Developing Caspase-1 Biosensors to Monitor Inflammation in Vitro and in Vivo

Loyola University Chicago Loyola eCommons

Dissertations Theses and Dissertations

2020

Sarah Talley

Follow this and additional works at: https://ecommons.luc.edu/luc_diss

Part of the Immunology and Infectious Disease Commons

Recommended Citation Talley, Sarah, "Developing Caspase-1 Biosensors to Monitor Inflammation in Vitro and in Vivo" (2020). Dissertations. 3827. https://ecommons.luc.edu/luc_diss/3827

This Dissertation is brought to you for free and open access by the Theses and Dissertations at Loyola eCommons. It has been accepted for inclusion in Dissertations by an authorized administrator of Loyola eCommons. For more information, please contact [email protected].

LOYOLA UNIVERSITY CHICAGO

DEVELOPING CASPASE-1 BIOSENSORS TO MONITOR INFLAMMATION IN VITRO AND IN VIVO

A DISSERTATION SUBMITTED TO THE FACULTY OF THE GRADUATE SCHOOL IN CANDIDACY FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

PROGRAM IN INTEGRATIVE CELL BIOLOGY

SARAH TALLEY

CHICAGO, IL

AUGUST 2020

TABLE OF CONTENTS

LIST OF FIGURES v CHAPTER ONE: INTRODUCTION 1 CHAPTER TWO: REVIEW OF THE LITERATURE 5 Overview 5 Structure of Inflammasomes 6 Function of Inflammasomes 8 NLRP1 8 NLRP3 14 NLRC4 21 AIM2 24 PYRIN 28 Noncanonical Inflammasome Activation and Pyroptosis 31 Inflammatory Caspases 36 Caspase-1 36 Other Inflammatory Caspases 40 Biosensors and Novel Tools to Monitor Inflammation 42 CHAPTER THREE: MONITORING CASPASE-1 BIOSENSOR ACTIVATION IN VITRO 45 Materials and Methods 45 Cell Culture 45 293T Transfections and Luciferase Reporter Assay 45 Generation of Stable Cell Lines 45 Monocyte Derived Macrophages 46 In vitro Inflammasome Activation Assays 46 Salmonella Infection 46 Statistical Analysis 47 Results 47 Caspase-1 Biosensor Design 47 Screen to Identify Putative Biosensors 48 Monitoring Biosensor Activation in THP-1 Cells 49 Monitoring Biosensor Activation in MDMs 52 IQAD biosensors are activated in response to inflammatory and apoptotic stimuli 53 CHAPTER FOUR: GENERATION OF CASPASE-1 REPORTER MICE 55 Materials and Methods 55 Generation of Caspase-1 Reporter Mice 55 In vitro Inflammasome Activation Assays 55 Western Blotting 56 Lactate Dehydrogenase (LDH) Measurements 56 IVIS Imaging 57 Statistical Analysis 57 Results 57 Generation of Caspase-1 Biosensor Transgenic Mice 57 Monitoring Biosensor Activation in BMDMs 59 IQAD Biosensors are Not Activated in Response to Some Inflammatory Stimuli 60 CHAPTER FIVE: MONITORING CASPASE-1 BIOSENSOR ACTIVAITON IN VIVO 64 Materials and Methods 64 S. aureus Culture and Murine Systemic Infection Model 64 DSS-Induced Colitis Model 64 Graft-Versus-Host Disease Model 65 Statistical Analysis 65 Results 66 Staphylococcus aureus Bacterial Infection Model 66 DSS-Induced Colitis Model 72 Graft-Versus-Host Disease Model 79 EPS Treatment Reduces T cell Activation in GVHD Mice 86 CHAPTER SIX: SUMMARY AND DISCUSSION 89 REFERENCE LIST 95 VITA 113

LIST OF FIGURES

Figure Page 1. Composition of Canonical and Noncanonical Inflammasomes 7

2. Mechanism of NLRP1 Inflammasome Activation 12

3. Model of NLRP3 Activation on Dispersed trans-Golgi Network 17

4. Schematic of Post-Translation Modifications Regulating NLRP3 Activity 18

5. Schematic of Pyrin Inflammasome Activation. 30

6. Mechanism of Noncanonical Inflammasome Activation and Pyroptosis. 34

7. Caspase-1 Structure, Activation and Inactivation 38

8. Caspase-1 Biosensor Design 48

9. Caspase-1 Biosensors are Activated in 293T Cells 49

10. Caspase-1 Biosensors are Activated in THP-1s 50

11. Caspase-1 Biosensors are Activated in THP-1s in Response to NLRP3 and AIM2 Agonists 50

12. Biosensors are Activated during Bacterial Infection in THP-1 Cells 52

13. Caspase-1 Biosensors are Activated in MDMs 53

14. IQAD Biosensors are Activated in Cells in Response to Inflammatory Stimuli but Secreted from Apoptotic Cells 54

15. Generation of Caspase Reporter Mice 58

16. Biosensor Signal ex vivo IQAD Mice 59

17. Caspase-1 Biosensors are Activated in BMDMs 60

18. Caspase-1 Biosensors are Activated in BMDMs in Response to Some Stimuli 62

19. Caspase-1 Biosensors are Not Activated by Nigericin in BMDMs 63

20. Subcutaneous S. aureus Infection and Cytodex-3 Beads Cause Biosensor Activation in the Skin 67

21. Caspase-1 Biosensors are Activated in S. aureus–Infected Mice 69

22. Biosensor Signal, Bacterial Burden and Inflammatory Cytokines and Chemokines Detected in Tissues Isolated from S. aureus-Infected Mice 71

23. Biosensor Signal and IL-1β Correlate with Bacterial CFU in the Kidneys 72

24. Biosensor Signal Correlates with Inflammatory Markers in the Kidneys 73

25. Biosensor Signal and IL-1β do Not Correlate with Bacterial CFU in the Liver and Heart 73

26. Biosensor Signal in the Colons in vivo 74

27. Mice Exhibit Some Clinical Measurements of Colitis on Day 5 and Day 7 75

28. Caspase-1 Biosensors are Activated in Inflamed Areas of the Colon during Colitis – Males 76

29. Caspase-1 Biosensors are Activated in Inflamed Areas of the Colon during Colitis – Females 77

30. Biosensors are Activated in Regions of Caspase-1 Activation and Inflammatory Cytokine / Chemokine Production in the Colon 78

31. Caspase-1 Biosensors are Activated in the Brain during Colitis 79

32. Biosensors are Activated in Alloreactive Donor Cells during GVHD 81

33. Mice Exhibit Clinical Measurements of GVHD 82

34. Increased Inflammatory Cytokines in the Serum of GVHD Mice 82

35. IQAD Biosensor is Activated in Donor T cells in Mice with GVHD 84

36. IQAD Biosensor is Activated in Alloreactive Donor T cells during GVHD 85

37. Caspase-1 Activation in Donor T cells in GVHD Spleens 85

38. Biosensor Activation is Attenuated in EPS-Treated Mice in vivo 87

39. EPS-Treated Mice Exhibit Reduced Biosensor Signal in Tissues and Reduced Clinical Disease 88

CHAPTER ONE

INTRODUCTION

Inflammation is an integral part of our host defense against various pathogens, but when uncontrolled, can drive the pathology of numerous diseases. Many of these conditions are exacerbated with age, and their incidences will only increase as the general population ages and lifespans increase. A key mediator of inflammatory responses is caspase-1. The recognition of pathogen associated molecular patterns (PAMPs) and damage associated molecular patterns (DAMPs) drives the formation of multiprotein complexes termed inflammasomes, which recruit and activate caspase-1. Caspase-1 cleaves numerous downstream targets, including the proinflammatory cytokines proIL-1β and proIL-18, leading to their maturation and release from cells. The processing and release of these cytokines and other inflammatory effectors mediate inflammation and trigger immune responses.

Inflammation plays a central role in numerous diseases and novel methods to study the initiation of early inflammatory events and the progression of inflammation over time would provide experimental value. The primary assays used to measure inflammasome activation are the proteolytic cleavage of procaspase-1 to caspase-1 by immunoblot or secretion of IL-1β in the serum by ELISA, both of which have notable limitations. Measuring active caspase-1 in vitro by immunoblot is technically challenging and often only detectable in the supernatant from cells stimulated directly with drugs that potently activate caspase-1. Measuring caspase-1 activation during inflammation in vivo requires tissue removal from sacrificed animals, preventing the dynamic measurement of caspase-1 in individual animals overtime. Furthermore, immunoblot detection of caspase-1 activation is often not sensitive enough to detect modest and/or transient activation of caspase-1 that can drive very consequential levels of inflammation in disease pathogenesis. Other common methods used to detect inflammation involve harvesting tissues for microscopic of qRT-PCR analyses to examine the histopathology or the presence of

certain inflammatory markers. These techniques also require tissue removal from sacrificed animals.

Serum IL-1β levels measured by ELISA can be used as an indirect measurement of inflammasome activation. During systemic inflammation, enough IL-1β can be secreted in the bloodstream to be detected by ELISA, but this is often not the case, especially when inflammation is localized to specific tissues.

Even if enough IL-1β is secreted, this assay gives no insight into where the inflammation is coming from.

We therefore sought to develop a biosensor of caspase-1 to overcome the technical limitations associated with monitoring inflammasome activation in vivo. Developing biosensors that directly monitor caspase-1 activation in real time would allow for the identification of specific tissues where inflammation is occurring in living animals and the ability to monitor how this inflammation changes in these animals over time.

To better understand the role of caspase-1 in various disease settings, we developed a series of bioluminescent sensors that allow us to detect and quantify caspase-1 activation. We employed a

circularly permuted form of luciferase in which the N- and C-terminal domains necessary for

bioluminescence are physically separated by a linker region. Caspase-1 target sequences were cloned into

this linker region, so that upon caspase-1 activation, these sequences are cleaved allowing the two domains of luciferase to come together and luminescence can be quantified. In our initial screen, we identified biosensors that exhibited a significant increase in bioluminescence following caspase-1 activation. We hypothesized that caspase-1 biosensors would be activated in cells and animals in response to inflammatory stimuli that promote caspase-1 activation. The goal of this research is to further develop these caspase-1 biosensors into experimental tools that can be used to study inflammation in the context of numerous disease models. We will use a number of inflammatory insults and infection models to identify biosensors that best monitor caspase-1 activation and determine the sensitivity and utility of these biosensors in vitro (Aim 1) and in vivo (Aim 2).

Specific Aim 1: Define the sensitivity and specificity of caspase-1 biosensors in vitro

We will use human THP-1 cells, a monocyte derived macrophage cell line, human monocyte derived

macrophages and murine bone marrow derived macrophages expressing caspase-1 biosensors and 3

quantify luminescence in response to various inflammatory stimuli that activate NLRP3, NLRC4, and

AIM2 inflammasomes. Stimuli to be tested include TLR agonists along with drugs that induce potassium efflux, agents that induce lysosomal destabilization, UV exposure, synthetic dsDNA and dsRNA, and bacterial-derived products such as flagellins. Using validated caspase-1 biosensors, we will monitor the kinetics of biosensor activation in response to intracellular and extracellular bacterial infections in vitro.

In addition to measuring bioluminescence, we will measure caspase-1 activation by immunoblot and the production of proinflammatory cytokines/chemokines by ELISA and CBA to confirm caspase-1 activation in this system.

Specific Aim 2: Determine the utility of caspase-1 biosensors in monitoring inflammation in vivo

We will generate transgenic mice expressing caspase-1 biosensors and identify lines in which biosensors are expressed in all tissues. Using the following models of inflammation, we will monitor biosensor activation in live animals and determine if areas of bioluminescence correspond to locations of caspase-1 activation in tissues:

1. Staphylococcus aureus infection model – We will systemically infect mice with Staphylococcus

aureus (or PBS control) via retro-orbital injection to induce potent inflammation in animals.

Luminescence will be measured 0, 24, 48, 72 and 96h post-infection using the IVIS. Tissues will be

harvested and luminescence quantified ex vivo. To determine if luminescence intensity correlates with

bacterial burden and/or markers of inflammation, infected tissues will be homogenized and plated to

enumerate CFU. We will also measure active caspase-1 and the expression of inflammatory cytokines

and chemokines in tissue homogenates by western blot and CBA, respectively.

2. Colitis model – Mice will be given 2% dextran sodium sulfate (DSS) or PBS sham ad libitum in the

drinking water for 5-7 days to induce acute colitis in animals. We will measure luminescence in vivo

prior to DSS and 7 days post-DSS using the IVIS. We will harvest intestinal tissues on day 7 and

measure luminescence ex vivo. Sections that exhibit strong bioluminescence and control sections that

do not exhibit luminescence will be harvested for histopathological assessment. We will monitor

weight loss, colonic shortening and blood in the stool to determine disease severity. 4

3. Graft-versus-host disease model – To monitor biosensor activation in alloreactive donor cells,

allogeneic BALB/c or control syngeneic CD45.1 C57Bl/6 mice will be injected via the tail vein with

splenocytes and bone marrow cells from caspase-1 biosensor mice. Biosensor activation will be

monitored using the IVIS following transfer. Day 7 post-transfer, mice will be sacrificed, and

biosensor signal will be quantified in the spleen, intestines and lymph nodes. To determine if

probiotics ameliorate GVHD, donor and recipient mice will be given 3 injections of

exopolysaccharide (EPS) isolated from Bacillus subtilis on days -7, -5 and -3 prior to GVHD

induction. GVHD severity will be measured over time in live animals and in excised tissues on day 7.

Impact: Caspase-1 biosensors may serve as an outstanding tool to monitor inflammation in live animals, which would have both experimental and economic value and aid in the development of therapeutic strategies to augment or thwart inflammation.

CHAPTER TWO

REVIEW OF THE LITERATURE

Overview

Tschopp and colleagues first described the existence of an inflammasome in 2002, when they discovered a high-molecular weight, multiprotein complex containing NALP1 (now NLRP1), Pycard

(ASC – apoptosis-associated speck-like protein containing a CARD) and proinflammatory caspases that

were induced under inflammatory conditions (1). Since this landmark finding, several inflammasomes

have been discovered, each activated by specific PAMPs and host-derived molecules that are indicative of cellular damage. Inflammasome assembly leads to the activation of cascades that coordinate the appropriate inflammatory response necessary for pathogen clearance. Consequently, inflammasomes are integral to our immune system and host defense.

Numerous cell types, most canonically cells of the innate immune system, express cytosolic, nuclear and membrane-associated pattern recognition receptors (PRRs), which are capable of sensing general danger signals, PAMPs and DAMPs. A subgroup of cytosolic PRRs in the Nod like receptor

(NLR) family and AIM2 (absent in melanoma 2)-like receptor (ALR) family are capable of forming inflammasomes (2-6). Other PYD-containing proteins, such as Pyrin also, can also form functional inflammasomes in response to PAMP/DAMP recognition (2-5). Assembly of these ‘canonical’ inflammasomes provides a platform for the recruitment and activation of caspase-1. Caspase-1 cleaves numerous downstream targets, including the proinflammatory cytokines proIL-1β and proIL-18; the processing and release of these cytokines activate other immune cells to initiate a robust inflammatory response.

In addition to ‘canonical’ inflammasomes, the ‘noncanonical’ inflammasome can form in response to recognition of intracellular LPS by other inflammatory caspases, human caspase-4 and -5 or

mouse caspase-11 (2-5, 7-10). While inflammasomes have been widely studied in cells of the myeloid lineage, it is becoming increasingly clear that inflammasomes function in a wide variety of cell types, including B cells (11, 12), T cells (13-15), neurons (16-19), astrocytes (20, 21), and microglia (20, 22-24), hepatocytes in the liver (25), keratinocytes in the skin (26-31), and epithelial cells in the intestine (32-39).

Since its discovery nearly two decades ago, studies on the cell and molecular biology of inflammasomes and their critical role in immune responses began to unravel in parallel with the realization that hyperactivation of inflammasomes underlies the pathology of many inflammatory diseases

(40). While inflammasomes evolved to recognize microbial pathogens or sterile injuries, uncontrolled inflammasome activation has become a liability, driving the pathogenesis of many non-communicable diseases, which are becoming increasingly prevalent particularly in aging populations. This chapter will discuss the six major inflammasomes and associated components, the identification of caspase-1 as the critical driver of inflammation and the downstream consequences of caspase-1 activation in cells, as well as the role of inflammasomes in human diseases.

Structure of Inflammasomes

As their name suggests, nucleotide-binding oligomerization domain (NOD), leucine-rich repeat

(LRR)-containing protein (NLRs) are characterized by the presence of a NACHT domain as well as a variable number of LRRs, and are broken up into two subgroups: NLRP inflammasomes containing a

PYD domain and NLRC inflammasomes containing a CARD (caspase recruitment domain) at the amino- terminus (Fig. 1) (2, 3, 41). The structure of most inflammasome-forming proteins is conserved between human and rodents, except for NLRP1. Human NLRP1 contains a PYD domain at the N-terminus, while this domain is absent in rodents. NLPR1 is also unique in that it contains a Function to Find domain

(FIIND) that is not found in any other sensor. AIM2 belongs to the PYHIN (Pyrin and HIN domain) family of proteins, which function as immune sensors of foreign DNA (42). AIM2 inflammasomes are composed of a DNA-binding HIN200 (hematopoietic interferon-inducible nuclear proteins with a 200- amino-acid repeat) domain with an N-terminal PYD motif (3). Pyrin, or TRIM20, is a member of the

TRIM family, containing four structural units including a B30.2/SPRY domain, a coiled-coil (CC) 7

domain, a zinc finger domain (Bbox) and a PYD motif (43). For all canonical inflammasomes, receptor

oligomerization leads to recruitment of procaspase-1 via CARD-CARD interactions (either direct or via

the adapter, ASC) and the proximity-induced proteolytic activation of caspase-1. An inflammasome that

forms independent of caspase-1 has been termed ‘noncanonical’. Noncanonical inflammasome activation

occurs in response to the recognition of components of intracellular gram-negative bacteria (LPS) by the

CARD domain of mouse caspase-11 or the analogous human caspases 4/5 (7-10).

Figure 1. Composition of Canonical and Noncanonical Inflammasomes. NLR family members shown include NLRP1, mouse Nlrp1 (Nlr1b), NLRP3 and NLRC4 all of which contain NACHT domains (blue) and LRRs (green). NLRP1 and NLRP3 inflammasomes further contain the adaptor ASC, consisting of a PYD (orange) and CARD (yellow) domains, while mouse Nlrp1 and NLRC4 are capable of directly recruiting caspase-1 via their N-terminal CARD, though ASC can be involved. AIM2 is comprised of a DNA binding HIN200 domain (grey) and a PYD, allowing it to interact with DNA and subsequently oligomerize ASC. The overall architecture of Pyrin resembles that of its TRIM family members, with a C-terminal SPRY/B30.2 domain (blue), followed by a coiled-coil domain (purple) and B-box domain (green). Unlike other TRIM proteins, Pyrin contains a PYD domain at the N-terminus, allowing it to interact with ASC to form an inflammasome. All canonical inflammasomes recruit and oligomerize caspase-1. The noncanonical inflammasome forms following the recognition of intracellular LPS by other inflammatory caspases, caspase-11 (mouse) or caspase-4/5 (human). Guanylate binding proteins (GBPs) (green) help facilitate the interaction between LPS and caspases.

Nlrp1b and NLRC4 can directly recruit procaspase-1 via their CARD motifs, while NLRP3,

AIM2, Pyrin and NLRP1 utilize the bipartite PYD-CARD adaptor protein ASC (3). ASC is involved in 8

inflammasome formation via homotypic PYD-PYD and CARD-CARD interactions, acting as an adapter between these sensors and procaspase-1 (Fig. 1). Oligomerization of sensors nucleates the recruitment and polymerization of ASC. Structural analysis revealed that ASC oligomerizes through its PYD, forming large helical filamentous structures (44, 45). The CARD domain remains exposed from these filaments, providing a platform for caspase-1 recruitment (44). ASC filaments cluster into visible specks, ~1μm in diameter, forming large macromolecular complexes (46). The final inflammasome complex is star shaped, with caspase-1 radiating from central ASC and receptor complex.

While not all inflammasomes require ASC, ASC serves as a signal amplification mechanism for inflammasome activation and cytokine processing (2). ASC specks are not only the site of caspase-1 recruitment into inflammasome complexes, but may also play a role in the dissemination of inflammation.

Several groups have demonstrated that ASC secreted from dying cells can exists in the extracellular environment, and that this ASC remains active and can be taken up into naïve cells where it can nucleate the polymerization of more ASC molecules, activate caspase-1 and promote IL-1β processing (46-48).

