mice compared with control WT mice, indicating that α-toxin alone is sufficient for inducing lethality.

Several observations suggest that disruption of the endothelial barrier, and perhaps the epithelial barrier as well, is why autophagy inhibition is catastrophic in the presence of α-toxin. Similar to ATG16L1^HM and lc3b^−/− mice, which lack autophagy in all cell types, mice lacking ATG16L1 primarily in endothelial cells (Atg16l1^flox/flox-Tek/Tie2-Cre) are susceptible to lethality during blood and lung infection with USA300. Also, ATG16L1^HM mice intranasally inoculated with α-toxin display epithelial and endothelial cell death. Furthermore, we confirmed this protective role of ATG16L1 in cultured primary endothelial cells. α-toxin-treated endothelial cells from ATG16L1^HM mice are significantly more susceptible to cell death and pore formation compared with cells harvested from autophagy-sufficient control mice.

The protein level, but not transcript, of the α-toxin receptor ADAM10 is higher in endothelial cells from ATG16L1^HM mice compared with WT, providing a likely explanation for the increased susceptibility of cells lacking ATG16L1. We also observed increased ADAM10 levels when we inhibited autophagy in endothelial cells with the chloroquine derivative Lys05 or ammonium chloride; conversely, induction of autophagy with serum starvation or the anthracycline drug epirubicin decreases ADAM10 levels. Importantly, the increase in cell death and ADAM10 levels observed in endothelial cells upon autophagy inhibition is not seen in other cell types, revealing a cell type-specific function of this pathway. Finally, epirubicin treatment of WT mice prior to α-toxin administration improves survival, suggesting that pharmacological modulation of autophagy may be a viable therapeutic option for enhancing host tolerance to S. aureus α-toxin.

When taken together, these results support a model in which ATG16L1 and autophagy mediate defense against S. aureus strains such as USA300 by regulating susceptibility to α-toxin (Fig. 1). Therefore, in addition to the established role of autophagy in promoting resistance by degrading internalized bacteria, autophagy can promote tolerance to damage caused by major human pathogens.

Figure 1.

Autophagy can function as a resistance mechanism during infection by sequestering and degrading intracellular pathogens through xenophagy (right panel). Autophagy is also critical for maintaining cellular homeostasis during stress conditions, such as prevention of cell death and damage upon membrane damage, and thereby contributes to tolerance to Staphylococcus aureus α-toxin (left panel).

An important unanswered question is what causes this differential regulation of ADAM10 by autophagy between different cell types. Autophagy may have an essential role in maintaining plasma membrane homeostasis in endothelial cells where cell integrity is critical for preserving blood vessel barrier function. Alternatively, there is little known about ADAM10 trafficking, which may be regulated in endothelial cells in a manner distinct from other cell types. Transmembrane proteins such as ADAM10 may therefore be recycled regularly by autophagy via bulk degradation or in response to damage by pore-forming toxins. This raises 2 important concerns that require consideration in future investigations. First, autophagy may have an important function in maintaining tolerance to other pore-forming toxins produced by various bacteria, both through its function of recycling transmembrane receptor proteins as well as degrading damaged membranes. Second, autophagy deficiency may lead to an inability of barrier cells to maintain membrane homeostasis under conditions of stress.

Identifying the molecular mechanism by which autophagy inhibition leads to dysregulation of ADAM10 will have considerable impact not only as a novel function for autophagy, but also because ADAM10 is a metallopeptidase with many substrates, including the amyloid precursor protein implicated in Alzheimer disease. Understanding how manipulation of the autophagy pathway modulates ADAM10 could have considerable therapeutic impact in humans.

Funding

This work was supported by US National Institute of Health (NIH) grants R01 DK093668 (KC), T32 GM0738 (KM), AI100853 (KM), F30 DK098925 (KM), R01AI099394 (VJT), R01AI105129 (VJT) and American Heart Association 12GRNT12030041 (KC).