Human GBP1 facilitates the rupture of the Legionella-containing vacuole and inflammasome activation.

IMPORTANCE: Inflammasomes are essential for host defense against intracellular bacterial pathogens like Legionella, as they activate caspases, which promote cytokine release and cell death to control infection. In mice, interferon (IFN) signaling promotes inflammasome responses against bacteria by inducing a family of IFN-inducible GTPases known as guanylate-binding proteins (GBPs). Within murine macrophages, IFN promotes the rupture of the Legionella-containing vacuole (LCV), while GBPs are dispensable for this process. Instead, GBPs facilitate the lysis of cytosol-exposed Legionella. In contrast, the functions of IFN and GBPs in human inflammasome responses to Legionella are poorly understood. We show that IFN-γ enhances inflammasome responses to Legionella in human macrophages. Human GBP1 is required for these IFN-γ-driven inflammasome responses. Furthermore, GBP1 co-localizes with Legionella and/or LCVs in a type IV secretion system (T4SS)-dependent manner and promotes damage to the LCV, which leads to increased exposure of the bacteria to the host cell cytosol. Thus, our findings reveal species- and pathogen-specific differences in how GBPs function to promote inflammasome responses.

Inflammasomes primarily restrict cytosolic Salmonella replication within human macrophages.

Salmonella enterica serovar Typhimurium is a facultative intracellular pathogen that utilizes its type III secretion systems (T3SSs) to inject virulence factors into the host cell and colonize the host. In turn, a subset of cytosolic immune receptors respond to T3SS ligands by forming multimeric signaling complexes called inflammasomes, which activate caspases that induce interleukin-1 (IL-1) family cytokine release and an inflammatory form of cell death called pyroptosis. Human macrophages mount a multifaceted inflammasome response to Salmonella infection that ultimately restricts intracellular bacterial replication. However, how inflammasomes restrict Salmonella replication remains unknown. We find that caspase-1 is essential for mediating inflammasome responses to Salmonella and subsequent restriction of bacterial replication within human macrophages, with caspase-4 contributing as well. We also demonstrate that the downstream pore-forming protein gasdermin D (GSDMD) and ninjurin-1 (NINJ1), a mediator of terminal cell lysis, play a role in controlling Salmonella replication in human macrophages. Notably, in the absence of inflammasome responses, we observed hyperreplication of Salmonella within the cytosol of infected cells, and we also observed increased bacterial replication within vacuoles, suggesting that inflammasomes control Salmonella replication primarily within the cytosol and also within vacuoles. These findings reveal that inflammatory caspases and pyroptotic factors mediate inflammasome responses that restrict the subcellular localization of intracellular Salmonella replication within human macrophages.

Lipid A Variants Activate Human TLR4 and the Noncanonical Inflammasome Differently and Require the Core Oligosaccharide for Inflammasome Activation.

Detection of Gram-negative bacterial lipid A by the extracellular sensor, myeloid differentiation 2 (MD2)/Toll-like receptor 4 (TLR4), or the intracellular inflammasome sensors, CASP4 and CASP5, induces robust inflammatory responses. The chemical structure of lipid A, specifically its phosphorylation and acylation state, varies across and within bacterial species, potentially allowing pathogens to evade or suppress host immunity. Currently, it is not clear how distinct alterations in the phosphorylation or acylation state of lipid A affect both human TLR4 and CASP4/5 activation. Using a panel of engineered lipooligosaccharides (LOS) derived from Yersinia pestis with defined lipid A structures that vary in their acylation or phosphorylation state, we identified that differences in phosphorylation state did not affect TLR4 or CASP4/5 activation. However, the acylation state differentially impacted TLR4 and CASP4/5 activation. Specifically, all tetra-, penta-, and hexa-acylated LOS variants examined activated CASP4/5-dependent responses, whereas TLR4 responded to penta- and hexa-acylated LOS but did not respond to tetra-acylated LOS or penta-acylated LOS lacking the secondary acyl chain at the 3' position. As expected, lipid A alone was sufficient for TLR4 activation. In contrast, both core oligosaccharide and lipid A were required for robust CASP4/5 inflammasome activation in human macrophages, whereas core oligosaccharide was not required to activate mouse macrophages expressing CASP4. Our findings show that human TLR4 and CASP4/5 detect both shared and nonoverlapping LOS/lipid A structures, which enables the innate immune system to recognize a wider range of bacterial LOS/lipid A and would thereby be expected to constrain the ability of pathogens to evade innate immune detection.