The final piece of the inflammasome puzzle is caspase-1. Procaspase-1 exists as an inactive zymogen. Following sensor oligomerization and ASC speck formation, procaspase-1 monomers are recruited into the inflammasome complex through their CARD domain. Procaspase-1 dimerizes and undergoes proximity-induced autoproteolytic processing, gaining full enzymatic activity (49, 50).

Caspase-1 cleaves numerous target proteins that mediate inflammatory responses. All canonical inflammasomes utilize caspase-1 to elicit their effector function, though other inflammatory caspases can activate inflammatory pathways through noncanonical inflammasome activation.

Function of Inflammasomes

NLRP1

Prior to its discovery, it was known that various microbial agents were capable of activating caspase-1, but the mechanism for caspase-1 activation remained unknown. Tschopp and colleagues hypothesized that NLRP1 might form a complex with caspase-1 via CARD-CARD interactions, analogous to Apaf-1-caspase-9 apoptosome formation (1). They detected a protein complex they termed 9

‘the inflammasome’ that consisted of NLRP1, ASC and inflammatory caspases that processed IL-1β (1).

NLRP1 is 1473 amino acids long and 165.9 kDa, making it the largest NLR-forming inflammasome (51).

Although the NLRP1 inflammasome was the first inflammasome discovered, the physiological role of

this inflammasome is not well understood. Studying NLRP1 has been complicated by the fact that, while

NLRP1 is found in most mammalian species, it is extremely polymorphic and has undergone extensive

genetic diversity within species.

Humans encode a single NLRP1 gene, while mice encode a number of NLRP1 orthologs (Nlrp1a,

Nlrp1b, Nlrp1c), which share structural features of human NLRP1 but lack a PYD (Fig. 1). The most well-studied NLRP1 inflammasome is Nlrp1b, which is activated in response to Bacillus anthracis Lethal

Toxin (LT) in certain strains of mice, driving caspase-1 activation and pyroptosis (Fig. 2) (52, 53). LT is an A/B toxin consisting of a cell-binding protein, protective antigen (PA), and the zinc-metalloprotease lethal factor (LF) (54). PA engages host cell membranes and forms a channel, translocating LF into the cytosol (54). LF cleaves components of mitogen-activated protein kinase (MAPK) signaling pathway to inactivate immune signaling and impair host defense, which is required for virulence (54). Nlrp1 in rodents has evolved to sense bacterial proteases as a mechanism to counteract this strategy. However, unlike rodent Nlrp1, human NLRP1 is not responsive to anthrax LT (55).

Mouse Nlrp1b has multiple polymorphic alleles, some of which confer susceptibility to LT, while others confer resistance; macrophages from mouse strains that carry alleles 1 or 5 are susceptible

(BALB/c and 129S1 strains) while those that carry alleles 2, 3, or 4 are resistant (C57Bl/6) (52).

Catalytically inactive LF mutants are unable to elicit caspase-1 activation and cell death (53, 56), confirming that LF proteolytic activity is essential for Nlrp1 inflammasome activation. LF protease directly cleaves Nlrp1 at the N-terminus, inducing inflammasome/caspase-1 activation, IL-1β release, and pyroptosis (57-59). This cleavage site is present in LT-susceptible rat strains but absence in resistant strains (57), suggesting that Nlrp1 proteolysis is required for inflammasome activation in rats.

It remained unclear whether proteolysis of Nlrp1 was limited to certain isoforms or represented a general mechanism for NLRP1 activation across mammalian species (55), until work by the Vance 10

laboratory revealed that proteolytic cleavage of human and rodent NRLP1 is a general mechanism for

NLRP1 inflammasome activation (55). In humans, cleavage at the linker between the PYD and NACHT

domains removes the N-terminal PYD domain, which functions to maintain NLRP1 in an auto-inhibited

state (26, 55). Vance and colleagues further demonstrated that this linker region, which spans ~240 amino

acids (51), exhibits a strong signature of positive selection, leading the authors to conclude that NLRP1 is

evolving rapidly to become an effective sensor for novel bacterial proteases and DAMPs (55).

While LF cleavage of Nlrp1 is only detected in LT-susceptible strains in rats, LF cleavage of

Nlrp1 can occur in both LT-responsive and LT-unresponsive cells in mice (58), suggesting that another

mechanism may regulate inflammasome formation in response to LT. Consistent with this idea, mouse

and human NLRP1 undergo auto-proteolytic cleavage within the FIIND, and this cleavage is required for

inflammasome formation and caspase-1 activation (60-62). FIIND spontaneously cleaves itself,

generating two polypeptides which remain non-covalently associated, and ASC is recruited to the C-

terminal CARD of processed NLRP1 (Fig. 1-2) (61). In addition to LT, small molecule inhibitors of two serine proteases, DPP8 and DPP9, can activate Nlrp1b, driving caspase-1 activation and pyroptosis (63).

FIIND auto-proteolysis is required for Nlrp1b inflammasome activation induced by DPP8/9 inhibitors, however, unlike LT, these inhibitors do not induce N-terminal cleavage (63). This indicates that while

FIIND cleavage is required for Nlrp1 inflammasome activation, N-terminal cleavage is not.

Interestingly, Nlrp1b inflammasome activation in response LT and DPP8/9 inhibitors requires the proteasome (53, 63, 64), but proteasome inhibitors do not block activation of other inflammasome, suggesting a novel requirement of proteasomal degradation for activation of Nlrp1. Two groups concomitantly discovered that proteasomal degradation of the amino-terminal domains of Nlrp1b resulted in the release of the C-terminal fragment that activates caspase-1 in a process called “functional degradation” (Fig. 2) (65, 66). Chui et al. used a genome-wide CRISPR-Cas9 knockout screen to identify genes involved in Nlrp1b-induced pyroptosis and found an enrichment in genes associated with the N-end rule pathway in LT-treated macrophages, but not in macrophages treated with DPP8/9 inhibitors (65), consistent with an earlier finding that the N-end rule is required for LT-mediated caspase-1 activation and 11

cytotoxicity (67). Sandstrom et al. further demonstrated that the proteasomal degradation of Nlrp1b N- terminus released the C-terminal FIIND-CARD fragment, which was sufficient to activate caspase-1 and form a functional inflammasome (Fig. 2) (66). This model explains why FIIND autoproteolysis is required for Nlrp1 inflammasome activation. Since the proteasome is a processive protease, Nlrp1 would be fully degraded in the absence of FIIND processing, but the break in the polypeptide chain prevents the

C-terminal FIIND-CARD fragment from being degraded (68). This is also an appealing model because it provides a functional rationale for the C-terminally placed CARD, which would be degraded if it was located on the N-terminus like it is on the other NLR-inflammasome forming proteins (Fig. 1) (68).

The CRISPR screen by Chui et al. revealed that the N-end rule did not apply to DPP8/9 inhibitor- mediated Nlrp1 inflammasome activation (65). Mass spectrometry analysis and immunoprecipitation assays demonstrated that DPP9 interacts with human NLRP1 FIIND and functions to maintain NLRP1 in an inhibited state to prevent NLRP1 inflammasome activation (Fig. 2) (69). Because NLRP1 degradation drives its activation, constitutive turnover could result in unwanted inflammasome activation, and DPP8/9 may function to prevent the accumulation of bioactive FIIND-CARD fragments that activate caspase-1

(68). Thus, N-terminal cleavage by LT and inhibition of DPP8/9 both relieve inhibitory mechanisms that are meant to prevent Nlrp1 inflammasome activation. Analysis of gain-of-function mutations in NLRP1

(discussed below) suggests that other regulatory mechanisms likely exist to maintain NLRP1 in an inhibited state (69).

Figure 2. Mechanism of NLRP1 Inflammasome Activation. FIIND undergoes auto-processing, resulting in two non-covalently associated peptides. DPP8/9 maintains Nlrp1 in an autoinhibited state and inhibition of DPP8/9 as well as N-terminal proteolysis (e.g. by anthrax lethal toxin) relives this inhibition. Subsequent proteasomal degradation of the N-terminal domains of NLRP1 liberates the C-terminal CARD-containing fragment that can form a functional inflammasome with caspase-1. The E3 ubiquitin ligase, IpaH7.8 generated by Shigella ubiquitinates Nlrp1, driving its proteasomal degradation and inflammasome activation. T. gondii activates Nlrp1 via an unknown mechanism that is independent of N-terminal and FIIND proteolysis.

Additional pathogens have been shown to activate NLRP1. Listeria monocytogenes and Shigella flexneri were shown to activate Nlrp1b by reducing cytosolic ATP levels and inducing metabolic stress

(70). In the case of S. flexneri, an E3 ubiquitin ligase secreted from this bacterium, IpaH7.8, directly ubiquitinates Nlrp1b, driving its degradation and activation (Fig. 2) (66). Other groups have also shown that mouse Nlrp1b and rat Nlrp1 are activated in response to Toxoplasma gondii infection, driving IL-1β processing and pyroptosis (71, 72), though N-terminal and FIIND cleavage were not detected in response to T. gondii infection (71), meaning either Nlrp1 processing may not be required for inflammasome activation or, more likely, that an unidentified intermediate may drive Nlrp1 inflammasome activation in response to T. gondii (Fig. 2). Interestingly, macrophages expressing Nlrp1 variants that confer LT- resistance were found to respond to T. gondii, driving caspase-1 activation and pyroptosis, whereas cells expressing LT-susceptible Nlrp1 variants supported parasite growth (72). Taken together, these data suggest that NLRP1 isoforms respond to different pathogens, potentially via different mechanisms. 13

NLRP1 plays a role in numerous human diseases, though the bona fide ligand(s) for human

NLRP1 and the mechanism of activation are not well understood. The most significant advances in the field have come from the identification of gain-of-function mutations in NLRP1. A number of NLRP1 mutations have been described in two monogenic epidermal disorders, multiple self-healing palmoplantar carcinoma (MSPC) and familial keratosis lichenoides chronica (FKLC) (26). MSPC patients develop benign proliferative epithelial skin lesions in areas of the skin lacking hair follicles (palms and soles) (26).

Lesions spontaneously resolve ~6 months, but they display similar histologic features of squamous cell carcinomas and can result in malignancy (26, 51). KLC is characterized by the presence of keratotic lesions with lichenoid histopathological features. MSPC patients were found to have mutations in the

PYD domain of NLRP1 (A54T, A66V, M77T), while FKLC patients had deletions in the LRR domain and preceding linker (F787del, R843del) (26). In vitro, cells expressing MSPC and FKLC mutations exhibit increased ASC speck formation, pro-IL-1β processing and secretion, and increased cell death (26).

However, the PYD domain of human NLRP1, unlike other PYD-containing inflammasomes, does not nucleate ASC; the CARD-containing C-terminal domain is sufficient to oligomerize ASC and activate inflammasome signaling (26), as demonstrated in rodents. Zhong et al. detected increased ASC speck formation by simply removing the PYD domain of NLRP1, supporting the hypothesis that NLRP1-PYD plays a role in maintaining NLRP1 in an auto-inhibited state (26). The authors found that WT-NLRP1 exists as a monomeric species, while MSPC, FKLC and ΔPYD NLRP1 mutants spontaneously oligomerize, suggesting that both the PYD and LRR domains are required to maintain NLRP1 in an autoinhibited form and prevent aberrant inflammasome activation (26).

Another autoinflammatory disease associated with NLRP1 is autoinflammation with arthritis and dyskeratosis (AIADK) also known as NLRP1-associated autoinflammation with arthritis and dyskeratosis

(NAIAD), characterized by fevers, dyskeratosis and arthritis (69, 73). AIADK results from mutations between the NACHT and LRR domains (R726W) or within the FIIND (P1214R) (69). The gain-of- function mutation in the FIIND stands out amongst the others, which all occur within or near the autoinhibitory domains of NLRP1. Work by Zhong et al. revealed that P1214R disrupted NLRP1-DPP9 14

association (69), further supporting a regulatory role of DPP9 in maintaining NLRP1 inhibition in human

cells.

All of these auto-activating mutations in NLRP1 are linked to various skin diseases (26, 69, 73),

suggesting a role for NLRP1 in the skin in humans. Consistent with this, human keratinocytes in

particular express high levels of NLRP1, as well as ASC, caspase-1, IL-1β and IL-18 (26, 31). NLRP1 is also expressed in PBMCs and lymphoid tissue (26, 74), raising the question as to why these mutations fail

to cause systemic inflammatory diseases, such as those seen in patients harboring mutations in NLRP3

(68). DPP8/9 inhibition was shown to activate NLRP1 in human PBMCs, resulting in significant cell

death and IL-1β processing, indicating that these cells are competent for NLRP1 inflammasome activation. It could be that NLRP1 functions to provide barrier protection, analogous to NLRC4 in the intestine. In support of this, while NLRP1 is the primary inflammasome functioning in human skin, mouse keratinocytes express very low levels of inflammasome proteins Nlrp1, Nlrp3, Aim2 and Asc (31), making them somewhat refractory to inflammasome activation. This could be because human skin is more directly exposed to exogenous DAMPs, while mice have fur that provides an extra layer of protection.

NLRP3

The most well-studied inflammasome is NLRP3. NLRP3 activation involves a two-step process.

First, PRR recognition of microbial ligands or sterile priming initiates a signaling cascade that results in

NF-κB activation and the upregulation of proinflammatory cytokines and NLRP3 so that the cell is primed to quickly respond to a second insult. A diverse panel of PAMPs and DAMPs serve as ‘signal 2’, including bacterial pore-forming toxins, potassium efflux, extracellular ATP, mitochondrial dysfunction, reactive oxygen species, calcium mobilization, particulate matter, lysosomal disruption, and pathogens, which results in the assembly of NLRP3 inflammasomes (75-81). Potassium efflux is thought to be a crucial upstream mediator of NLRP3 activation (2, 79), but the mechanism by which potassium efflux triggers NLRP3 activation remains unclear. 15

Factors that regulate NLRP3 inflammasome activation have been extensively studied. In addition

to its transcriptional regulation, NLRP3 inflammasome activation is, in part, coordinated by cell biology –

NLRP3 and associated factors must be localized to specific cellular compartments in order for the

inflammasome to form. Furthermore, numerous post-translational modifications are in place to regulate

NLRP3 activity under physiological conditions and in response to NLRP3-activating stimuli. These inherent checkpoints are critical to ensure NLRP3 inflammasomes are turned on at the appropriate time

and downregulated following resolution of tissue damage to prevent excessive inflammation.

Many cellular host factors are required for NLRP3 inflammasome assembly, and the availability

of these host factors in the cell regulates NLRP3 activity. NIMA-related kinase 7 (NEK7), a serine- threonine kinase that modulates microtubule dynamics during mitosis, has recently been identified as a regulator of NLRP3 inflammasome assembly and activation (82-84). In non-mitotic cells or interphase,

NEK7 is available and binds to the LRR of NLRP3 in response to potassium efflux and ROS, though the catalytic activity of NEK7 is not required for NLRP3 activation (83, 84). NEK7 instead serves as a scaffold for NLRP3 inflammasome activation, promoting its self-oligomerization. Bruton’s tyrosine kinase (BTK) also plays an essential role in NLRP3 inflammasome activation. BTK physically interacts with ASC and NLRP3 and functions as a platform protein for the interaction between these two proteins

(85). The Kanneganti lab recently identified DDX3X as a component of the NLRP3 inflammasome complex (86). DDX3X interacts with the NACHT domain of NLRP3 and facilitates its oligomerization

(86). DDX3X is required for the pro-survival assembly of stress granules, cytoplasmic compartments of proteins and RNA that form and enable cells to survive in response to various stressors (86). Upon investigating the cross-talk between these two pathways, the authors found that inflammasome activation and stress granule assembly are mutually exclusive processes that both depend on DDX3X and that both must compete for DDX3X availability in the cell (86). The authors proposed a model where DDX3X acts as a live-or-die checkpoint in stressed cells – if DDX3X assembles into stress granules, it is unavailable for inflammasome activation and the cell survives, whereas if DDX3X is incorporated into the NLRP3 inflammasome, the cell undergoes pyroptosis (86). 16

Various stress conditions can drive ROS production from the mitochondria, and numerous groups

have shown that mitochondrial ROS is a potent activator of NLRP3. Accumulation of mitochondrial

ROS, resulting from either inhibition of complex I or III of the mitochondrial respiratory chain or

inhibition of mitophagy, which is the cell’s mechanism for clearing damaged mitochondria, leads to

increased NLRP3 activation and IL-1β production (75). Because ROS are short-lived and can only act as

messengers for a short distance, NLRP3 likely resides in close vicinity to the mitochondria (75). In steady

state, NLRP3 is found colocalized with the endoplasmic reticulum (ER), but under inflammatory

conditions, NLRP3 relocates to the perinuclear space and colocalizes with ER and mitochondria

structures (75). Juxtaposition between ER and mitochondria forms mitochondria-associated ER membranes (MAMs) (87). MAMs could be the location of NLRP3 activation and ASC recruitment (75).

Microtubules also play a role in coordinating NLRP3 inflammasome activation in cells by facilitating the dynein-dependent transport of ASC on mitochondria to NLRP3 on the ER (88).

Recent work from James Chen’s group reported that the disassembly of the trans-Golgi network

(TGN) is the common stress event driving NLRP3 activation (Fig. 3) (89, 90). The authors used cell fractionation assays and fluorescent microscopy approaches to better understand the subcellular localization of NLRP3 and found that NLRP3 colocalizes with the TGN38, a marker of TGN, in response to diverse NLRP3 agonists (89). Notably, the TGN appears disassembled into vesicle-like structures, while cis and medial Golgi networks remain intact (89). NLRP3 colocalizes with dispersed TGN in ASC knockout cells, indicating this process occurs upstream of ASC recruitment to the inflammasome complex and caspase-1 activation (89). The authors further demonstrated that TGN disassembly exposes negatively charged phosphatidylinositol-4-phosphates (PtdIns4P) that interact via ionic bonding to a conserved polybasic KKKK motif within NLRP3 (89, 90). In this way, the dispersed TGN serves as a scaffold for the recruitment and oligomerization of NLRP3 and subsequent polymerization of ASC (89,

90). ER-Golgi intermediate compartments (ERGIC) mediates trafficking and sorting of cargo between these two organelles, and considering both the ER and Golgi are essential for NLRP3 assembly, it is likely that ERGIC plays a role in NLRP3 activation as well. 17

Figure 3. Model of NLRP3 Activation on Dispersed trans-Golgi Network. TGN disassembly in response to NLRP3 activators triggers exposure of PI4P microdomains that recruit NLRP3 through ionic bonding between negatively charged PI4P and positively charged lysine residues between the PYD and NACHT domain of NLRP3.

Prior to stimulation, numerous post-translational modifications maintain NLRP3 in an inactive conformation. NLRP3 activity is positively regulated by a number of post-translational modifications that include phosphorylation and ubiquitination (Fig. 4). Phosphorylation of the Y861 residue in the LRR domain of NLRP3 negatively regulates its activity. In response to various signal 2 agonists, protein tyrosine phosphatase non-receptor 22 (PTPN22) de-phosphorylates NLRP3 at this residue, thus acting as a positive regulator of NLRP3 inflammasome activation (91). Phosphorylation of the PYD domain of

NLRP3 at S5 also negatively regulates its activity, causing electrostatic repulsion of PYDs, thus disturbing NLRP3 assembly (92). Protein phosphatase 2A (PP2A) dephosphorylates NLRP3 at S5 during priming, licensing NLRP3 for activation (92). Another key priming event for NLRP3 activation is JNK1- mediated phosphorylation of NLRP3 at position S194 (S198 in human), downstream of TLR signaling

(93). Phosphorylation at this site promotes the deubiquitination of NLRP3 and facilitates its self- association (93). In addition to triggering the transcriptional upregulation of NLRP3, priming licenses inflammasome activation by initiating the deubiquitination of NLRP3 by the deubiquitinase (DUB)

BRCC3 (BRC36 in humans) (94, 95). Protein kinase D (PKD) signaling emanating from the Golgi is also 18

required for NLRP3 activation (96). In response to NLRP3 activators, DAG accumulates at the Golgi,

triggering the recruitment and activation of PKD (96). PKD phosphorylates NLRP3 in the NBD at

position S293 (S295 in human), driving the release of NLRP3 from MAMs, the recruitment of ASC and

assembly of the inflammasome (96).

NLRP3 activity is negatively regulated by a number of post-translational modifications (Fig. 4).

Ubiquitination can induce the proteasomal or autophagic degradation of NLRP3. An E3 ubiquitin ligase

in the Skp-Cullin-F-box family member, F-Box L2 (FBXL2) inhibits NLRP3 activity by inducing its

polyubiquitination and subsequent proteasomal degradation (97). LPS priming induces the degradation of

FBXL2, extending the lifespan of NLRP3 for inflammasome formation (97). Another E3 ubiquitin ligase,

TRIM31, also induces the ubiquitin-mediated proteasomal degradation of NLRP3 and is thought to be a

critical feedback suppressor of inflammasome activation, since priming induces the expression of

TRIM31 instead of inhibiting its activity (98). Dopamine serves as an endogenous inhibitor of NLRP3 by

activating DRD1 signaling to induce NLRP3 ubiquitination and autophagic degradation via the E3

ubiquitin ligase, MARCH7 (99).

