UC Berkeley

To protect against infectious disease, the innate immune system must mount a rapid response that distinguishes between self and non-self, and between pathogen and harmless commensal. Invasion of the host cell cytosol, either directly or through the injection of effector proteins, is a uniquely pathogen-associated event. Therefore, the innate immune system includes numerous cytosolic sensors to detect foreign molecules or the disruption of intracellular homeostasis. Among these sensors is the NBD–LRR family of proteins, which oligomerize into large complexes called “inflammasomes”, and serve as a molecular platform for the activation of the protease CASPASE–1.

The outcome of innate sensing in the cytosol is determined by the downstream effector functions of each pathway. In most cases, this occurs through the transcriptional induction of effector proteins, such as inflammatory and chemotactic cytokines. In contrast, inflammasome effector functions are carried out by CASPASE–1 in the absence of de novo transcription or translation, making inflammasome activation one of the most rapid innate signaling responses.

Experiments in mice deficient for CASPASE–1 or other inflammasome components, have revealed the importance of inflammasome activation in restricting bacterial pathogens in vivo. Bacterial restriction is achieved by the effector functions of CASPASE–1, the best characterized of which are the processing and secretion of the inflammatory cytokines IL–1β and IL–18, and the induction of a lytic cell death called pyroptosis. The relative contribution of CASPASE–1 effector functions to immunity varies depending on the disease model; however, in several cases inflammasome-dependent protection is completely retained in IL–1β/IL–18–/– mice, suggesting cytokine-independent inflammasome functions. In this dissertation I explore the molecular mechanisms that regulate cytokine-independent inflammasome activity, and describe a novel CASPASE–1 effector function in vivo.

The dissertation begins with an overview of inflammasome activation and function, highlighting some of the important areas for future research. Next, I describe a full-genome siRNA screen we undertook to identify novel proteins required for inflammasome signaling, and in particular for pyroptosis. Although the screen led to the unbiased identification of known inflammasome components, all novel candidate genes failed to validate after further analysis. Perhaps this negative result indicates a functional redundancy among the CASPASE–1 substrates required for pyroptosis.

In chapter 3, we examine how autoproteolysis determines the effector functions of CASPASE–1. We show that a non-cleavable allele of CASPASE–1 retains the ability to initiate pyroptosis, but fails to process IL-1β/IL–18. This finding explains the previous observation that in the absence of the inflammasome adaptor protein ASC, inflammasome activation leads to cell death without release of cytokines. Furthermore, we show that pyroptosis and cytokine processing can be initiated from spatially and structurally distinct inflammasome complexes, suggesting a mechanism for tuning the outcome of CASPASE–1 activation.

In chapter 4, we identify a novel CASPASE–1 effector function using an in vivo model of inflammasome activation. We show that systemic cytosolic delivery of flagellin, a potent inflammasome agonist, leads to massive vascular fluid loss and can kill mice in less than 30 minutes. This unexpected response is dependent on CASPASE–1, but independent of IL–1β and IL–18. Instead, CASPASE–1 activation leads to a cellular calcium influx that triggers the biosynthesis of eicosanoids — a family of inflammatory lipids that includes the prostaglandins and leukotrienes. Mice deficient in COX–1, an enzyme required for synthesis of prostaglandins, are resistant to the rapid pathology caused by inflammasome activation.

Our findings therefore link the rapid sensing capacity of inflammasomes to the potent activity of eicosanoids. Activated within minutes, this pathway represents one of the most rapid cellular innate immune responses described to date, and suggests a model for the initiation of inflammation at the site of infection. In the 5th and final chapter, I discuss this model and other emerging themes in inflammasome research, highlighting several mouse models I have developed to uncover additional inflammasome functions in vivo. Although to date the emphasis has been on IL–1β and IL–18, the findings reported in this dissertation highlight how much there is to learn about the cytokine–independent functions of inflammasomes.