The role of Staphylococcus aureus Panton-Valentine leukocidin (PVL) in mammalian macrophages

Methicillin-resistant Staphylococcus aureus (MRSA) causes skin infections and life-threatening necrotizing pneumonia in otherwise healthy individuals. Nearly all of these community-acquired MRSA (CA-MRSA) strains express the secreted pore-forming leukotoxin Panton-Valentine leukocidin (PVL). However, its role in pathogenesis remains controversial, as it fails to kill mice and its immune cells. Recently, PVL was shown to bind human, but not mouse, complement receptor 5a (C5aR). While this explained species and cell specificity, the underlying mechanism that leads to PVL pore formation and cell death remains poorly described. Thus, the major focus of the work described in this thesis was to characterize PVL cytotoxicity and identify host factors that mediate cell death. To further probe its function, I generated recombinant PVL, its subunits, LukS-PV and LukF-PV and utilized S. aureus culture supernatants from wild type, ∆pvl and complemented ∆pvl strains. Notably, both native and recombinant PVL were highly cytotoxic to PMA-differentiated human THP-1 macrophages in vitro whereas the subunits LukS-PV or LukF-PV showed no effect (Chapter Two). Intriguingly, human THP-1 monocytes, HeLa epithelial cells and RAW264.7 mouse macrophages remained refractory to PVL toxicity, consistent with the absence of human C5a receptor (hC5aR) expression. While binding to hC5aR is essential for PVL-induced killing of macrophages, I have now shown for the first time that PVL specifically interacts with several anionic phospholipids, including phosphatidylserine (PS), phosphatidic acid (PA) and cardiolipin (CL) (Chapter Three). Surprisingly, these lipids are enriched on different organellar membranes in host cells, suggesting that PVL kills macrophages by targeting intracellular membranes. To gain a spatial-temporal view on the PVL-macrophage interaction, single-cell analysis by live-cell imaging was established. This showed that PVL caused mitochondrial and lysosomal damage prior to cell death, which can be partially prevented by lysosomal cathepsin B inhibitors (Ca-074Me) (Chapter Three). Intriguingly, PVL cytotoxicity was blocked by inhibiting osmotic lysis and potassium efflux (using glycine and potassium chloride enriched media, respectively), and NLRP3 inflammasome activity (glyburide and MCC950) in mammalian macrophages (Chapter Four). In addition, PVL triggered caspase-1 activation and subsequently induced inflammatory responses in mammalian macrophages. Surprisingly, loss of caspase-1 activity did not affect PVL-mediated cell death, but genetic deletion of other inflammatory caspases, such as caspase-4 or caspase-5, in THP-1 macrophages reduced PVL cytotoxicity compared to wild type and caspase-8 deficient cells (Chapter Four). Furthermore, pharmacological inhibition of apoptotic caspases (using pan-caspase inhibitors such as Q-VD-OPh) and RIPK1 (using necrostatin-1s, nec-1s) did not protect mammalian macrophages from PVL-mediated cell death. These observations suggest that PVL kill macrophages by targeting different intracellular membranes and activating several host cell death factors. Collectively, I have identified important host factors that are targeted by PVL to induce macrophage cell death and inflammation. This may lead to host-directed compounds to prevent MRSA-induced immune evasion during infections and the development of new therapeutic treatment options.