Figure 4. Schematic of Post-Translation Modifications Regulating NLRP3 Activity. Positive regulators (blue) and negative regulators (red) of NLRP3 inflammasome activation via phosphorylation/dephosphorylation or ubiquitination/deubiquitination.

Cryopyrin-Associated Periodic Syndromes (CAPS) collectively describes a clinical continuum of

three systemic autoinflammatory disorders, Muckle-Wells syndrome (MWS), familial cold-induced 19

autoinflammatory syndrome (FCAS) and chronic infantile neurologic cutaneous articular syndrome

(CINCA) / neonatal onset multi-system inflammatory disease (NOMID), all of which result from mutations in NLRP3 and manifest as nearly continuous episodes of inflammation (100). MWS, FCAS, and CINCA/NOMID disorders are classified as distinct disease entities, although they share several overlapping symptoms, and some patients even present with clinical features that overlap with more than one CAPS subtype (100, 101). CINCA/NOMID is the most severe of the three (101). Patients exhibit nearly continuous symptoms of inflammation beginning during infancy, and later develop CNS symptoms including chronic aseptic meningitis, cerebral atrophy, increased intracranial pressure, seizures, and progressive hearing loss (101). MWS presents as episodic flares of inflammation associated with rash, fever, arthralgia, progressive hearing loss, with some patients developing serum amyloid A amyloidosis

(101). Inflammation in FCAS patients is triggered by cold exposure and is characterized by fever, rash, headaches, and arthralgia (101). Both MWS and FCAS patients can have daily baseline fatigue and flu- like malaise (100).

To date, more than 200 CAPS-associated mutations have been documented (85). Most of these mutations are located in exon 3, which encodes NACHT/NACHT-associated domains (100, 101). These gain-of-function mutations lead to hyperactive NLRP3 assembly and activation that occurs in the absence of signal 2, resulting in increased tissue inflammation that drives various disease symptoms. CAPS patients respond well to IL-1 targeting therapy (Anakinra) (102-104), supporting a role for IL-1β in CAPS pathogenesis. Other IL-1 targeting agents used include a neutralizing IL-1β antibody canakinumab and a soluble decoy IL-1 receptor rilonacept.

Recently there has been a lot of interest in targeting the NLRP3 inflammasome directly. In addition to CAPS, NLRP3 has been implicated in the pathogenesis of a number of diseases, including autoimmune diseases, metabolic disorders including type 2 diabetes and obesity, atherosclerosis, gout, neurodegenerative disease, lung disease, liver disease, kidney disease, and ageing (3, 6, 105). Such a drug would need to specifically attenuate NLRP3 activity, without affecting the other inflammasomes necessary for antimicrobial immunity and host defense. Coll et al. recently described the development of 20

a small molecule inhibitor of NLRP3 called MCC950 (106). MCC950 specifically inhibits the activation

of NLRP3, caspase-1 and IL-1β maturation, while NLRP1, NLRC4 and AIM2 remain fully functional

(106). MCC950 also inhibits NLRP3 activation in vivo, attenuating experimental autoimmune encephalomyelitis (EAE) severity and increasing survival in a mouse model of CAPS (106). MCC950

directly interacts with the NLRP3 NACHT domain, proximal to the Walker B motif, and blocks the

ability of NLRP3 to hydrolyze ATP (107), a critical step for NLRP3 inflammasome activation (108). CY-

09, another NLRP3 inhibitor, also directly binds the NACHT domain in the Walker A motif of NLRP3,

which competes binding of ATP to inhibit NLRP3 inflammasome activation (109). CY-09 is protective in mouse models of CAPS and type 2 diabetes (109). More recently, the drug Tranilast was found to specifically inhibit NLRP3. Tranilast is an analog of tryptophan metabolite and is used as an anti- inflammatory (110). Mechanistically, Tranilast inhibits IgE-induced histamine release from mast cells, but more recently it was also shown to bind the NACHT domain of NLRP3 and inhibit NLRP3-NLRP3 interactions to prevent inflammasome oligomerization (110). This drug has strong clinical potential as an anti-inflammatory, since it is well tolerated by patients at doses of up to 600mg/day over the course of months (110). The recent report by James Chen’s group demonstrated that a polybasic region of NLRP3 is required for its recruitment to dTGN (89). This newly identified polybasic region may be a promising therapeutic target.

Another approach to regulate NLRP3 activity is to inhibit the host factors that are involved in

NLRP3 inflammasome activation or upstream signaling events that drive NLRP3 assembly. Inhibition of

BTK, a component of the NLRP3-inflammasome complex, can be beneficial for the treatment of acute inflammatory diseases. The BTK inhibitor Ibrutinib has a neuroprotective effect on brain ischemia by suppressing inflammasome/caspase-1 activation in infarcted areas of injured brains (85). Oridonin is the major active ingredient of Rabdosia rubescens, a widely used herbal medicine for the treatment of inflammatory diseases (111). Oridonin directly binds the NACHT domain of NLRP3 at position C279 to prevent NLRP3 interaction with NEK7 (111). This drug was shown to have both preventative and therapeutic effects in mouse models of MSU-induced peritonitis and gouty arthritis as well as diabetes 21

(111). The ketone metabolite β-hydroxybutyrate (BHB) was also found to specifically suppress NLRP3 activation, without affecting NLRC4, AIM2 or noncanonical inflammasome responses (112). BHB inhibits NLRP3 inflammasome activation by preventing potassium efflux, ASC oligomerization and IL-

1β production (112). Ketone bodies serve as alternative energy sources for various tissues during periods of nutrient deprivation and adherence to low carbohydrate diets. During these periods, metabolic signals like BHB dampen innate immune responses to spare ATP. Mice fed a ketogenic diet, which elevates serum BHB levels, were protected in a mouse model of FCAS, suggesting that dietary or pharmacological approaches that elevate BHB can reduce the severity of NLRP3-dependent inflammatory diseases (112).

A number of other NLRP3-targeting drugs have been developed in addition to those mentioned here

(reviewed in (85)).

NLRC4

While NLRP3 is the most well-studied inflammasome, NLRC4 is perhaps the best understood, likely due to the structural biology that has helped to define the inflammasome paradigm. All components of the NLRC4 inflammasome pathway have been identified, including the trigger (bacterial ligands), the sensor (NAIP proteins), the nucleator (NLRC4), the adaptor (ASC) and the effector (caspase-1). NLRC4 is a critical component to host defense against bacterial pathogens.

NLRC4 was initially named Ipaf (ICE-protease activating factor), identified as a pro-apoptotic protein capable of activating caspase-1, until the structure revealed that it belonged to the NLR family of proteins (40, 113). Like other inflammasome-forming proteins, NLRC4 expression is upregulated by proinflammatory stimuli, however basal levels of NLRC4 are sufficient for inflammasome activation in various cell types (40). So unlike the NLRP3 inflammasome, which is tightly regulated by transcriptional mechanisms (i.e. ‘signal 1’), NLRC4 inflammasome activation is primarily regulated at the post- transcriptional level, through ligand binding (NAIP proteins) to induce the conformational activation of

NLRC4 and phosphorylation of S533 in the LRR domain of NRLC4.

NLRC4 is capable of recognizing a diverse panel of bacterial ligands by utilizing different sensor proteins in the NLR family apoptosis inhibitory protein (NAIP) family. In mice, Naip1 and Naip2 activate 22

Nlrc4 in response to conserved T3SS rod or needle proteins, while Naip5 and Naip6 serve a similar

function to Naip1/2, activating Nlrc4 in response to cytosolic flagellin (114, 115). In contrast to mice,

humans encode a single genetic locus for NAIP, but the gene has several transcript variants that encode for unique NAIP isoforms that can activate NLRC4 in response to T3SS needle protein and cytosolic flagellin (116, 117). Human NAIP has the highest sequence homology with murine Naip1 (118).

Exactly how NAIP-binding of bacterial ligands drove NLRC4 inflammasome activation remained unclear until 2013, when Hu and colleagues resolved the X-ray crystallography structure of monomeric

NRLC4, demonstrating that NLRC4 resides in an autoinhibited form, with intramolecular interactions between the C-terminal LRR domain and the nucleotide binding domain (NBD) of NACHT resulting in the shape of a question mark (119). Further cryo-EM studies revealed NAIP binding initiates assembly of

9-11 NLRC4 molecules, forming a wheel-like shape (120, 121). Following NAIP binding, the C-terminal

LRR domain of NLRC4 opens via a ~90˚ rotation along a hinge region between the LRR and NBD, exposing a catalytic surface for NLRC4 oligomerization and autoactivation (121). Mutations in certain domains of NLRC4, such as the hinge region, can allow the protein to adopt an open conformation in the absence of NAIP, leading to NLRC4 autoactivation and the development of severe autoinflammatory diseases (discussed below) (40, 122). Once oligomerized, NLRC4 recruits procaspase-1 into the inflammasome complex via CARD-CARD interactions. Compared to the other NLR-forming inflammasomes, NLRC4 is unique in that it can directly recruit caspase-1 via its C-terminal CARD.

While NLRC4-caspase-1 activation is well-established to occur in the absence of ASC (123), interaction with ASC greatly augments caspase-1 activation (124, 125).

In addition to ligand binding by NAIPs, NLRC4 is post-translationally modified by the δ isoform of Protein Kinase C (PKCδ) or Leucine Rich Repeat containing Kinase-2 (LRRK2), both of which phosphorylate Serine 533 (S533) within the LRR domain (126, 127). NLRC4 S533A mutants fail to recruit caspase-1, assemble ASC and process proIL-1β (126). Naip5 is required for flagellin-induced caspase-1 activation but is dispensable for S533 phosphorylation, suggesting that Naip-mediated sensing of flagellin and Nlrc4 phosphorylation are independent processes that converge upon inflammasome 23

activation (128). Interestingly, flagellin from Helicobacter pylori was shown to induce Nlrc4 S533 phosphorylation but not inflammasome activation (128). These data led to the conclusion that S533 phosphorylation primes NLRC4 for activation by NAIP proteins, but phosphorylation alone is not sufficient to induce NLRC4 inflammasome and caspase-1 activation (128).

While NLRC4 activity is primarily regulated at the post-translational level, the expression of

Naip genes (118) and Prkcd, the gene encoding PKCδ (129), is regulated by IFN regulatory factor 8

(IRF8), suggesting that IRF8 may play a role in various aspects of the NLRC4 inflammasome pathway. In vivo, Irf8-/- mice are more susceptible to S. typhimurium infection, exhibiting significant weight loss and mortality comparable to Nlrc4-/- mice (118). These data demonstrate a role for IRF8 in dictating NLRC4- mediated innate immune response against pathogenic bacteria in vitro and in vivo (130).

The first description of human diseases associated with NLRC4 occurred in 2014, when two groups described autoinflammation with recurrent Macrophage Activation Syndrome (MAS) and a syndrome of neonatal-onset enterocolitis and periodic fever caused by gain-of-function mutations in

NLRC4 (131, 132). Canna et al. identified a mutation (T337S) in a highly conserved region within the

NACHT domain of NLRC4, called helical domain 1 (HD1), that they hypothesized would interfere with maintaining NLRC4 in an autoinhibited state, leading to increased NLRC4 inflammasome activation and

MAS (131). Monocytes and macrophages from MAS patients exhibited elevated cytokine secretion, cell death and spontaneous ASC aggregation consistent with a hyperinflammatory phenotype, and in vitro, increased spontaneous caspase-1 activation was detected in cells expressing T337S NLRC4 compared to

WT (131). IL-1R antagonist (anakinra) therapy, which successfully treats patients with CAPS, ameliorated systemic inflammation in the patient with the T337S NLRC4 mutation, but high IL-18 levels and macrophage hyper-responsiveness persisted (131). Romberg et al. identified a mutation (V341A) also within HD1, in a 23 day old patient that died from infantile enterocolitis and autoinflammation (132). In vitro, V341A NLRC4 promoted spontaneous caspase-1 activation, ASC oligomerization and increased

IL-1β and IL-18 production and cell death in LPS-primed macrophages (132). As neonatal-onset enterocolitis resolves by 1 year of age in survivors, the authors hypothesized that this chronic 24

inflammatory state could be exacerbated in the infant’s gut due to constant exposure to bacterial ligands, and as the host-microbe interaction matures, it eventually reaches homeostasis and becomes less inflammatory (132). Other NLRC4-related autoinflammatory diseases have been reported, resulting in conjunctivitis, rash and skin lesions that vary in severity and localization, pain and swelling of joints from a S445P mutation (133) and urticarial-like rash, as well as arthritis and fever induced by exposure to cold stimuli that results from a H443P mutation in NLRC4 (134). All of the mutations described thus far occur near the putative ADP/ATP binding site, which makes sense considering that bound ADP locks NLRC4 in an autoinhibited conformation (119).

AIM2

In 2008, Tschopp and colleagues observed the existence of an inflammasome that responded to viral, bacterial, genomic and synthetic DNA (135). Their results demonstrated that ASC and caspase-1 were required for IL-1β secretion in response to cytosolic DNA, but NLRP3 was dispensable, leading the authors to conclude that the DNA sensor formed an alternative inflammasome (135). Subsequent groups identified the IFN-inducible HIN-200 family member, absent in melanoma 2 (AIM2), as the cytoplasmic

DNA-sensing inflammasome (136-139). Recognition of DNA leads to the oligomerization of AIM2, providing a platform for ASC recruitment and caspase-1 activation, culminating in pyroptosis (138). The generation of Aim2-/- mice further confirmed a role for the AIM2 in host defense against viral and bacterial pathogens in vivo (140-142).

AIM2 recognizes dsDNA structures non-specifically, enabling AIM2 to form inflammasomes and activate caspase-1 in response to diverse DNA ligands. This is accomplished through electrostatic attraction between negatively charged dsDNA and positively charged HIN domain residues (143). AIM2 binds both strands of dsDNA, across the major and minor grooves, which explains its specificity for dsDNA instead of ssDNA (143). Optimal AIM2 activation, as measured by IL-1β release, occurs in the presence of ~80bp of transfected dsDNA, a length that would be able to accommodate ~20 AIM2 HIN domains (143). In the absence of DNA, AIM2 resides in an auto-inhibited state, with its PYD domain 25

interacting with the HIN domains (143). Once bound, AIM2 oligomerizes along the dsDNA lattice,

leading to ASC recruitment and inflammasome assembly (143).

AIM2 is an important sensor to our host defense because of its ability to sense DNA, a central component of many pathogens. But, for AIM2 to recognize pathogen-derived DNA, the DNA must somehow get exposed in the cytosol. For example, bacteriolysis of intracellular bacteria, such as Listeria monocytogenes, in the cytoplasm releases DNA that can be sensed by AIM2, triggering inflammasome

activation and pyroptosis (144). A critical component of cell-autonomous immunity to Gram-negative

bacteria is the production of type I IFNs and the induction of IFN-stimulated GTPases, including

guanylate-binding proteins (GBPs) and interferon-regulated GTPases (IRGs), which target vacuolar and

cytosolic pathogens (145-149) and enable pathogen recognition by cytosolic sensors. The concerted role

of GBPs and IRGs is most well-studied in response to the bacterium Francisella novicida (F. novicida),

which is the common model for Francisella tularensis in mice. F. novicida infection is sensed by the

cGAS-STING pathway, activating type I IFN signaling and IRF1 expression. IRF1 regulates the

expression of GBPs, which induce bacteriolysis to promote AIM2 recognition of bacterial DNA (150,

151). Specifically, GBP2 and GBP5 are required for F. novicida-induced AIM2 activation (150, 151).

The interferon-inducible protein IRGB10, is also essential for F. novicida-induced AIM2 activation (149).

It is thought that GBPs control the recruitment of IRG10 to bacteria, where IRG10 localizes to bacterial

membranes and facilitates the release of bacterial DNA into the cytosol for recognition by AIM2 (149).

Importantly, type I IFNs and IRF1 do not contribute to AIM2 activation in response to some DNA

ligands, such as mouse cytomegalovirus (mCMV) or transfected dsDNA (151).

AIM2 also plays a role in sensing viral DNA, including vaccinia virus and mCMV, as mice

deficient in Aim2 exhibit reduced inflammatory cytokine production and attenuated IFN responses, both

of which are critical to control of viral replication (140). While AIM2 is well-established to sense viral

DNA (137, 140), the mechanisms of how viral DNA is exposed in the cytosol for AIM2 recognition are

not well-defined. Notably, not all DNA viruses trigger AIM2 inflammasome activation, as many have

evolved strategies to thwart ALR sensing. For example, herpes simplex virus 1 (HSV-1) still elicits a 26

strong IL-1β response in the absence of AIM2 (140), and recently, it was shown that HSV-1 blocks AIM2

inflammasome activation through the tegument protein VP22. Many DNA viruses have likely evolved to

evade AIM2 sensing, since AIM2 inflammasome activation triggers rapid pyroptosis, an unfavorable

environment for viral replication. RNA viruses, such as influenza A (IAV), and other pathogens can also

indirectly elicit AIM2 activation by triggering the release of endogenous DNA from infected cells, which

serves as a ligand for AIM2 (152, 153).

Given AIM2’s ability to recognize dsDNA in a sequence-independent manner, it is not surprising

that endogenous DNA is also a ligand for AIM2 inflammasome activation. Drugs/stimuli that induce

changes to the nuclear envelope or DNA damage activate the AIM2 inflammasome. HIV-protease

inhibitors, specifically Nelfinavir, were recently shown to disrupt nuclear envelope integrity, driving

DNA relocalization into the cytoplasm where it activates the cGAS-STING pathway and primes AIM2

inflammasome activation, driving IL-1β release and cell death (154). HIV-infected patients on

antiretroviral therapy (ART) often experience inflammatory sequelae and this provides a potential

mechanism contributing to chronic inflammation in these individuals. Recently, Hu et al. showed that

AIM2 could sense DNA in the nucleus (38). In their model, Aim2-/-, Asc-/-, and caspase-1/11-/- mice

were resistant to subtotal body irradiation (SBI)-induced lethality and intestinal damage (38). They further demonstrated that AIM2 localized to the nucleus upon radiation exposure and colocalized with gamma-

H2AX, a marker for dsDNA breaks (38). The authors concluded that AIM2 might be recruited to sites of radiation-induced DNA damage and mediate inflammasome activation and cell death (38). Since AIM2 is capable of sensing DNA following disruption of the nuclear envelope, this raises the question as to how

DNA sensors are not activated during mitosis. It could be that AIM2 is regulated by an unknown mitotic factor(s) analogous to NEK7 for NLRP3.

Unlike all of the other inflammasomes described here, no gain-of-function mutations in AIM2 have been described to date. This could be because NLRs can oligomerize independent of their ligand, whereas AIM2 requires dsDNA for its oligomerization (155). However, aberrant AIM2 activity has been linked to many human diseases. AIM2 and its ALR family member, IFI16, play a role in systemic lupus 27

erythematosus (SLE), an autoimmune disease caused by abnormal self-DNA recognition processes, characterized by the production of auto-antibodies against various nuclear antigens. Usually, cytosolic

DNA is rapidly degraded by DNases and extracellular DNA is rapidly cleared. In SLE patients, impaired dsDNA degradation and defective clearance of apoptotic cells or neutrophil extracellular traps (NETs) result in the accumulation of self dsDNA, driving auto-antibody production, activation of proinflammatory responses and a type I IFN signature (156). IFI16 contains a PYD and two HIN binding domains and is capable of forming inflammasomes with ASC and caspase-1 (15). The lupus susceptible

New Zealand Black (NZB) strain is characterized by increased expression of the IFI16 orthologue, p202.

Unlike IFI16, p202 lacks a PYD domain and is incapable of forming an inflammasome. However, it has been shown to bind DNA and antagonize Aim2 activity, suggesting a role for AIM2 in this disease (157,

158). It is thought that p202 evolved in certain strains of mice to minimize Aim2-inflammatory responses, but the consequence for this is an increased susceptibility to Lupus (158). While the ALR family of DNA sensors are well appreciated to be involved in SLE pathology, their exact functions in this disease are not completely understood.

AIM2 plays a role in numerous sterile inflammatory diseases. In all cases, healthy cells engulf necrotic cell DNA, a DAMP triggering AIM2 inflammasome activation and pyroptosis. In patients with acute and chronic kidney diseases, nucleosomes and dsDNA are released during necrotic cell death, and these DAMPs have been shown to activate the AIM2 inflammasome (159). Aim2-/- mice exhibit reduced renal injury, inflammation and fibrosis in a mouse model of kidney injury induced by unilateral urethral obstruction, and increased AIM2 expression is detected in the tubules and infiltrating leukocytes in nephrectomy samples from patients with human kidney diseases (159). Cytosolic DNA triggers AIM2 activation and IL-1β production in human keratinocytes, contributing to the pathogenesis of psoriasis

(29). AIM2 is also upregulated in atopic dermatitis lesions, suggesting a role for AIM2 in both chronic and acute skin diseases (160), though the source of cytosolic DNA in psoriatic keratinocytes remains unknown. AIM2 may be involved in many neuronal diseases as well. AIM2 is thought to be detrimental during ischemic brain injury, as Aim2-/- mice have reduced injury and improved neurological scores after 28

middle cerebral artery occlusion (161). CSF isolated from patients with TBI contain increased levels of cell-free dsDNA compared to non-trauma patients, which triggered AIM2 activation and caspase-1 cleavage in embryonic cortical neurons in vitro (162). However, AIM2 regulates neuronal morphology

(18) and is protective against various pathogens such as enteroviruses (17), so its expression in the CNS, particularly in neurons, is quite important.

Because of its ability to sense self-DNA, AIM2 activity must be tightly regulated to prevent autoinflammation. As mentioned above, the identification of p202 as a negative regulator of Aim2 activity has led some to suggest that this protein was evolutionary selected in certain mouse strains to prevent aberrant

Aim2 activity (158). A recently identified isoform of IFI16, IFI16-β, also negative regulates AIM2 activation similar to mouse p202 (163). Like p202, IFI16-β contains two DNA binding HIN domains but lacks a PYD, so it is unable to form an inflammasome (163). IFI16-β mRNA levels are also upregulated in lupus (163). Another negative regulator of AIM2 inflammasome activation is TRIM11. Ubiquitinated

TRIM11 targets AIM2 for autophagic degradation in response to DNA stimulation (164). Certain DNA sequences, such as the TTAGGG repeat found in telomeric DNA, abrogate nucleic acid sensing pathways, and work from the Fitzgerald lab revealed that a suppressive oligodeoxynucleotide termed

A151, a synthetic ssDNA species containing four repeats of the TTAGGG motif, was capable of binding

AIM2 and IFI16 and inhibiting downstream inflammatory responses and cell death (165). A more comprehensive understanding of AIM2 regulation will be important for developing therapeutics to prevent AIM2-induced inflammation.

Pyrin

Pyrin is the newest addition to the group of inflammasome-forming proteins. Pyrin is encoded by the gene MEFV, and mutations in MEFV result in an autoinflammatory disease called familial

Mediterranean fever (FMF), characterized by spontaneous episodes of fever and inflammation (166).

Pyrin (TRIM20) is a member of the TRIM family of proteins, containing many of the structural elements of its family members including B-box, coiled-coil and SPRY domains, except the typical RING motif, which confers E3 ubiquitin ligase activity, is switched out for a PYD motif. In its native state, Pyrin is 29

thought to exist as a trimer (167). Most FMF-associated gain of function mutations are clustered in the

B30.2/SPRY domain of Pyrin, resulting in constitutive caspase-1 activation (discussed below) (166).

The function of Pyrin was only recently uncovered in a landmark paper by Xu et al., where they

found that bacterial Rho-activating toxins and effector proteins trigger Pyrin inflammasome activation

(168). Many bacterial and viral pathogens that replicate in host cells depend on the cytoskeleton to enable

their uptake and movement within cells, and so sensors that detect modulations in cytoskeletal dynamics

would be advantageous for pathogen detection. Rho GTPases, which act as molecular switches to

coordinate changes in the actin cytoskeleton and other cellular activities, are frequently targeted by

bacterial toxins during infection (169, 170). Xu et al. found that a toxin protein secreted by Clostridium

difficile (TcdB) triggered caspase-1 activation and ASC speck formation, independently of NLRP3,

NLRC4, AIM2 and NOD1/2 (168). Further analysis into the mechanism revealed that TcdB and other

bacterial toxins and effector proteins (Vibrio parahaemolyticus VopS, Histophilus somni IbpA-Fic1/2,

lostridium difficile TcdA, Burkhoderia cenocepacia TecA and Clostridium botulinum C3) modify the

Switch I region of RhoA, resulting in its inactivation and toxin-induced Pyrin activation (Fig. 5) (168). In this way, Pyrin serves as a guard protein, surveying the activity of host proteins to sense if they are modified by pathogens (43). Inactivation of RhoA drives Pyrin inflammasome activation via the RhoA effector kinases PKN1 and PKN2, which bind and phosphorylate Pyrin (171). 14-3-3 regulatory proteins bind phosphorylated Pyrin and, in turn, block inflammasome activation (171, 172). Less PKN and 14-3-3 binds Pyrin in FMF-knock in mice expressing mutations in human SPRY, indicating that this domain is important for maintaining Pyrin in an autoinhibited state (171).

A number of drugs are used to treat patients with FMF, including colchicine, a drug that disrupts microtubules. Colchicine and other microtubule polymerization inhibitors were found to inhibit Pyrin inflammasome activation downstream of dephosphorylation and 14-3-3 dissociation but upstream of ASC speck formation (172, 173). In support of a role for microtubule dynamics in Pyrin inflammasome formation, FMF mutations in Pyrin remove this obligatory requirement for microtubules (173). 30

Figure 5. Schematic of Pyrin Inflammasome Activation. Rho-family proteins are active when bound to GTP and inactive when bound to GDP. GTPase-activating proteins (GAPs) inactivate Rho proteins by facilitating GTP hydrolysis, while guanine nucleotide exchange factors (GEFs) activate Rho proteins by inducing GDP-GTP exchange. Bacterial infection leads to RhoA modification and inactivation by bacterial toxins or effectors (red), inhibiting PKN phosphorylation of Pyrin which triggers 14-3-3 dissociation (red arrow), leading to Pyrin-ASC- caspase-1 activation. Microtubule dynamics plays an important role in Pyrin inflammasome assembly, as microtubule targeting drugs prevent Pyrin-mediated ASC aggregation.

Pyrin may play a role in regulating the activity of other inflammasomes and associated

components. The RING domain present in most TRIM proteins possess E3 ubiquitin ligase activity, allowing TRIM proteins to conjugate a wide range of different polyubiquitin linkages that modify target proteins, for example targeting cargo for degradation (174). Pyrin lacks E3 ligase activity but can directly

bind numerous autophagic mediators, including ULK1, Beclin and ATG16L via its B-box, coiled-coil

and/or SPRY domains (175). Consistent with earlier reports demonstrating that the SPRY domain of

Pyrin interacts with inflammasome components NLRP3 and caspase-1 (176, 177), Pyrin was also shown

to degrade NLRP1, NLRP3 and procaspase-1 through autophagy, suggesting a regulatory role for Pyrin 31

(175). Disease-associated mutations in the SPRY domain of Pyrin could perturb Pyrin-mediated degradation of other inflammasomes, such as NLRP3, leading to excessive caspase-1 activation and elevated IL-1β production (175).

FMF results from mutations within the B30.2/SPRY domain of Pyrin (most frequently M680I,

M694V, V726A). FMF is characterized by recurrent episodes of fever, abdominal/chest pain, arthritis, skin rash, or more severe secondary serum amyloid A protein amyloidosis, and is treated with IL-1 inhibitors or colchicine (178). In addition to FMF, a number of other Pyrin inflammasome disorders have been described. Pyrin-associated autoinflammation with neutrophilic dermatosis (PAAND) results from a mutations in Pyrin (S242R, E244K), resulting in disruption of a 14-3-3 binding motif and subsequent

Pyrin inflammasome activation (179, 180). These patients present with severe neutrophil-mediated dermatosis, arthralgia, myalgia, cystic acne and longer episodes of fever, compared to FMF patients

(178). In vitro, these mutations drive spontaneous ASC speck formation, caspase-1 activation and inflammatory cytokine secretion (179, 180). Patients with mutations between the coiled-coil and B30.2 domain at position 577 (T577N/A/S) exhibit a similar autoinflammatory phenotype as patients with FMF, presenting with recurrent fevers, serositis, amyloidosis, and signs of systemic inflammation (181, 182).

PBMCs isolated from one these patients showed increased IL-1β/cytokine responses following LPS stimulation (181). The high presence of FMF mutations in multiple populations support the idea that such mutations provide an evolutionary benefit for people with a subclinical inflammatory phenotype, where increased Pyrin activity may have been protective against endemic infections throughout history (178).

Noncanonical Inflammasome Activation and Pyroptosis

The term pyroptosis (from the Greek root pyro, relating to fire) was coined by Cookson and

Brennan in 2001 as a way to describe a caspase-1 dependent form of programmed cell death that was observed in macrophages upon Salmonella typhimurium infection (183). In contrast to apoptosis where unwanted cells are removed quietly, pyroptosis ‘sounds the alarm’, releasing inflammatory mediators and alarmins, which benefits the host by promptly recruiting immune cells and driving inflammation (183,

184). Ten years later, Vishva Dixit’s group found that caspase-11 was required for noncanonical 32

inflammasome-induced cell death (7). They determined that it was the loss of caspase-11, not caspase-1

that protected mice from LPS-induced lethality (7). Triggering of the noncanonical inflammasome was

later shown to be driven by the presence of cytosolic LPS, specifically the penta- or hexa-acyl lipid A moiety of LPS from gram negative bacteria, as a means of protection against bacteria that escape the vacuole (Fig. 6) (9, 10, 185).

Cytosolic LPS from Gram-negative bacteria is thought to directly bind caspase-11 and human caspase-4/5 resulting in oligomerization and activation of these caspases (8), however, recent work from the Broz group and others has challenged this paradigm. Outer membrane vesicles (OMVs) produced by

Gram-negative bacteria can enter cells via endocytosis and deliver LPS into the cytosol, which explains how LPS from extracellular bacteria can activate the noncanonical inflammasome (186). However, the lipid A moiety of LPS that is required for caspase-11 activation is highly hydrophobic and embedded within the OMV (187). Santos et al. hypothesized that another factor was required to facilitate the interaction of LPS with caspase-11 (187). They found that caspase-11 bound less efficiently to LPS in the absence of GBPs in vitro, and GBP KO mice injected with OMVs exhibited reduced levels of inflammatory cytokines and chemokines and reduced endotoxemia in vivo (187). These data indicate a requirement for GBPs in LPS sensing and caspase-11 activation, though whether GBPs directly bind LPS or do so indirectly through intermediate proteins remains to be determined (187). Intriguingly, imaging work by Man et al. depicts GBPs and IRGs encased within a layer of LPS in E. coli, which would suggest that GBPs and IRGs are directly accessible to the lipid A moiety of LPS within the bacterial membrane

(149), though it is difficult to conceive how GBPs and IRGs gain access to the inner part of Gram- negative bacterial membranes.

Another group recently demonstrated that LPS delivery into the cytosol requires hepatocyte- released high mobility group box 1 (HMGB1) (188). HMGB1 produced by hepatocytes during endotoxemia and bacterial sepsis binds LPS, mediates its uptake into lysosomal compartments in macrophages and endothelial cells via the receptor for advanced glycation end-products (RAGE), and permeabilizes the phospholipid bilayer of lysosomes to facilitate LPS leakage into the cytosol and 33

subsequent caspase-11 activation (188). It remains to be determined if other unknown factors facilitate

caspase-4/5/11 recognition of LPS. Structural analysis of the LPS-caspase4/5/11 noncanonical

inflammasome are needed to better understand the inflammasome paradigm and to provide a basis for the

identification of noncanonical inflammasome inhibitors, which would be invaluable for developing

treatments for sepsis.

While caspase-11 activation drives pyroptosis by itself, NLRP3 inflammasome activation is

required for cytokine secretion. The NLRP3 inflammasome is activated downstream of LPS sensing by

inflammatory caspases, driving caspase-1 activation and IL-1β processing (Fig. 6) (7, 10, 185, 189).

Exactly how caspase-11 activation triggers NLRP3 activation remains unclear, although recent work from

Ruhl et al. investigated this link and found that caspase-11 activation leads to a drop in intracellular potassium levels, a common danger signal for NLRP3 inflammasome activation (190). This could, in part, explain the mechanism of NLRP3 activation downstream of caspase-11.

It is likely that caspases-4/5/11 are capable of recognizing molecules other than LPS. Leishmania species was recently found to trigger noncanonical inflammasome activation (191). The parasite membrane glycoconjugate lipophosphoglycan (LPG) is recognized by caspase-11 in macrophages infected with Leishmania (191). Purified LPG does not activate caspase-4/5/11, confirming the involvement of an additional molecule for LPG recognition (191).

The mechanism of noncanonical inflammasome activation-induced cell death remained unknown, until 2015, when two groups concurrently identified Gasdermin D (GSDMD) as the mediator of pyroptosis (192, 193). Shi et al. identified GSDMD in a genome-wide CRISPR screen in BMDMs, while

Kayagaki et al. used a forward genetic screen with mutagenized mice to link GSDMD and pyroptosis

(192, 193). GSDMD is cleaved by inflammatory caspases (caspase-1, 4, 5 and 11), releasing its N- terminal domain (~31kDa) that perforates the plasma membrane and mediates IL-1β release, cell swelling and osmotic lysis (Fig. 6) (192-195). Biophysical analysis estimates GSDMD forms pores with a diameter of ~10-30nm, much larger than the diameter of cleaved IL-1β, which is ~4.5nm and, therefore, could easily pass through GSDMD pores (194, 195). Importantly, the N-terminal fragment of GSDMD 34

specifically binds phospholipids that are present on the internal leaflet of viable mammalian cell

membranes, so GSDMD release into the extracellular environment does not cause the death of

neighboring cells (194). Phospholipids are also present on the outer leaflet of phagosomes (194), meaning

that GSDMD pores may provide a mechanism for bacterial escape into the cytosol. N-terminal GSDMD

can also bind cardiolipin, a component of bacterial membranes, and is able to directly kill both Gram-

negative and Gram-positive bacteria in vitro (194), providing another mechanism of antimicrobial host

defense.

Figure 6. Mechanism of Noncanonical Inflammasome Activation and Pyroptosis. Inflammatory caspases 4/5/11 sense intracellular LPS, driving cleavage of GSDMD into its C and N-terminal fragments. The N-terminal of GSDMD oligomerizes in the plasma membrane, forming pores and triggering pyroptosis. Caspase-1 is activated in parallel, it processes IL-1β and the bioactive cytokine is released from GSDMD pores. Other molecules including IL-1α and HMGB1 are also released. The formation of GSDMD pores in the plasma membrane can lead to pyroptosis. 35

Pyroptosis and IL-1β secretion are temporally associated, so it is often proposed that IL-1β is passively released from pyroptotic cells. However, IL-1β can be secreted from living cells, providing an argument against an obligatory requirement of cell lysis for IL-1β release. In addition to its role in driving pyroptosis, GSDMD pores serve as a conduit for the passage of IL-1β across intact lipid bilayers (Fig. 6)

(196, 197). Experimentally preventing pyroptosis (by treating cells with an osmoprotectant like glycine) inhibits the release of LDH and cell death but does not block pore formation and IL-1β secretion (196).

Pyroptosis is also not a required outcome of caspase-1 activation in specific cell types. Inflammasome activation in neutrophils drives IL-1β secretion without concomitant cell death (198). Similarly, monocytes release IL-1β in the absence of pyroptosis (199, 200). As both neutrophils and monocytes are recruited early to sites of infection and are potent drivers of inflammatory and anti-microbial responses, this intrinsic resistance to pyroptotic cell death likely allows these innate immune cells time to produce high quantities of IL-1β and perform their effector functions without dying in a caspase-1 dependent manner. Pyroptosis is an important mechanism to prevent sustained inflammatory responses, by killing cells that would otherwise continue to propagate inflammation. Notably, neutrophils are remarkably short-lived, with a half-life ~6-8h and circulating monocytes have a half-life ~1 day. Pyroptosis may be more important in cells that are longer-lived, like tissue-resident macrophages. There could also be an additional regulatory step required for the initiation of pyroptosis in these cells that have yet to be identified.

Importantly, GSDMD-independent mechanisms of IL-1β secretion exist (201, 202). Elegant work by Kate Schroder’s group dissected the role of GSDMD on IL-1β secretion. They found that early IL-1β release depends on GSDMD, but this is followed by GSDMD-independent IL-1β release in non pyroptotic primary myeloid cells, such as neutrophils or monocytes (201). This indicates that at least two mechanisms for IL-1β secretion exist and that the GSDMD-independent secretory pathway may be the primary route for IL-1β release in certain cell types or in specific contexts (201). Notably, ectopic expression of mature IL-1β but not proIL-1β is slowly released from macrophages in the absence of inflammatory stimuli, caspase-1 or GSDMD, indicating that maturation is not only necessary but 36

sufficient for IL-1β secretion (201). Under physiological conditions (neutral pH), proIL-1β is negatively charged and repelled by the plasma membrane. IL-1β maturation exposes a positively charged polybasic motif that enables electrostatic interactions with membranes (201). Specifically, mature IL-1β is recruited to negatively charged phosphatidylinositol 4,5-bisphosphate (PIP2)-enriched plasma membrane domains

(201). At the EM level, plasma membrane projections and surface ruffles can be seen decorated with PIP2 and mature IL-1β, while proIL-1β resides in intracellular locations away from the plasma membrane

(201). This relocalization likely supports both GSDMD-dependent and GSDMD-independent IL-1β release from the cell.

Inflammatory Caspases

Caspase-1

Caspases belong to a family of aspartate-specific cysteine proteases that have been well conserved throughout evolution (3). In 1989, two groups reported the identification of a protease responsible for cleaving IL-1β in monocytes and leukocytes (203, 204), long before the seminal 2002 study that led to the identification of inflammasomes. The enzyme was identified three years later as IL-

1β converting enzyme (ICE) (205), later named caspase-1. Caspase-1 is present in cells as an inactive zymogen. Upon stimulation by a myriad of microbial and endogenous ligands, caspase-1 is recruited into inflammasomes via its N-terminal CARD that interacts with the CARD domain within polymerized ASC or, in some cases, directly with the oligomerized inflammasome (e.g., NLRC4 or mouse Nlrp1b).

Following recruitment into the inflammasome complex, procaspase-1 undergoes proximity-induced proteolytic activation. Active caspase-1 cleaves several inflammatory mediators, including proIL-1β and proIL-18, into their mature biologically active forms.

For a long time, it was thought that active caspase-1 was comprised of a tetramer of two p20 and two p10 subunits, but recent work from Kate Schroder’s group revealed that in cells, the dominant species of active caspase-1 are full-length p46 and a transient p33/p10 species on inflammasome complexes (Fig.

7) (49). Recruitment of caspase-1 monomers into inflammasomes leads to the oligomerization of caspase-

1 and autoproteolytic cleavage at the interdomain linker (IDL), yielding a fully active p33/p10 dimer 37

complex. Further autoprocessing at the CARD domain linker (CDL) results in a p20/p10 species that rapidly dissociates from the inflammasome and inactivates (49). Thus, the inflammasome is not merely a platform for caspase-1 recruitment and activation; instead, caspase-1 requires association with inflammasomes to retain its activity. In this way, the inflammasome-caspase-1 complex functions as a holoenzyme, regulating the activation and self-intrinsic inactivation of caspase-1 (49). In support of this, caspase-1 displays promiscuity toward many substrates only when its concentration exceeds a threshold of ~50-100nM (206). The inflammasome hub likely facilitates this concentration requirement for caspase-

1, and dissociation of caspase-1 p20/p10 from the ASC speck results in inactivity (49).

Figure 7. Caspase-1 Structure, Activation and Inactivation. Caspase-1 structure and subunits (A) identified species (B) and pathway of activation and inactivation (C). 1. Caspase-1 monomers are recruited to the ASC speck, 2. Caspase-1 dimerizes and gains some activity, 3. Caspase-1 gains full activity following autoproteolytic processing at the IDL, 4. Caspase-1 undergoes self-cleavage at the CDL, 5. p20/p10 dissociates from inflammasomes and loses activity.

Boucher et al. further demonstrated that inflammasome platform size and cell type dictates

caspase-1 activity duration. Larger ASC speck formation correlates with increased caspase-1 activation

and more rapid protease inactivation (49). Consist with these data, when they assessed NLRC4

inflammasome activation in Asc-/- cells, in which caspase-1 can still be activated by its direct recruitment

via CARD-CARD interactions, they found that the dominant active species was p46 and that this species retained activity for a longer duration compared to active caspase-1 in Asc+/+ cells (49). This sheds light on a previous phenomenon that active caspase-1 (p20) could not be detected in the supernatant of Asc-/- cells stimulated with NLRC4 ligands, despite the detection of cell death and low levels of IL-1β 39

processing, which would otherwise suggest that caspase-1 is active (207). Together, these data indicate that inflammasomes dimerize caspase-1 into an active but unprocessed form in the absence of ASC, and this caspase-1 species can initiate pyroptosis, while the fully processed caspase-1 species is required for optimal cytokine cleavage (Fig. 7C) (49, 207, 208). Cell type may also dictate caspase-1 activity duration, as active caspase-1 p46 and p33/p10 species were detected in neutrophils up to 8h following nigericin stimulation, compared to the 15 minute window in macrophages (49). The finding that the inflammasome complex and cell type govern the kinetics of caspase-1 activation makes sense considering that the larger the inflammasome/ASC speck, the more caspase-1 binding sites available, which would allow more rapid caspase-1 activation and subsequent inactivation. Additionally, cells express different levels of caspase-1, meaning caspase-1 activation kinetics may be regulated by the cell-type specific caspase-1 pools available.

This self-intrinsic mechanism to terminate caspase-1 activity provides a way for cells to limit abrogated inflammatory responses, potentially without inducing cell death. Pyroptosis is likely the predominant and easiest way for cells to downregulate inflammatory responses, particularly in cells like macrophages where robust inflammasome activation results in a strong and rapid burst in caspase-1 activity. However, in cell types such as dendritic cells, monocytes and neutrophils, where more moderate or weak inflammasome activation occurs, slower kinetics of caspase-1 activation may allow for cytokine processing and release while also allowing the cell time to repair GSDMD pores (208).

The critical finding in this study is that caspase-1 is only active within the inflammasome complex. Most inflammasomes, namely NLRP3, localize to specific areas in the cell, meaning that caspase-1 substrates must come in direct contact or close proximity to inflammasomes in order to be cleaved by caspase-1. The proposed 15 minute window for active caspase-1 in nigericin-stimulated macrophages would add an additional constraint for caspase-1 target selection, as this time frame may not allow significant trafficking of target proteins to areas of active inflammasomes prior to rapid caspase-1 inactivation. These data support the previous observations that while caspase-1 is quite promiscuous in its substrate repertoire, in cells, the caspase-1 substrate repertoire is relatively small (49, 206, 209). 40

In addition to dictating the kinetics of caspase-1 activation, the inflammasome size and cell type may also direct caspase-1 target proteolysis. Walsh et al. found that caspase-1 substrate cleavage was dependent on the concentration of caspase-1 (206). For example, caspase-1 cleaves its well establish

target proIL-1β at a wide range of concentrations ranging from 1-100nM, while caspase-1 promiscuity

increased at higher concentrations ~100nM (206). In contrast to robust inflammasome activation in

nigericin-stimulated macrophages, moderate inflammasome activation or ASC-independent inflammasomes may result in a lower concentration of caspase-1 within the signaling hub that favors caspase-1 cleavage of specific proteins. In other words, caspase-1 cleavage of target substrates may depend on the inflammatory stimuli driving caspase-1 activation (209).

Other Inflammatory Caspases

In addition to caspase-1, human caspase-4 and caspase-5 are involved in initiating inflammatory responses. CASP4 and CASP5 are the human orthologues of murine Casp11, but the current understanding of caspase-4/5 activation and the mechanisms that regulate these caspases is still limited.

Like caspase-11, caspase-4/5 undergo autoproteolysis following recognition of cytosolic LPS, driving

noncanonical inflammasome activation. While caspases-4/5 do not cleave proIL-1β or proIL-18 into

bioactive cytokines, they each can cleave GSDMD to drive pyroptosis (82, 192, 193, 210), which

subsequently triggers NLRP3/caspase-1 activation and IL-1β/IL-18 secretion. In addition to the different

caspases in humans and mice, human GBPs differ from murine Gbps due to multiple gene amplification

in mice, and the Irg resistance system present in mice (i.e. IRG10) is absent in humans (211, 212).

Therefore, the mechanisms of noncanonical inflammasome activation are different in humans.

Recently, caspase-4 was shown to be central to the inflammasome response in human

macrophages, as CASP4 deletion protects cells from cell death in response to LPS transfection (213).

Cytosolic LPS derived from Francisella or Bacteroides vulgantus is not sensed by caspase-11 in

BMDMs; however, caspase-4 in hMDMs recognizes both species of LPS, albeit with lower efficiency

than LPS from E. coli (211). LPS from these bacteria are under-acylated, composed of tetra-acylated or

monophosphoryl penta-acylated lipid A (211), as opposed to the hexa-acylated lipid A species required 41

for caspase-11 oligomerization and activation. This suggests that human caspase-4 has evolved to sense

more diverse LPS ligands. Ectopically expressing human caspase-4 in BMDMs renders them susceptible

to Francisella LPS-mediated cell death, confirming that human caspase-4 has an intrinsic broader

sensitivity to LPS species (211). Like in mice, human GBPs cooperate with caspases for LPS sensing and

noncanonical inflammasome activation. Specifically, GBP2, and to a lesser extent GBPs 1, 3 and 4,

promote caspase-4 recognition of F. novicida and tetra-acylated LPS in human macrophages, but GBP2 is

not involved in caspase-4 activation in response to hexa-acylated LPS (211). F. novicida infection in

BMDMs elicits Aim2 inflammasome activation, but AIM2 is not the primary sensor of this bacterium in

hMDMs, likely due to the absence of IRG10 in human cells, which could impair the bacteriolysis event

required for dsDNA exposure (211). While caspase-4 is critical for LPS sensing, caspase-5 is required for inflammasome responses to bacterial infections and acts syngersitically with caspase-4 to regulate pyroptosis and downstream IL-1β production (213).

Caspase-8 is a well-established initiator caspase that plays a central role in apoptosis. More recently, caspase-8 has emerged as a critical player in inflammatory responses. Like caspase-1, caspase-8 processes IL-1β and IL-18 into their mature form (214-220) and can be recruited into inflammasome complexes, although caspase-8 is less efficient at triggering cytokine maturation compared to caspase-1

(125, 219). NLRP3/ASC inflammasomes recruit caspase-8 in the absence of caspase-1, although in contrast to the rapid kinetics of caspase-1 activation and substrate cleavage, caspase-8 drives late maturation of bioactive IL-1β (~2-4h post nigericin stimulation) and cell death (219). An inflammasome can also form with Naip5, NLRC4 and caspase-8, and caspase-8 recruitment into this inflammasome complex requires ASC (125, 221). Interestingly, caspase-8 recruitment into the NLRC4 inflammasome occurs in WT cells, however caspase-8 is only activated in conditions when caspase-1 or GSDMD are inhibited (125). GSDMD-dependent pyroptosis suppresses caspase-8 activation, but in the absence of

GSDMD, inflammasomes can engage a delayed, alternative form of lytic cell death (222). This inflammasome/caspase-8/GSDMD-independent proinflammatory form of cell death is thought to be unique among the other cell death modalities (222). Caspase-8, therefore, functions as a backup strategy 42

to guarantee cell death when components of the pyroptotic pathway are inhibited. Caspase-8 may drive

cell death by activating caspase-3/7 to target gasdermin-E (also known as DFNA5), which also induces

pore formation (125, 223, 224), though DFNA5 was shown to be dispensable for cell death following

NLRC4/caspase-8 activation in vitro (225). Regardless of the mechanism, inflammasome/caspase-8

activation-dependent cell death may be an essential host defense strategy against pathogens that have evolved to inhibit canonical inflammatory components.

Biosensors and Novel Tools to Monitor Inflammation

Monitoring the proteolytic activation of caspase-1 has remained the gold-standard technique for

detecting inflammatory responses, since a myriad of PAMPs and DAMPs can activate numerous

inflammasomes, all of which culminate in the activation of caspase-1. The direct analysis of caspase-1

cleavage by western blot requires the removal of tissues from sacrificed animals or harvesting cells at

defined time points. This can result in limited accuracy since the analysis requires different cell

populations or animals at each time point and prevents the dynamic measurement of

inflammasome/caspase-1 activation over time. Other assays that allow caspase-1 activation and

inflammatory responses to be quantified with high spatial and temporal resolution would contribute

substantially to the study of inflammatory diseases.

A BRET-based biosensor was developed in 2012, in which the proIL-1β protein was fused to the

C-terminus of RLuc8 (the donor) to the N-terminus of Venus (the acceptor) (226). Energy transfer between the donor and acceptor occurs when proIL-1β is intact and RLuc8 and Venus are in close

proximity (<100Å), which results in Venus emission at ~535nm (226). Caspase-1 mediated cleavage of proIL-1β causes the N- and C-terminals to move apart and this is accompanied by a loss in energy transfer that can be quantified (226). This R-proIL-1β-V sensor allows for real-time ratiometric quantification of processed IL-1β in macrophage cell lines and BMDMs following nigericin stimulation

(226), a DAMP that drives robust IL-1β processing. By combining the BRET technique with live cell microscopy, BRET signals can also be quantified in individual single cells before and after the addition of 43

inflammasome-activating stimuli (226). This technique provides a significant improvement in the temporal resolution of IL-1β dynamics over standard WB and ELISA assays.

In 2013, Veit Hornung’s group published the iGLuc biosensor, a proIL-1β-gaussia luciferase fusion construct in which caspase-1-mediated cleavage leads to monomerization of this protein and increased luciferase activity that can be quantified (227). In response to inflammasome-activating stimuli,

iGLuc-expressing macrophages secrete the processed reporter alongside endogenous IL-1β in the

supernatant, allowing the sensor to be monitored in the supernatant without lysing cells (227). This construct was indirectly applied in vivo by transferring cells expressing the reporter construct into animals

i.p and monitoring biosensor activation following listeria infection (227). Promega published a luciferase

substrate called z-WEHD aminoluciferin in 2017, which is cleaved in response to inflammasome-

activating stimuli (228). In this system, cells are incubated 1:1 with a lytic reagent containing a

proteasome inhibitor and the luciferin substrate and luminescence can be quantified (228). While all of

these probes effectively monitored inflammasome activation in vitro, they have limitations in vivo. These

include the limitations associated with tissue autofluorescence, poor tissue penetration in the case of

substrates or short-lived signal following biosensor release into the extracellular space.

Recently, a non-invasive fluorescent caspase-1 probe was developed that can detect caspase-1

activation in cells and in inflamed tissues. The Cas-1 probe was synthesized by conjugating the WEHD-G

caspase-1 substrate to Cy5.5 and Black Hole Quencher-3 (BHQ-3), a quencher of fluorescent dyes that

emit in the red/far-red spectrum (229). Under normal conditions, the quencher suppresses Cy5.5

fluorescence, but following caspase-1 mediated cleavage of the WEHD-G linker, the quencher is

separated from Cy5.5 and the fluorescent signal can be detected and quantified (229). The membrane

permeable Cas-1 probe is added to cells prior to the addition of inflammatory stimuli, and fluorescent

activity can be monitored over time in cells where caspase-1 is active (229). The probe could also be

injected intravenously into mice at various time points following treatment with inflammatory stimuli,

such as LPS or DSS, to monitor fluorescence in vivo 30 minutes to 2 hours after injection or in tissues ex

vivo (229). The probe was also applied to study caspase-1 activation in the context of cancer models and 44

Alzheimer’s disease (229). The ability to inject a probe and non-invasively track inflammatory responses has significant advantages over other technologies, as it has potential clinical applications (229).

Advances in the development of protease biosensors, including circularly permutated firefly luciferases led to the development of a caspase-3/7 specific bioluminescent reporter called GloSensor, published in 2013 (230-232). Galban et al. generated tissue-specific transgenic mice expressing the

GloSensor apoptotic reporter to monitor apoptotic responses in vivo (230). In transgenic mice where Cre- activity localized to the pancreas, they observed GloSensor activation in the pancreas following repeated administration of cerulein, an inducer of apoptotic cell death due to pancreatitis (230). The caspase3/7

GloSensor was also used for high throughput screens to test pharmacological compounds that induce caspase3/7 activation (230). This biosensor has since been used by other groups to study various other proteases (233) and was the basis of the development of our caspase-1 biosensors.

CHAPTER THREE

MONITORING CASPASE-1 BIOSENSOR ACTIVATION IN VITRO

Materials and Methods

Cell Culture

293T and THP-1 cells were obtained from the American Type Culture Collection (ATCC). Cells were cultured at 37˚C in 5% CO2 in DMEM (293T cells) or RPMI (THP-1 cells / MDMs), respectively, supplemented with 10% Fetal Bovine Serum (FBS) (Gibco) with 100 U/mL penicillin, 100 U/mL streptomycin and 10 μg/mL ciprofloxacin. MDMs were additionally cultured with 50 ng/mL GM-CSF and M-CSF. THP-1 cells were treated with 100 ng/mL PMA for 48h for differentiation into macrophages.

293T Transfections and Luciferase Reporter Assay

293T cells were plated at ~70 percent confluence in DMEM. Cells were transfected with a pGLO sensor construct containing the indicated target sequence, along with ASC and procaspase-1 expression plasmids using polyethylenimine (PEI, Polysciences). 24-36h later, cells were lysed in Passive Lysis

Buffer (Promega). Firefly luciferase substrate (Promega) was added to each well and luminescence (RLU) was quantified.

Generation of Stable Cell Lines

Caspase-1 biosensors containing the indicated caspase-1 recognition sequence were synthesized

(IDT) and cloned into pLVX (XhoI and EcoRI). Lentiviral vectors were generated by transfecting 293T with equal ratio of each biosensor, psPAX2 packaging construct (from Dr. Didier Trono, NIH AIDS

Reagent Program, Cat. #11348) and vesicular stomatitis virus glycoprotein (VSVg) using PEI. After 48h, supernatant was harvested, filtered through a 0.45 μM filter (Millipore) and added to cells. Cells were transduced by spinoculation (1,200g, 2h at 13˚C) (234). After 48h, the media was replaced with media containing 5 μg/mL puromycin.

Monocyte Derived Macrophages

To generate MDMs, PBMCs were isolated from healthy donors and monocytes were positively

selected using a human CD14 positive selection kit following the manufactures protocol (STEMCELL).

MDMs were differentiated for 7 days with GM-CSF and M-CSF (50ng/mL). After differentiation, SIV-

VSVg VLPs were added to cells overnight. The next day, cells were transduced with lentiviruses to

express the indicated biosensor. Inflammasome activation assays were performed 3 days later.

In vitro Inflammasome Activation Assays

Differentiated THP-1 cells were plated in 48 well or 24 well plates and stimulated with the

indicated stimuli. Cells were primed with 10 ng/mL LPS (from E. coli 055:B5, Sigma-Aldrich) for 4h

followed by 5 mM ATP (Invivogen) or 6.5uM nigericin (Sigma-Aldrich) 3h. Supernatant and cells were

harvested for luciferase assays. Alternatively, cells were transfected with 10 μg/mL LPS, 10 μg/mL

poly(dA:dT) (Sigma-Aldrich) or 100 μg/mL poly(I:C) (Sigma-Aldrich) with lipofectamine in media

without antibiotics. 6h later, supernatant and cells were harvested for ELISA and luciferase assays. For

caspase-1 inhibitor experiments, THP-1 cells were treated with Z-WEHD-FMK (R&D systems) or

DMSO control for one hour prior to LPS stimulation. For in vitro luciferase measurements, cells were

lysed and bioluminescence was quantified as described above.

Salmonella Infection

The Salmonella strain used in these experiments contained a duel fluorescence reporter system

described previously, called pDiGc (235). Bacteria were cultured overnight then subcultured 1:33 in LB

with arabinose for 3h. OD was measured and bacteria were spun down 10,000g for 2 minutes, washed

twice in PBS and then resuspended in RPMI. 2.5 x 107 bacteria were added to 500,000 IQAD or GSDMD

THP-1 cells per well in a 24 well plate (MOI = 50). Bacteria were spun down onto THP-1 cells for 5 minutes. The Mock coverslip was fixed immediately following spinoculation. 30 minutes post-infection, media was replaced with media containing 100μg/mL gentamycin to kill bacteria that had not entered the cells. 1.5h post-infection, the media was replaced with media containing 15μg/mL gentamycin, as excess antibiotics could be taken up by macrophages and cause intracellular killing of the bacteria. Coverslips or 47

cells were harvested at the indicated time point and fixed for immunofluorescence or lysed for luciferase

measurements.

Statistical Analysis

Statistical significance was assessed using GraphPad Prism software (GraphPad Software, Inc.).

Student t test was employed when comparing a treatment group relative to NT. One way ANOVA was

performed when comparing a group of data, followed by either a Turkey Post Hoc test or Dunnett’s

Multiple Comparison test to determine significance. Data are represented as means +/- standard error of

the mean (SEM) or standard deviation (SD). P < .05 was considered significant.

Results

Caspase-1 Biosensor Design

To better understand the role of caspase-1 in various disease settings, we developed a series of

bioluminescent sensors that readily detect caspase-1 activation. We employed a circularly permuted form

of luciferase, in which the N and C-terminal domains necessary for bioluminescence are physically

separated by a flexible hinge region (231). Putative caspase-1 target sequences were cloned into this

linker region, with the expectation that upon caspase-1 activation, these sequences would be cleaved,

allowing the two domains of luciferase to come together (Fig. 8). This results in a strong gain in

luciferase activity, so that the degree to which caspase-1 is activated can be quantified. The biosensors

generated contained the following caspase-1 target sequences: two cleavage sequences within proIL-1β

(214, 236), proIL-18 (237), procaspase-7 (238-240), gasdermin-D (192), GAPDH (241), a synthetic caspase-1 target, FESD and the well-established consensus sequence for all caspase-1 targets, WEHD

(242) (Fig. 8). 48

Figure 8. Caspase-1 Biosensor Design. Schematic of the biosensor design. The indicated caspase-1 target amino acid sequence (red) was cloned into the circularly permuted luciferase (promega) and subsequently cloned into a lentiviral backbone to express this construct in cells. Caspase-1 cleavage of its target sequence will result in a gain in luciferase signal, which can be quantified in cells.

Screen to Identify Putative Biosensors

We tested the ability of these biosensors to measure caspase-1 activation by transiently expressing each construct in 293T cells. To reconstitute an inflammasome pathway in these cells, we additionally co-transfected procaspase-1 and ASC (226, 227) or a GFP sham, and luminescence was measured. We detected robust activation of biosensors containing all caspase-1 target sequences (Fig. 9).

Figure 9. Caspase-1 Biosensors are Activated in 293T Cells. 293T cells were transfected with the indicated biosensor or pcDNA control in the presence of human procaspase-1 or ASC expression plasmids, or GFP sham control. 36h post-transfection, cells were lysed and luciferase was quantified. Data are expressed as fold RLU relative to luciferase detected in the GFP sham.

Monitoring Biosensor Activation in THP-1 cells

To evaluate putative caspase-1 biosensors in a more biologically relevant setting, we transduced

the human monocytic THP-1 cell line with lentiviral vectors expressing these biosensors. THP-1 cells can

be terminally differentiated into macrophages with PMA, giving us a relevant cell line for studying

caspase-1. Following differentiation, cells were primed with LPS for 4h, then stimulated with the ATP or nigericin, a bacterial toxin derived from Streptomyces hygroscopicus that acts as a potassium ionophore.

Both stimuli drive potassium efflux, providing signal 2 for NLRP3 inflammasome activation. We detected the most robust biosensor activation in our THP-1 lines expressing biosensors with either the

IQAD or GSDMD target sequence, as well as cells expressing the first caspase-1 target sequence within pro-IL-1β at position D26 (Fig. 10). Surprisingly, little to no bioluminescence was detected with our

WEHD biosensor, despite the high signal we measured in 293T cells (Fig. 9-10). The other biosensors produced moderate luciferase signals and may be more valuable in studying caspase-1 activation in other models of inflammation. For example, caspase-1 targets enzymes involved in glycolysis and was shown to process GAPDH in response to Salmonella infection, so the GAPDH biosensor may be relevant to metabolic or bacterial infection studies (241). Together, these data indicate that caspase-1 biosensors are functioning in vitro. 50

Figure 10. Caspase-1 Biosensors are Activated in THP-1s. THP-1 cells were transduced with the lentiviruses to generate cell lines expressing biosensors with the indicated caspase-1 target sequence. Cells were differentiated for 48h, then treated with 10ng/mL LPS for 4h followed by 5mM ATP or 6.5μM nigericin, and luciferase signal was quantified 3h later. NT = no treatment.

Based on data from our biosensor screen in THP-1 cells, we moved forward with our two best biosensors, IQAD and GSDMD. To further characterize our IQAD biosensor in vitro, cells were transfected with LPS, poly(I:C) or poly(dA:dT) and bioluminescence was quantified 6h later. We detected strong biosensor signal in response to poly(I:C) and poly(dA:dT), drugs that activate NLRP3 and AIM2 inflammasomes, respectively (135, 137, 243), while minimal signal was measured in response to intracellular LPS, which activates the noncanonical inflammasome (Fig. 11).

Figure 11. Caspase-1 Biosensors are Activated in THP-1s in Response to NLRP3 and AIM2 Agonists. THP-1s expressing the IQAD or GSDMD biosensors were differentiated for 48h, then transfected with lipofectamine in the presence of 10μg/mL LPS, 100μg/mL poly(I:C) or 10μg/mL poly(dA:dT) and luciferase was measured 6h later.

We next wanted to determine if caspase-1 biosensors were turned on during an infection. The

enteric pathogen, Salmonella enterica serovar Typhimurium (S. typhimurium), is a gram negative rod- 51

shaped intracellular bacterium that infects both humans and animal hosts. This bacterium is capable of activating inflammasomes via multiple pathways. TLR4 senses bacterial LPS, a major constituent of the outer membrane of gram-negative bacteria such as S. typhimurium, while TLR5 can recognize extracellular flagellin, both of which provide signal 1 for NLRP3 priming (244). Intracellular S. typhimurium can also disrupt host cell metabolism, driving alterations in mitochondrial ROS and NLRP3 activation (245, 246). The presence of cytosolic flagellins and the T3SS of S. typhimurium are capable of activating NLRC4 (115, 116). Both NLRP3 and NLRC4 inflammasomes are critical for host defense against S. typhimurium, as mice lacking both NLRs are significantly more susceptible to infection (244,

247). To determine if the IQAD and GSDMD biosensor were activated in response to S. typhimurium infection, THP-1 cells expressing each biosensor were infected and biosensor activation and bacterial replication were measured. We used a dual fluorescence reporter system in which GFP is constitutively expressed while arabinose-induced DsRed is diluted as the bacteria replicate (235), enabling us to identify

infected cells and determine if the bacteria are replicating by immunofluorescence analysis, while

monitoring biosensor activation in parallel. Biosensor activation gradually increased from 1h post

infection to 6h post infection, and then started to decline by 8h to 24h (data not shown). Importantly, we

detected increased luciferase signal 6h post-infection, at a time point where the bacteria were actively

replicating in the host cells (Fig 12).

Figure 12. Biosensors are Activated during Bacterial Infection in THP-1 Cells. Differentiated IQAD or GSDMD-expressing THP-1 cells were infected with Salmonella at an MOI of 50. Bacteria were spun down onto cells for 5 minutes. 30 minutes post-infection, the media was replaced with media containing gentamycin to kill extracellular bacteria. At the indicated time points, coverslips were harvested for immunofluorescence to assess bacterial replication or cells were lysed for luciferase reads to measure biosensor activity. Data are representative from one experiment, expressed as fold change in luciferase relative to the uninfected cells. NT = no treatment.

Monitoring Biosensor Activation in MDMs

Next, we isolated monocytes from human PBMCs and differentiated these cells into macrophages

(MDMs). Primary human macrophages express retroviral restriction factors making them difficult to

transduce with lentiviruses. SAMHD1 is a dNTPase that regulates dNTP pools and is a potent restriction

factor for HIV-1, limiting reverse transcription. The SIV accessory protein, lentiviral protein x (Vpx), targets SAMHD1 for proteasomal degradation, increasing dNTP pools and relieving reverse transcription

(248, 249). To circumvent SAMHD1 activity, we pretreated cells with vpx-containing virus like particles

(VLPs) and then transduced these MDMs with lentiviral vectors expressing biosensors of interest.

Biosensor signal was measured following treatment with the indicated inflammatory stimuli. Consistent with THP-1 cells, we detected activation of all biosensors in response to NLRP3 and AIM2 activation

(Fig. 13, data not shown). Together, these data reveal that caspase-1 biosensors are functioning in vitro in various cell types in response to stimuli that activate different inflammasomes.

Figure 13. Caspase-1 Biosensors are Activated in MDMs. Monocytes were purified from PBMCs isolated from healthy donors (CD14+ selection) and differentiated into macrophages with GM-CSF and M-CSF for 7 days. Cells were treated with SIV-VSVg VLPs on day 8, and then transduced with lentiviral vectors to express the indicated biosensor on day 9. 3 days later, cells were treated with LPS and ATP or nigericin or transfected with poly(I:C) or poly(dA:dT) and biosensor activation was quantified.

IQAD Biosensors are Activated in Response to Inflammatory and Apoptotic Stimuli

While the IQAD recognition sequence in procaspase-7 is a well-established target of caspase-1 in vitro and in vivo (238-240), procaspase-7 is also activated by initiator caspases caspase-8, -9 and -10 during apoptosis (238). To confirm the IQAD biosensor was cleaved by caspase-1, cells were treated with an inhibitor of caspase-1, z-WEHD-FMK, then stimulated with LPS and ATP as before. We detected a dose-dependent decrease in biosensor signal in the presence of increasing concentrations of the caspase-1 inhibitor (Fig. 14A). To determine if the IQAD biosensor was activated in cells during apoptosis, THP-1 cells were treated with staurosporine and biosensor activation was quantified. While we detected significant biosensor activation in cell lysates following LPS and ATP treatment, we observed minimal biosensor activation in lysates following staurosporine treatment (Fig. 14B). However, we did observe significant biosensor signal in the supernatant of staurosporine treated cells, while we did not observe a similar increase in the supernatant of cells treated with LPS and ATP (Fig. 14B). These data suggest that although the IQAD biosensor can be activated in response to inflammatory and apoptotic stimuli, the biosensor is released from apoptotic cells and retained in cells following inflammasome activation.

Figure 14. IQAD Biosensors are Activated in Cells in Response to Inflammatory Stimuli but Secreted from Apoptotic Cells. IQAD THP-1 cells were pre-treated with the indicated concentration of z-WEHD-FMK for one hour, then treated with or without LPS and ATP and bioluminescence was quantified (A). Biosensor activation measured in the cell lysate or supernatant from IQAD THP-1 cells stimulated with LPS and ATP as before, or with 10μM staurosporine for 16h (B).

CHAPTER FOUR

GENERATION OF CASPASE-1 REPORTER MICE

Materials and Methods

Generation of Caspase-1 Reporter Mice

Caspase-1 biosensors containing the IQAD target sequence were synthesized (IDT) and cloned into the EcoRI and XhoI sites of pCAG. SalI and HindIII digested DNA containing the CAG promoter and intron, caspase-1 biosensor and polyA sequence was purified and microinjected into zygotes of

C57Bl/6 mice (performed by Northwestern Transgenic and Targeted Mutagenesis Laboratory). Progeny were genotyped by PCR of the tail DNA and transgene-positive mice were identified. The founder line that exhibited ubiquitous and robust transgene expression in all tissues was expanded. For transgene expression quantification, mouse tissues were homogenized, RNA was isolated (NucleoSpin RNA Plus,

Macherey-Nagel) and converted to cDNA (GoScript Reverse Transcription System, Promega). Transgene or mouse β-actin expression was quantified by qRT-PCR using SYBR green (Bio-Rad). Caspase-1 biosensor mice are maintained as heterozygote. All experiments were performed according to protocols approved by the Institutional Animal Care and Use Committee (IACUC) at Loyola University Chicago.

All animals were housed in pathogen-free conditions at Loyola University Chicago medical campus

(Maywood, IL); animals used for S. aureus experiments were housed in a BSL-2 facility at Loyola

University Chicago medical campus (Maywood, IL).

In vitro Inflammasome Activation Assays

BMDMs were plated in 48 well or 24 well plates and stimulated with the indicated stimuli. The following drug concentrations were used: 10ng/mL LPS 4h (priming), 1μg/mL LPS 16h (overnight),

10μg/mL LPS (transfection) 6h, 5mM ATP 3h, 6.5μM nigericin 3h, 1μg/mL Pam2CSK4 4h, 10μg/mL

Pam3CSK4 4h, 100μg/mL curdlan 24h, 125μg/mL MSU 4h, 2μg/mL flagellin 6h, 100μg/mL poly(I:C)

6h, 10μg/mL poly(dA:dT) 6h. Cells were with LPS for 4h followed by ATP or nigericin for 30m-1h.

Alternatively cells were transfected with LPS, poly(I:C) or poly(dA:dT). Luciferase was quantified as

described above. IL-1β ELISA was performed according to the manufacturer’s instructions (RayBiotech).

Cytometric bead array (CBA) was performed according to the manufacturer’s instructions and samples

were analyzed on an LSRFortessa (BD Biosciences). The log2fold change was calculated for each

treatment group and heatmaps were generated using RStudio.

Western Blotting

Cells were stimulated with LPS and ATP as described above. Supernatant and cells were

harvested 3h after ATP treatment. Whole-cell lysates were prepared by lysing cells in NP-40 lysis buffer

(100 mM Tris pH 8.0, 1% NP-40, 150 mM NaCl) containing a protease inhibitor cocktail (PIC) (Roche) for 30 minutes on ice. Lysates were collected, mixed with 2x Laemmli sample buffer and incubated at

100˚C for 5 minutes. For detection of caspase-1 in the supernatant, proteins were concentrated from 500

μL of supernatant by methanol/chloroform precipitation. Protein pellets were suspended in lysis buffer with protease inhibitor cocktail (PIC) and sample buffer and incubated at 100˚C for 5 minutes. For detection of proteins in tissues, tissues were homogenized in lysis buffer with PIC and homogenates were shaken on ice for 30 minutes – 1 hour. Homogenates were centrifuged at 3900 rpm for 10 minutes, lysates were collected and protein concentration was quantified by BCA (Pierce, Thermo Fisher). Proteins were mixed with 2x Laemmli sample buffer and boiled as described above. Proteins were loaded onto a 4-15% gradient gel (Bio-Rad) and transferred to a nitrocellulose membrane (Bio-Rad). Membranes were incubated with anti-caspase-1 (Adipogen) or anti-β-actin (Santa Cruz Biotechnology) followed by anti- mouse IgG conjugated to HRP (Thermo Scientific). Antibody complexes were detected using

SuperSignal West Fempto Chemiluminescent Substrate (Thermo Scientific) and chemiluminescence was measured in a FluorChem E machine (ProteinSimple).

Lactate Dehydrogenase (LDH) Measurements

LDH cytotoxicity assay was performed according to the manufacturer’s instructions (Pierce,

Thermo Scientific). Briefly, supernatants were collected and centrifuged to remove cell debris. Maximum 57

LDH activity was determined in supernatant from cells incubated with 10x Lysis Buffer. Background

LDH activity present in the serum was measured using culture media.

IVIS Imaging

Mice were anesthetized using 2% isofluorane/air mixture delivered by the Xenogen IVIS

Spectrum XGl-8 Gas Anesthesia system and injected i.p or i.v with a single dose of 150 mg/kg VivoGlo

Luciferin for in vivo bioluminescence imaging. Anesthetized mice were placed in the IVIS imaging chamber and imaged with the IVIS 100 Imaging System (Xenogen) 10 minutes after i.p administration of the luciferase substrate or immediately following i.v. administration of the substrate. Bioluminescence was quantified and analyzed use Living Image software.

Statistical Analysis.

Statistical significance was assessed using GraphPad Prism software (GraphPad Software, Inc.).

Student t test was employed when comparing a treatment group relative to NT. One way ANOVA was

performed when comparing a group of data, followed by either a Turkey Post Hoc test or Dunnett’s

Multiple Comparison test to determine significance. Data are represented as means +/- standard error of

the mean (SEM) or standard deviation (SD). P < .05 was considered significant.

Results

Generation of Caspase-1 IQAD Biosensor Transgenic Mice

Transgenic mice expressing the IQAD biosensor were generated by microinjection of a transgene

containing the biosensor under control of the chicken β-actin promoter and cytomegalovirus (CMV) early

enhancer for ubiquitous transgene expression (250-252) (Fig. 15A). The transgene was purified and microinjected into zygotes of C57Bl/6 mice (Fig. 15B). Offspring were genotyped by PCR of the tail

DNA to identify positive founders. Four founder lines were initially identified (Fig. 15C), one of which successfully delivered positive progeny. This line exhibited ubiquitous transgene expression in all tissues and was expanded and used for all future experiments (Fig 15D-E).

Figure 15. Generation of Caspase Reporter Mice. Schematic of transgene, containing the CAG promoter -- CMV enhancer and chicken β-actin promoter followed by a chimeric intron driving expression of the IQAD biosensor followed by a β-globulin poly(A) signal (A). The digested transgene was purified and microinjected into zygotes (B) and founder lines were identified by PCR of tail DNA (C). RNA was isolated from various tissues in transgenic and negative control mice, converted to cDNA and qRT-PCR was performed to identify founder lines with ubiquitous transgene expression in all tissues (D). Representative image of luciferase signal detected following tail vein injection of luciferase substrate in mouse from the founder line used in these experiments (E).

We next examined biosensor activation in internal organs ex vivo, including the thymus, which is known to have constitutively high levels of apoptosis due to ongoing negative selection of developing T cells. When the thymus was excised from positive transgenic mice and bioluminescence was measured, we observed minimal biosensor activation in the thymus ex vivo, relative to other tissues isolated (Fig.

16). Thus, although the IQAD biosensor is activated by apoptotic stimuli in vitro, its ability to serve as a biosensor of apoptosis in vivo is poor, potentially due to degradation of the biosensor in the extracellular environment or the rapid clearance of apoptotic cells in vivo (253, 254).

Figure 16. Biosensor Signal ex vivo IQAD Mice. Representative images of biosensor signal in caspase-1 biosensor mice. The thymus (red square) along with all other tissues isolated from the mouse. Thymus (1), heart (2), liver (3), lung (4), small intestine (5), spleen (6), large intestine (7) and kidney (8).

Monitoring Biosensor Activation in BMDMs

To determine if the IQAD biosensor was functioning in these transgenic mice, bone marrow cells

from transgenic mice were isolated and differentiated into macrophages (BMDMs). BMDMs were treated

with a panel of inflammatory stimuli and the biosensor activation generated from proteolytic cleavage of

caspase-1 and IL-1β secretion was measured. As expected, increased luciferase activity and IL-1β

secretion were detected in BMDMs stimulated with LPS and ATP (Fig. 17A-B) and activated caspase-1 was detected in the supernatant of LPS and ATP stimulated BMDMs by western blot analysis (Fig. 17C).

We also detected robust biosensor activation in BMDMs transfected with poly(dA:dT) and poly(I:C),

similar to THP-1 cells and MDMs (Fig. 17D). To determine if these treatments were inducing cell death,

lactate dehydrogenase (LDH) levels were measured in the supernatant, as LDH is normally found in cells

but is released when cell membrane integrity is compromised and is thus a measurement of cytotoxicity.

We did detect cytotoxicity in response to some inflammatory stimuli, including poly(dA:dT), which is

known to induce pyroptosis, and poly(I:C), while no significant cell death was detected in cells treated

with LPS and ATP (Fig. 17E). We also performed cytokine bead array (CBA) to assess cytokine and chemokine secretion in parallel with biosensor activation. In addition to IL-1β, LPS-primed cells treated with ATP secreted a panel of inflammatory cytokines, including, IL-1α, IL-6, and TNF as well as the chemokines keratinocyte chemoattractant (KC) CXCL1, monocyte chemoattractant protein-1 (MCP-1), macrophage inflammatory proteins (MIP-1α, MIP-1β) and RANTES (CCL5) (Fig. 17F). Together these 60

data suggest that caspase-1 biosensors are activated in BMDMs in response to various inflammatory

stimuli.

Figure 17. Caspase-1 Biosensors are Activated in BMDMs. BMDMs from caspase-1 reporter mice were stimulated with 10ng/mL LPS for 4h followed by 5mM ATP for 30m or 1h and luciferase was measured in cells, and IL-1β and caspase-1 were measured in the supernatant (A-C). BMDMs were transfected with lipofectamine and 10ng/mL poly(dA:dT) or 100ng/mL poly(I:C) and luciferase was quantified 6h later (D). LDH was measured in the cell supernatant of BMDMs stimulated with the indicated treatment (E). The dashed red line indicates the average background LDH activity detected in untreated BMDM supernatant (E). Cytokine bead array was used to quantify cytokine/chemokine levels in the supernatant of BMDMs treated with the indicated stimuli (F). The log2 fold change relative to the NT group was calculated and plotted as a heatmap, where each cytokine/chemokine is displayed on the x-axis (F).

IQAD Biosensors are Not Activated in Response to Some Inflammatory Stimuli

We also tested biosensor activation in response to other TLR agonist such as Pam2CSK4 and

Pam3CSK4, which trigger TLR2-6 or TLR2-1 heterodimers to provide signal 1. Pam2 or Pam3 with ATP

also induced robust biosensor activation and cytokine production, similar to LPS and ATP (Fig. 18A-C).

However, we did not detect biosensor activation in response to all stimuli tested, including monosodium urate crystals (MSU), curdlan, and flagellin transfection (Fig. 18A). Uric acid released from dying cells and can act as a DAMP to drive NLRP3 inflammasome activation (255). Gout is associated with MSU 61

deposition in joints and periarticular tissues due to high levels of uric acid in the blood (255). Increased

uric acid concentrations have also been detected in the serum of patients with MS (256). While MSU is

well established to drive NLRP3 and caspase-1 activation following phagocytosis of crystals, we did not

detect biosensor activation in response to MSU alone or MSU with LPS, although we did detect increased

production of inflammatory cytokines/chemokines by CBA suggesting that an inflammatory pathway was

activated (Fig. 18A-C). Fungal β-glucan, a major component of the fungal cell wall, can also activate the

NLRP3 inflammasome. The large particulate β-glucan curdlan activates NLRP3 by providing signal 1 via

dectin-1 and signal 2 following uptake into the cytosol (257). BMDMs treated with curdlan exhibited

little to no biosensor activation and minimal cytokine/chemokine responses relative to untreated cells

(Fig. 18A-C). We were also unable to detect biosensor activation in BMDMs transfected with flagellin

(Fig. 18A). Lipofectamine transfection of LPS also elicited minimal biosensor activation, though we did

detect cytokine/chemokine production by CBA (Fig. 18A-C). Cytosolic LPS activates the noncanonical inflammasome which ultimately drives caspase-1 activation. The LPS used in these experiments is serotype 055:B5 derived from E. coli, and previous reports have demonstrated serotype-dependent differences in noncanonical inflammasome activation (9). TLR4-priming is not serotype dependent (9), which could explain why we see efficient inflammasome activation in LPS-primed cells but not in transfected cells.

Figure 18. Caspase-1 Biosensors are Activated in BMDMs in Response to Some Stimuli. BMDMs were stimulated with the indicated stimuli and biosensor activation was measured in cell lysates (A), IL-1β levels were measured in the supernatant by ELISA (B) and other inflammatory cytokines/chemokines were quantified in the supernatant by CBA (C).

Similarly, despite consistently robust biosensor activation in THP-1 cells and human MDMs

treated with LPS and nigericin, we detected low luminescence in LPS and nigericin treated BMDMs (Fig

19A). We still detected IL-1β secretion by ELISA and caspase-1 activation by western blot and FLICA in

cells stimulated with nigericin, indicating that the treatment is activating caspase-1 (Fig. 19B-D). We also did not detect significant cytotoxicity as measured by LDH (Fig 19E) and no biosensor signal was detected in the cell supernatant of LPS and nigericin stimulated cells (Fig 19F), indicating that the sensor is not leaking out due to pyroptosis. LPS and nigericin-treated BMDMs exhibited robust expression of various cytokines and chemokines by CBA, similar to ATP –treated cells (Fig. 18C). Together these data

indicate that caspase-1 is active in BMDMs treated with LPS and nigericin, but biosensor activation did

not occur.

Figure 19. Caspase-1 Biosensors are Not Activated by Nigericin in BMDMs. BMDMs were treated with LPS and ATP or nigericin and luciferase was measured in cells (A), IL-1β was measured in the supernatant (B), caspase- 1 activation was measured by WB (C), caspase-1 activation was measured by quantifying FLICA-positive cells by immunofluorescence (D), LDH was measured in the supernatant (E), and luciferase was measured in cell lysates and supernatants (F).

CHAPTER FIVE

MONITORING CASPASE-1 BIOSENSOR ACTIVAITON IN VIVO

Materials and Methods

S. aureus Culture and Murine Systemic Infection Model

Caspase-1 biosensor mice were infected with 107 CFU S. aureus as described previously (258-

260). Briefly, S. aureus (USA300 strain LAC) cultures were grown in tryptic soy broth (TSB) overnight

at 37˚C and subcultured 1:100 in fresh TSB at 37˚C for 3 hours. Cultures were centrifuged, bacterial

pellets were washed in PBS and normalized to an inoculum of 108 CFU/mL. Mice were anesthetized via intraperitoneal injection with 250 mg/kg Avertin and infected via the retro-orbital plexus with 100 μL

PBS containing 107 CFU S. aureus. Tissues were isolated from infected animals and homogenized using an electric homogenizer. CFU were enumerated by plating tissue homogenates on tryptic soy agar (TSA) plates for 16 hours at 37˚C. Tissue homogenates were centrifuged at 3900rpm for 10 minutes and supernatant was collected for CBA analysis using BD CBA Flex sets and mouse inflammation kit (BD

Biosciences).

DSS-Induced Colitis Model

Caspase-1 biosensor mice were exposed to DSS as described previously (261). Briefly, mice were given molecular grade water with or without 2% DSS (40,000 kDa; MP Biomedicals) for five days. For female mice, DSS treatment was extended 48 hours and DSS concentration was doubled to 4% for the final 24 hours. Mice were weighed daily throughout the course of the experiment. On day 5 or day 7, mice were injected with luciferase substrate and imaged in vivo. Blood samples were collected and mice were sacrificed. Colons and brains were excised and bioluminescence was measured ex vivo. Colon length

was measured and fecal samples were assessed for blood using a hemoccult test (Beckman Coulter).

Tissue sections were fixed in 10% phosphate buffered formalin, paraffin embedded, processed and

stained with H&E by AML laboratories (Saint Augustine, Florida) or homogenized and used for western

blot analysis of caspase-1 activation. Colon samples were weighed and homogenized and 5μL of

homogenates were used for CBA (BD Biosciences).

Graft-Versus-Host Disease Model

On day 0, WT BALB/c mice were irradiated. Caspase-1 biosensor mice were sacrificed and bone marrow cells were harvested as described above. For splenocyte isolation, spleens were pressed through a sterile 100 μm cell strainer. The cell strainer was flushed with media to dislodge the cells. Cells were centrifuged for 5 minutes and pellets were resuspended in ACK lysis buffer (Gibco) and washed with medium. Splenocytes and bone marrow cells were counted and resuspended in 300 μL PBS. T cells were

purified (≥ 90% purity, confirmed by flow cytometric analysis of CD3 expression on purified

splenocytes) from total splenocytes by negative selection, using biotin anti-mouse B220 [RA3-6B2], biotin anti-mouse CD11b [M1/70] and biotin anti-mouse CD11c [N418], followed by incubation with magnetic nanoparticles conjugated to streptavidin (BD IMag Streptavidin Particles Plus, BD

Biosciences). Heparin was added prior to intravenous injection. Bone marrow cells (107) +/- 5 x 106

splenocytes or bone marrow cells + 1.3-1.5 x 106 T cells were injected into recipient mice. All donor and recipients were sex matched. Serum was collected from mice on day 7 and cytokine levels were assessed by CBA (BD Biosciences). Clinical scores were obtained by 3 independent researchers. Mice were given a score of 0-2 (2 being most severe) in each of the following categories: activity, posture, fur texture/grooming. Scores were totaled for each mouse. Weight loss was measured separately and was not included in the clinical score. Percentage of donor CD8+ and CD4+ T cells and their activation status was

analyzed by flow cytometry. All antibodies were obtained from Biolegend (anti-H-2Kd [SF11.1], anti-

CD45.1 [A20], anti-CD4 [GK1.5], anti-CD8a [53-6.7], anti-CD44 [IM7], anti-CD25 [PC61], anti-CD69

[H1.23F]).

Statistical Analysis

Statistical significance was assessed using GraphPad Prism software (GraphPad Software, Inc.).

Student t test was employed when comparing a treatment group relative to NT. One way ANOVA was 66

performed when comparing a group of data, followed by either a Turkey Post Hoc test or Dunnett’s

Multiple Comparison test to determine significance. Data are represented as means +/- standard error of

the mean (SEM) or standard deviation (SD). P < .05 was considered significant.

Results

Staphylococcus Aureus Bacterial Infection Model

To determine if caspase-1 biosensors could be used to monitor inflammatory responses in vivo, we chose Staphylococcus aureus (S. aureus) skin and systemic infection models. This gram positive bacterium is typically found as a commensal but can become pathogenic, causing local skin infections or more serious systemic illness, accompanied by bacteremia, abscess formation in tissues, endocarditis, osteomyelitis, and morbidity and mortality (262). The bacterium encodes a number of virulence factors that potently stimulate inflammatory responses (263, 264). S. aureus infection triggers PRRs and subsequent production of inflammatory cytokines and chemokines, leading to the recruitment of neutrophils and abscess formation in tissues (265). Specifically, many components of the bacterial wall and pore-forming toxins are capable of providing signal 1 via TLR2 activation and signal 2, driving inflammasome activation (263, 265, 266). Multiple inflammasomes are known to be triggered by S. aureus, mainly NLRP3 (266-270) and caspase-1 activation is critical, as caspase-1 deficient macrophages are impaired in cytokine production (271).

To test the caspase-1 biosensors in vivo, we first used a model of S. aureus skin infection.

Biosensor mice were shaved to remove hair and intradermally infected with WT S. aureus or PBS control in the presence of dextran-based microcarrier beads (cytodex-3), which cause agitation to exacerbate S. aureus infection in the skin. Immediately after infection, 24h, 48, and 72h post-infection, mice were injected with 150mg/kg VivoGlo Luciferin, a luciferase substrate designed for measuring luciferase in live animal studies, and subjected to IVIS imaging to quantify biosensor activation at the site of infection.

We found a significant increase in bioluminescence in infected and control animals that peaked 24h after infection (Fig. 20). One of the PBS control injected animals (mouse A) had a large, red bump at the injection site, and significant biosensor activation (Fig. 20), suggesting that the dextran-based 67

microcarrier beads used for agitation also caused inflammation and biosensor activation in these mice.

Even though we were unable to measure biosensor signal specifically resulting from S. aureus skin

infection, we were able to measure the onset and resolution of biosensor signal in the skin with this

model.

Figure 20. Subcutaneous S. aureus Infection and Cytodex-3 Beads Cause Biosensor Activation in the Skin. Representative images of caspase-1 biosensor mice inoculated with S. aureus or PBS with cytodex-3 microbeads, immediately after and 24h after infection (left) and image taken of mouse A 24h after inoculation with PBS with cytodex beads (right).

We next chose to monitor S. aureus infection in the bloodstream, which induces systemic

inflammation in mice. Caspase-1 biosensor mice were infected with S. aureus (or PBS control) via retro-

orbital injection and luminescence was quantified 0, 1, 2 and 3 days post-infection using an IVIS imaging system (Fig. 21A). For imaging, mice were injected i.p. with 150mg/kg VivoGlo Luciferin, a luciferase substrate designed for measuring luciferase in live animal studies. Biosensor signal measured in animals 68

on day 0 was set as background, and we quantified the change in this biosensor signal over the course of

disease in PBS and infected animals. Bioluminescence accumulated in infected animals but not PBS

controls over the course of the experiment (Fig. 21A). On day 3, infected and control mice were

sacrificed following in vivo imaging, and the bioluminescence of individual organs was asses ex vivo. We

detected increased luciferase signal in the livers, kidneys and hearts isolated from S. aureus-infected mice relative to the PBS controls (Fig. 21B). 69

Figure 21. Caspase-1 Biosensors are Activated in S. aureus–Infected Mice. Representative IVIS images of caspase-1 biosensor mice infected with S. aureus or PBS day 0, 1, 2, and 3 post-infection (A). Regions of interest (ROI) (red gates) were drawn, and max radiance (p/s/cm2/sr) was quantified within each ROI. Data are expressed as fold max radiance relative to the max radiance detected in individual animals on day 0 (A). IVIS images of control and infected tissues day 3 post-infection (B).

Biosensor signal was quantified in tissues and bacterial CFU and cytokine expression were also measured to determine if bacterial burden and/or inflammatory markers coincided with areas of bioluminescence. We detected bacterial colonization in all kidneys and livers and some hearts isolated from infected mice, consistent with the increased biosensor activation detected in these tissues (Fig. 22A-

B). We also found increased expression of a panel of inflammatory cytokines and chemokines in tissues isolated from infected mice, relative to uninfected controls (Fig. 22C-D). These data demonstrate that caspase-1 biosensors are activated in vivo in response to S. aureus infection, and that we can monitor changes in biosensor activation over the course of disease in living animals and in infected tissues ex vivo.

Figure 22. Biosensor Signal, Bacterial Burden and Inflammatory Cytokines and Chemokines Detected in Tissues Isolated from S. aureus-Infected Mice. Biosensor activation measured ex vivo in the kidneys and livers of control and infected mice (A). Tissues were homogenized in PBS and homogenates were plated on TSA plates to enumerate bacterial CFU (B). Homogenates were centrifuged and supernatant was collected for CBA analysis (C and D). ND = not detectable.

Lastly, we wanted to determine if there was a correlation between biosensor signal, bacterial burden and IL-1β in tissues. The tissue specific role of inflammation and bacterial persistence or clearance is not well defined. Inflammation seems to be important in the skin, driving bacterial clearance

(265, 266). In systemic infections, inflammation may be more harmful for the host causing sepsis, though there is conflicting evidence as to whether inflammation drives bacterial clearance or persistence in 72

different tissues (263, 266, 272). We detected the strongest inflammatory response and bacterial colonization in the kidneys, and we measured a positive correlation between CFU and IL-1β as well as

CFU and the biosensor signal (Fig. 23). There was also a trending correlation between the biosensor signal and the presence of various inflammatory cytokines and chemokines including IL-1, TNF, KC and

MIP1 (Fig. 24). However this was not the case in the liver and heart where there was little to no correlation between CFU, biosensor signal, and inflammatory markers (Fig 25). Profiling inflammatory responses ex vivo come with the caveat that these are static measurements of a dynamic process. Caspase-

1 may be activated earlier during disease, prior to the appearance of inflammatory cytokines/chemokines in tissues. A greater sample size may be required in order to determine the tissue-specific role of inflammation on bacterial survival or clearance.

Figure 23. Biosensor Signal and IL-1β Correlate with Bacterial CFU in the Kidneys. Correlation graphs of biosensor signal (p/s/cm2/sr) or IL-1β plotted against bacterial CFU in the kidneys from control (3) and S. aureus- infected (8) mice. 73

Figure 24. Biosensor Signal Correlates with Inflammatory Markers in the Kidneys. Correlation graphs of biosensor signal (p/s/cm2/sr) plotted against the indicated cytokine or chemokine in 3 control and 8 infected kidneys.

Figure 25. Biosensor Signal and IL-1β do not Correlate with Bacterial CFU in the Liver and Heart. Correlation graphs of biosensor signal (p/s/cm2/sr) or IL-1β plotted against bacterial CFU in the liver (top) or heart (bottom) from control (3) and S. aureus-infected (8) mice. 74

DSS-Induced Colitis Model

To measure biosensor activation during intestinal inflammation, we used an established model of

dextran sodium sulfate (DSS)-induced ulcerative colitis (UC) (261). We chose this model because, in the

absence of caspase-1, DSS-induced colitis is ameliorated, indicating caspase-1 activation is essential for

intestinal inflammation (273-275). Mice were treated with 2% DSS ad libitum in the drinking water for 5 or 7 days to induce acute colitis. On day 5 or day 7, following administration of the luciferase substrate, biosensor signal was measured in vivo and then mice were sacrificed and colons extracted for ex vivo analyses. Unfortunately, we were unable to detect bioluminescence in the colons of most mice with mild to moderate colitis in vivo (Fig. 26A), likely because the colon is embedded behind the small intestines in

these animals and monitoring small changes in biosensor signal in the abdomen of these mice is difficult

due to the strong background signal in that area. However, in some mice with more severe colitis, we

were able to detect increased bioluminescence in the abdomen (Fig. 26B).

Figure 26. Biosensor Signal in the Colons in vivo. Biosensor activation (p/s/cm2/sr) in 1 control and 3 DSS-treated mice, 5 days after exposure to 2% DSS (A). Mice were injected with luciferase substrate via the tail vein and biosensor activation was measured immediately (A). Biosensor activation (p/s/cm2/sr) in 1 control and 2 DSS- treated mice, 7 days after exposure to 2% DSS (B). Mice were injected i.p with luciferase substrate and biosensor activation was measured after 10 minutes (B). Bodyweight loss (%) is shown for each animal. Ex vivo, the colon was measured, as colonic shortening correlates with acute inflammation severity. We also monitored for blood in the stool using a Hemoccult test. In this model, mice treated with 75

2% DSS for 5 days exhibit mild symptoms of colitis – little to no weight loss, slight colonic shortening,

and blood in the stool of some but not all mice (Fig. 27A-C). By day 7, we detect increased weight loss in the DSS-treated mice, and most mice exhibit colonic shortening and positive Hemoccult tests (Fig. 27C, data not shown).

Figure 27. Mice Exhibit Some Clinical Measurements of Colitis on Day 5 and Day 7. Representative Hemoccult images and results (+ = positive test, - = negative test) in control and DSS mice (A). Representative experiment with weight loss, colon length and Hemoccult test results in in control and DSS mice 5 days after DSS exposure (B). Weight loss in mice on day 5 versus day 7 in control and DSS mice (C).

We detected biosensor activation in colons isolated from DSS-treated mice on day 5 in males and day 7 in males and females (Fig. 28-29). Biosensor activation appeared segmented throughout the colon in most mice or localized to the distal region. Areas where we detected biosensor activation and control regions that did not exhibit bioluminescence were sectioned and stained with H&E for histopathological assessment or were isolated for western blot analysis of active caspase-1. In tissue sections derived from

areas of biosensor activation, we found nodules with inflammatory infiltrate in the mucosa and

submucosa regions of the colon, as well as a loss of crypt integrity, reduction of goblet cells, disruption of

the epithelial barrier, neutrophil infiltration, edema and thickening of the muscularis mucosae in H&E

sections (Fig. 28-29). None of these inflammatory hallmarks were found in control sections from the same mice, although we did find crypt elongation in some proximal sections where no bioluminescence was detected (Fig. 28-29). Notably, we detected activated caspase-1 specifically in areas of biosensor activation (Fig. 30A-B). We also found increased expression of inflammatory cytokines/chemokines, 76

including IL-6, KC, IL-1β and TNF, in the colons isolated from DSS-treated mice relative to the untreated controls (Fig. 30C). Together, these data demonstrate that the IQAD biosensor is activated in inflamed regions of the colon in mice with colitis.

Figure 28. Caspase-1 Biosensors are Activated in Inflamed Areas of the Colon During Colitis --Males. Representative images of biosensor signal in the colons of male mice treated with DSS for 5 d (DSS 01, DSS 02) and 7 d (DSS 54). Representative images of tissue sections acquired and histopathology assessed in the colons of mice with regions indicated with a + or – for biosensor positive or negative, respectively.

Figure 29. Caspase-1 Biosensors are Activated in Inflamed Areas of the Colon During Colitis --Females. Representative images of biosensor signal in the colons of female mice treated with DSS for 7 d (DSS 08, 88, 14, 64). Representative images of tissue sections acquired and histopathology assessed in the colons of mice with regions indicated with a “P” (proximal) or “D” (distal). Black arrows point to histopathological features of colitis including loss of crypts (08D, 88D), epithelial damage (08D), and inflammatory infiltration (88D, 14D), as well as crypt elongation (14P).

Figure 30 - Biosensors are Activated in Regions of Caspase-1 Activation and Inflammatory Cytokine/Chemokine Production in the Colon. Tissue sections were isolated from regions of the colon where biosensor activation was detected (+) or control regions where no biosensor activation was detected (-) following DSS exposure (A). Tissues were homogenized and lysed for Western blot analysis of cleaved caspase-1 or β-actin (B). Heatmaps depict log2 fold change in cytokine/chemokine expression in the colons of DSS mice relative to the average of two control mice (C).

In addition to monitoring biosensor activation in the colons of mice with colitis, we wanted to

determine if local inflammation in the colon affected inflammatory responses in other tissues. Day 5 and day 7 following DSS exposure, we detected increased bioluminescence in the brain, relative to untreated control animals (Fig. 31A-B). Biosensor activation persisted in DSS brain slices after one week in vitro, suggesting that the increased bioluminescence detected in the DSS brains is not due to an increase in luciferin distribution into the CNS resulting from increased blood-brain-barrier permeability (Fig. 31C).

We hypothesized that the inflammatory response in the brain was the result of a systemic inflammatory

response, however we did not detect inflammatory cytokines/chemokines in the serum until day 7, after

we first detect biosensor activation (Fig. 31D). These data suggest that inflammasome activation occurs in the brain during colitis, but the mechanism driving this process remains unknown.

Figure 31. Caspase-1 Biosensors are Activated in the Brain during Colitis. Representative images and quantification of biosensor activation in the brains on day 5 and day 7 following DSS exposure relative to control brains from mice given normal drinking water (A-B). Serum was collected from mice prior to sacrificing and imaging and harvested for CBA analysis (C). Control and DSS brains were sliced into 200μm sections using a microtome and cultured for one week on membrane inserts to derive organotypic brain slice cultures. Luciferase substrate was added to cultures and the brain slices were imaged after 1 week in vitro (D).

Graft-Versus-Host Disease Model

Recently, caspase-1 activation has been reported to occur in T cells during immune responses, including those occurring in models of autoimmune disease and inflammasome activation in T cells has been demonstrated to contribute to the production of inflammatory cytokines (13-15, 276, 277). Graft- versus-host disease (GVHD) is a reaction that occurs following the transfer of donor immune cells (the graft) into an allogeneic host (278). Donor T cells recognize host alloantigens as foreign, resulting in the 80

activation of donor T cells and migration into host tissues including the skin, gastrointestinal tract and

liver (278-280). We hypothesized that the biosensor would become activated in alloreactive donor T cells,

and we could quantify bioluminescence in these cells and track their movement in the host over the

course of disease. Furthermore, because the biosensor is only expressed in the donor cell population, we

could specifically study inflammasome activation in donor cells during GVHD.

To assess the IQAD biosensor in the context of GVHD, bone marrow cells and splenocytes were isolated from biosensor mice and transferred into irradiated WT C57Bl/6 mice (syngeneic transfer) or

BALB/c mice (allogeneic transfer). As a control, bone marrow cells only were transferred into irradiated allogeneic recipients, as this is necessary to reconstitute the host hematopoietic system but does not cause

GVHD in mice (281-283). At various time points post transfer, mice were imaged in vivo and bioluminescence was quantified. As expected, we found no biosensor activation in the syngeneic C57Bl/6 mice or in the allogeneic BALB/c mice that received bone marrow cells only (Fig. 32A). However, robust biosensor signal was detected in vivo in BALB/c mice that received allogeneic bone marrow cells and splenocytes (Fig. 32A). Biosensor signal first appeared in an area near the site of the spleen and over time, the signal spread to the abdomen and ultimately to lymph nodes. On day 7, prior to euthanasia, mice

were imaged in vivo, and then sacrificed and biosensor activation was quantified in tissues. We detected

strong bioluminescence in the spleen and moderate signal in the intestines and mesenteric lymph nodes

(Fig. 32B-C). Most of the luciferase signal in the intestinal tissues came from the mesenteric lymph

nodes, but we also detected alloreactive cells throughout the small intestines (Fig. 32D). 81

Figure 32. Biosensors are Activated in Alloreactive Donor Cells during GVHD. Bone marrow cells (107) ± 5 and 106 splenocytes isolated from biosensor mice were transferred into irradiated C57BL/6 syngeneic or BALB/c allogeneic recipient mice via i.v. injection into the tail vein. Representative IVIS images of mice at the indicated time points following syngeneic or allogeneic transfer of biosensor cells (A). Max radiance was quantified within the ROI for each mouse and graphed as fold max radiance relative to the max radiance detected on day 2, after transfer of biosensor cells, prior to disease development (A). On day 7, tissues were harvested and biosensor activation was measured ex vivo in the spleen, MLN, and intestines (B and C). Representative images from two GVHD mice, where the intestines and MLN are spread to better assess where the biosensor signal is coming from within the small intestine (D).

By day 7, these mice exhibited increased clinical scores, as measured by decreased activity, hunched body posture and lack of grooming and had lost ~20-25% of their body weight, indicative of acute graft-versus-host disease (Fig. 33A-B).

Figure 33. Mice Exhibit Clinical Measurements of GVHD. Clinical scores (A) and weight loss (B) were measured on day 7 in control and GVHD mice.

We also detected increased levels of cytokines and chemokines in the serum of GVHD mice,

including TNF and IL-6 (Fig 34). Although we are unable to determine if these cytokines/chemokines are

donor or host derived, their presence in the serum is consistent with GVHD manifestation.

Figure 34. Increased Inflammatory Cytokines in the Serum of GVHD mice. Serum was harvested from control and GVHD mice on day 7 and CBA analysis was performed to quantify the presence of the indicated cytokines and chemokines.

To confirm the biosensor was activated in donor T cells, we performed the same experiment transferring BM cells and T cells isolated from caspase-1 biosensor mice. T cells were isolated by negative selection to make sure this population of cells remained untouched. Donor cells were transferred into syngeneic C57Bl/6 CD45.1 recipient or allogeneic BALB/c mice, and mice were sacrificed 7 days

post-cell transfer and biosensor activation was measured in vivo and ex vivo. As before, we detected 83

biosensor activation in allogeneic recipient mice, specifically in the spleen, mesenteric lymph node and

small intestines (Fig. 35A-C). We next isolated splenocytes and measured luciferase signal in vitro. We

detected increased luciferase signal per cell in splenocytes isolated from allogeneic recipient mice

compared to splenocytes isolated from the syngeneic control (Fig. 35D). Flow cytometric analysis of these splenocytes revealed increased CD8 to CD4 T cell ratio, a hallmark of GVHD (Fig. 35E). We also detected increased CD8 T cell activation in donor cells isolated from mice with GVHD (Fig. 35F). Since all mice received the same donor cell population, collectively these data suggest that the IQAD biosensor is specifically activated in alloreactive T cells in mice with GVHD.

Figure 35. IQAD Biosensor is Activated in Donor T cells in Mice with GVHD. T cells were isolated from splenocytes from IQAD mice and purified by negative selection (>90% purity) and injected with BM cells into irradiated syngeneic CD45.1 C57Bl/6 or allogeneic BALB/c recipient mice and biosensor signal was measured in vivo 7 days later (A). Figure depicts one female (F0) and one male (M0) alongside a control (A). Tissues were harvested and bioluminescence was measured ex vivo in the spleens and intestines / mesenteric lymph nodes (B). Biosensor signal was quantified in the spleens (C). Spleens were collected from control and GVHD mice and splenocytes were isolated, lysed and biosensor signal was measured in 5 million cells (D). For flow cytometry analysis, splenocytes were stained with either anti-CD45.1 or anti-H-2Kb to gate on donor cells in the syngeneic or allogeneic recipients, respectively. Within donor populations, CD8+ and CD4+ populations were measured and CD8-to-CD4 ratios were measured (E). Within the CD8+ and CD4+ , CD44+ and CD25+ cells were quantified (F).

Our in vitro data demonstrated that the IQAD biosensor could be turned on in response to both

inflammatory and apoptotic stimuli. Our data above suggest that the IQAD biosensor is activated in

alloreactive T cells, but another possible explanation could be that the biosensor is being turned on during

T cell-activation induced cell death. To discriminate between these possibilities, we isolated splenocytes

from mice with GVHD, gated on the donor cell population and then sorted the Annexin V positive and

Annexin V negative cells. An equal number of sorted cells were lysed and luciferase activity was

quantified. We did detect some biosensor activation in the Annexin V positive population of donor T 85

cells, which could be due to cell death during the isolation and sorting processes or this could indicate that

the biosensor is active in dead/dying T cells in vivo (Fig. 36). Importantly, we found increased biosensor activation in the Annexin V negative population (Fig. 36), indicating that the IQAD biosensor is active in living alloreactive donor T cells.

Figure 36. IQAD Biosensor is Activated in Alloreactive Donor T cells during GVHD. Spleens were harvested from GVHD mice, splenocytes were isolated and stained for host cells (H-2Kd) followed by Annexin V-FITC staining just prior to FACS. Cells were gated on donor cells (H-2kd-) and then sorted based on Annexin V staining. Cells were collected and the same number of cells were spun down and lysed for luciferase measurements.

To determine if caspase-1 is active in alloreactive T cells, spleens were harvested from GVHD or

control C57Bl/6 mice and the T cells were sorted for immunoblot. We used an antibody for caspase-1 that

recognizes the p20 subunit and found increased expression of p33 (Fig. 37), the active species of caspase-

1 in cells (Fig. 7). These data indicate that caspase-1 is active in donor T cells during GVHD.

Figure 37. Caspase-1 Activation in Donor T cells in GVHD Spleens. Spleens were harvested from control and GVHD mice, T cells were isolated and lysed for WB.

EPS Treatment Reduces T cell Activation in GVHD Mice

Exopolysaccharide (EPS) produced by the commensal bacterium, Bacillus subtilis, can protect mice from T cell-mediated inflammatory disease induced by the enteric pathogen Citrobacter rodentium

and sepsis caused by infection of mice with the human pathogen Staphylococcus aureus (258, 284-286).

EPS functions by inducing anti-inflammatory M2 macrophages in a TLR4-dependent manner, which

ultimately inhibit T cell activation and proliferation (284). We hypothesized that EPS could protect mice

from acute GVHD, and we could measure differences in biosensor activation between the protected EPS-

treated mice and mice with GVHD. Donor and recipient mice were injected with EPS or PBS prior to the

induction of GVHD, and biosensor activation was measured over the course of disease. As before, we

detected bioluminescence in the PBS-treated mice by day 5-7 post cell transfer (Fig. 38). However, no

biosensor activation was detected in the EPS-treated mice on day 7, and two of the three mice continued to have no detectable luciferase signal on day 14 (Fig. 38). 87

Figure 38. Biosensor Activation is Attenuated in EPS-Treated Mice in vivo. On days -7, -5 and -3, biosensor donor mice and BALB/c recipient mice were injected with 300mg/kg of EPS in PBS or PBS only control. On Day 0, 107 bone marrow cells + 5x106 splenocytes from biosensor mice were transferred into irradiated BALB/c mice via intravenous injection. As a control, bone marrow cells (107) only were transferred. Biosensor activation was monitored on the IVIS days 2, 5, 7 and 14 post-GVHD induction. Representative IVIS images of PBS- or EPS- treated mice at the indicated time points following allogeneic transfer of biosensor cells (A). Max radiance was quantified within the ROI (red box) for each mouse and graphed as fold max radiance relative to the max radiance (B).

Mice were sacrificed on day 7 after induction of GVHD, a time point where nearly all PBS mice

reached criteria, and bioluminescence was quantified in the spleen and lymph nodes of PBS and EPS-

treated animals. We measured a marked reduction in biosensor activation in the spleens and intestines

(that contained mesenteric lymph nodes) of EPS-treated mice relative to the PBS-treated group (Fig. 39A- 88

C). We also assessed body weight loss during disease and found that by day 7, whereas 12 of 13 (92%)

PBS-treated mice were moribund (i.e., lost more than 20% of body weight), only 6 of 15 (40%) EPS-

treated mice were moribund, indicating that 60% of the EPS-treated mice were protected (Fig. 39D).

These data suggest that EPS can improve clinical manifestations of acute GVHD and this can be monitored by biosensor activation in live animals and tissues ex vivo.

Figure 39. EPS-Treated Mice Exhibit Reduced Biosensor Signal in Tissues and Reduced Clinical Disease. PBS and EPS-tread mice were sacrificed on day 7, the spleen and small intestines with mesenteric lymph nodes were isolated and bioluminescence was quantified ex vivo (A-C). Weight loss was monitored over the course of disease and % weight loss on day 7 relative to initial weight on day 0 was determined (D).

CHAPTER SIX

SUMMARY AND DISCUSSION

In this study, we describe a bioluminescent sensor that quantitatively reports caspase-1 activation in cells and in living animals. We focused on caspase-1 since a diverse repertoire of PAMPs and DAMPS drive activation of a number of different inflammasomes, all of which culminate in the activation of caspase-1. The biosensor design was based on previously published circularly permuted luciferases that contain a protease recognition sequence between the two domains of the protein (230-233). We hypothesized that if a caspase-1 target sequence was placed into the luciferase construct, then we would detect luciferase signal whenever caspase-1 was active in a cell. We initially screened a number of caspase-1 target sequences and identified the strongest luciferase signal in cells expressing biosensors containing the IQAD or FLTD target sequences from procaspase-7 and gasdermin-D. Procaspase-7 is a well-established downstream target of caspase-1 (209, 238-240) and our data support this, as IQAD was cleaved in response to a variety of inflammatory stimuli in vitro and in vivo. Gasdermin-D is a well- established target of caspase-1 (193, 209) and we detected similar luciferase activity with our GSDMD biosensor as with IQAD.

While we call these ‘caspase-1 biosensors’ and can demonstrate that the biosensors are activated in response to a broad panel of inflammatory stimuli in vitro and inflammatory conditions in vivo where we detect active caspase-1, it is important to note that multiple caspases are capable of activating these biosensors. To date, no caspase-1 specific target sequence has been identified, making it impossible to design a biosensor that is specifically cleaved by caspase-1. The most well-established targets of caspase-

1 are proIL-1β, pro-IL-18, procaspase-7 and GSDMD, all of which are cleaved by other caspases and enzymes at the same amino acid sequence that is cleaved by caspase-1 (192, 214-218, 238, 287-290).

Furthermore, there is such significant cross-talk between inflammatory and apoptotic pathways that it is

virtually impossible to separate these two pathways, particularly in the context of an inflammatory

response in vivo.

The concept of ‘inflammatory’ and ‘apoptotic’ caspases is outdated, as it is now well established

that caspases integrate cell death and inflammatory pathways via a common axis (291, 292). Even if we

did generate a biosensor that was somehow specifically cleaved by caspase-1, in addition to being

activated during inflammation, this biosensor would also be turned on during apoptosis, since caspase-1 is

activated in this cell death pathway (290, 293, 294). Despite the lack of specificity of our biosensors, we

found that in vivo, the IQAD biosensor reliably directed us to locations of caspase-1 activation in tissues,

while we did not detect biosensor signal in locations of apoptosis in vivo or ex vivo (i.e., the thymus). We

suspect that the lack of biosensor activation during apoptosis is due to the fact that apoptotic cells are

rapidly cleared by phagocytes in vivo (253, 254). Recently, caspase-8 has emerged as a driver of

inflammation and numerous groups have demonstrated that functional inflammasomes can form with

caspase-8 (125, 219-222). This could explain why the IQAD biosensor has performed so well in vitro and

in vivo, since procaspase-7 is also cleaved by active caspase-8 (221, 238). So, while we were initially

concerned that our putative biosensors could be turned on by other caspases, the fact that these biosensors

are not specific for caspase-1 may make them more valuable for monitoring inflammatory responses.

We did not detect biosensor activation in response to all stimuli tested. Some treatments tested

may need to be optimized further, testing different concentrations, time points, drugs from different

companies. However, we were able to detect increased markers of inflammation in response to some of

these treatments, indicating that inflammatory cascades were activated. The finding that nigericin

stimulates biosensor activation in differentiated THP-1s and MDMs but not mouse BMDMs was

surprising, especially considering the high levels of caspase-1 activation detected by western blot and IL-

1β secretion from these cells. A possible explanation for this is a difference in caspase-1 target kinetics in

BMDMs compared to THP-1 cells. We did detect more rapid biosensor activation in BMDMs, where

bioluminescence peaked between 30 minutes and 1 hour, compared to THP-1 cells were bioluminescence accumulated over the course of 3 hours. Previous groups have proposed a 15 minute window for caspase- 91

1 activation in nigericin-stimulated BMDMs, where peak caspase-1 activation occurs between 15 and 30

minutes (49). The earliest time point for our luciferase measurements was 30 minutes. However, firefly

luciferase is a relatively stable protein with a half-life of ~2h (295) and, once activated, should retain

signal in living cells. Nonetheless, these data bring up an important caveat to our IQAD biosensor –it may

not be activated every time caspase-1 is activated in cells.

Surprisingly, we did not detect robust biosensor activation in cells transfected with LPS, relative

to poly(I:C) and poly(dA:dT). Intracellular LPS activates the noncanonical inflammasome, leading to

active caspase-1, we therefore suspected to detect strong bioluminescence. Instead, we detect ~2-6 fold

increase in biosensor activation with all biosensors in human and mouse cells. We also detect some IL-1β

production by ELISA, indicating that the inflammsome is active (data not shown). As mentioned earlier,

we used serotype 055:B5 derived from E. coli, and serotype-dependent differences in noncanonical inflammasome activation have previously been reported (9). It will be interesting to perform further experiments using different LPS serotypes in cells expressing the IQAD or GSDMD biosensors. Since

GSDMD is cleaved by caspases-4/5/11 as well as caspase-1, activation of this biosensor may be more robust compared to IQAD.

Our ultimate goal was to use caspase-1 biosensors to monitor inflammatory responses in vivo. To

this end, we chose a S. aureus infection model of systemic inflammation and colitis model of tissue-

specific inflammation to validate our biosensor in living animals and tissues ex vivo. We detected

bioluminescence that accumulated over time in mice infected with S. aureus and in infected organs ex vivo. We also detected biosensor signal in the colons and brains of mice with colitis. One caveat to using transgenic mice expressing IQAD biosensors is the level of background bioluminescence. These mice can exhibit up to ~106 p/sec/cm2/sr biosensor signal under normal conditions, with the highest signal

occurring at the site of injection (tail or peritoneal cavity). In our experiments, we overcame this by

quantifying the fold change in biosensor activation relative to the signal measured on day 0. In our S.

aureus infection model where we detected robust biosensor signal in infected animals, we were able to

measure significant signal over background. However, the background signal can be problematic when 92

measuring small changes in bioluminescence in the abdomen area, as was the case with our colitis model.

Additionally, detecting biosensor signal in vivo from certain tissues, like the kidneys, was difficult,

because these organs are physically obstructed by other organs in the mouse. We have recently developed

a system to surgically implant windows in various areas of the mouse (abdominal or cranial) which will

help us monitor low levels of bioluminescence in live animals in tissues that are hard to measure, such as

the brain.

The high background biosensor activity was primarily detected in the abdomen. Some of this

signal could be due to the food in the intestines. Alfalfa (chlorophyll) in mouse chow is very auto-

luminescent (296), and in fact, if we place normal mouse food in the IVIS machine, we detect significant

signal from these food pellets (data not shown). In some experiments, such as the colitis model, we used

Alfalfa-free chow to reduce this background signal in the colon. Even in mice fed this diet, the biosensor is still activated in the abdomen in vivo and intestine ex vivo. There may be some biological significance to low basal levels of caspase-1 activation in some tissues like the intestines, where cells are constantly being exposed to microbes. Low-grade inflammation may be important for maintaining immune homeostasis.

Despite the background signal in vivo, the ability to detect distinct areas of caspase-1 activation within tissues proved to be advantageous in our colitis model. Biosensor signal reliably identified colonic regions of inflammation in a model of DSS. Notably, we detected active caspase-1 in tissues or regions of tissues that exhibited strong bioluminescence. These data highlight the potential of caspase-1 biosensor mice as guides to identifying locations in animals or within tissues where inflammatory responses are occurring.

In our colitis model, biosensor signal was detected in distal sites from the diseased organ. We consistently measured higher signal in the brains isolated from mice exposed to DSS for 5-7 days. While there have been previous reports that colitis drives neuroinflammation (297), the mechanism for how this is happening remains elusive. DSS causes colitis by breaking down the epithelial barrier, which allows bacteria that normally reside in the lumen to activate immune cells in the lamina propria, driving an 93

inflammatory response. One hypothesis is that gut microbiota, a bacterial product or bacterial metabolite

is driving inflammation in the brain. Alternatively, upon sensing inflammation in the colon, the vagus

nerve may transmit this information to the brain, activating an inflammatory response in the CNS. A

better understanding of how the brain senses peripheral dysbiosis and how tissue-specific inflammatory

responses can drive changes in the brain is critical in numerous fields.

We next wanted to evaluate the utility of our biosensor during graft-versus-host disease. We

speculated that due to the aforementioned high background in the biosensor mice, it would be very

difficult to detect subtle changes in biosensor activation in GVHD animals (the host). However, if we transferred hematopoietic cells expressing the biosensor into irradiated WT mice, then we would be able to specifically monitor biosensor activation in immune cell populations (the donor). Our data demonstrate that the IQAD biosensor was activated in alloreactive T cells in specific tissues, primarily the spleen, in irradiated allogeneic recipient mice. Donor cells were specifically activated in the context of GVHD, since little to no biosensor was detected in the allogeneic mice that received BM cells only or in the syngeneic mice that received the same population of BM cells and splenocytes. Biosensor signal was detected in Annexin V negative cells ex vivo. We also detected active caspase-1 (p33 species) in donor T cells isolated from GVHD mice, but not in T cells isolated from WT mice, indicating that caspase-1 is active in alloreactive, living T cells.

Accumulating evidence has supported a role for NLRP3 inflammasome and caspase-1 activation and IL-1β production in T cells (13, 15) and, more generally, in autoimmune diseases (276, 277). Other groups have reported a T cell-intrinsic inflammasome that drives the production of IL-1β via NLRP3 and caspase-8 (14), and it is possible that caspase-8 is driving IQAD sensor activation in this model.

Regardless of the caspase responsible for biosensor cleavage, the ability to monitor alloreactive T cell populations in vivo highlights the potential use of caspase-1 biosensors for monitoring T cell responses in

the context of various autoimmune diseases.

To determine if we could use IQAD biosensors to assess the severity of disease, we monitored

biosensor activation during the course of acute GVHD in control mice and mice pre-treated with EPS. We 94

showed previously that EPS protects mice from several T-cell mediated diseases (258, 284, 286) and we hypothesized that EPS might also protect mice from acute GVHD. We found that pre-treatment with EPS prior to the induction of GVHD significantly improved clinical outcomes of the disease, as measured by weight loss and mortality. Importantly, we detected reduced biosensor activation in the EPS-treated mice compared to the control group. Because our data demonstrate that biosensor activation was due to the activation of alloreactive donor T cells, we can conclude that EPS affects (directly or indirectly) the activation of alloreactive donor T cells. Furthermore, biosensor activation allowed the severity of disease to be readily monitored and revealed the positive influence of EPS treatment in this setting. These data suggest that caspase-1 biosensors may be valuable for testing novel treatments in live animals.

In conclusion, we developed a bioluminescent reporter that detects caspase-1 activation. This biosensor is activated in response to a variety of inflammatory stimuli and murine models of human disease. Inflammation is a dynamic process, and the ability to track and quantify the initiation and onset of inflammatory responses in individual animals over time has both experimental and economic value.

Ultimately, we hope that this biosensor, when used concurrently with other conventional assays that measure inflammasome/caspase-1 activation will help to identify novel sites of inflammation in animals and tissues, which may facilitate the development and evaluation of future therapies for inflammatory diseases.

REFERENCE LIST

1. Martinon F, Burns K, Tschopp J. 2002. The inflammasome: a molecular platform triggering activation of inflammatory caspases and processing of proIL-beta. Mol Cell 10:417-426. 2. Broz P, Dixit VM. 2016. Inflammasomes: mechanism of assembly, regulation and signalling. Nat Rev Immunol 16:407-420. 3. Lamkanfi M, Dixit VM. 2014. Mechanisms and functions of inflammasomes. Cell 157:1013- 1022. 4. Vanaja SK, Rathinam VA, Fitzgerald KA. 2015. Mechanisms of inflammasome activation: recent advances and novel insights. Trends Cell Biol 25:308-315. 5. Sharma D, Kanneganti TD. 2016. The cell biology of inflammasomes: Mechanisms of inflammasome activation and regulation. J Cell Biol 213:617-629. 6. Schroder K, Tschopp J. 2010. The inflammasomes. Cell 140:821-832. 7. Kayagaki N, Warming S, Lamkanfi M, Vande Walle L, Louie S, Dong J, Newton K, Qu Y, Liu J, Heldens S, Zhang J, Lee WP, Roose-Girma M, Dixit VM. 2011. Non-canonical inflammasome activation targets caspase-11. Nature 479:117-121. 8. Shi J, Zhao Y, Wang Y, Gao W, Ding J, Li P, Hu L, Shao F. 2014. Inflammatory caspases are innate immune receptors for intracellular LPS. Nature 514:187-192. 9. Kayagaki N, Wong MT, Stowe IB, Ramani SR, Gonzalez LC, Akashi-Takamura S, Miyake K, Zhang J, Lee WP, Muszynski A, Forsberg LS, Carlson RW, Dixit VM. 2013. Noncanonical inflammasome activation by intracellular LPS independent of TLR4. Science 341:1246-1249. 10. Hagar JA, Powell DA, Aachoui Y, Ernst RK, Miao EA. 2013. Cytoplasmic LPS activates caspase-11: implications in TLR4-independent endotoxic shock. Science 341:1250-1253. 11. Svensson A, Patzi Churqui M, Schluter K, Lind L, Eriksson K. 2017. Maturation-dependent expression of AIM2 in human B-cells. PLoS One 12:e0183268. 12. Singh VV, Kerur N, Bottero V, Dutta S, Chakraborty S, Ansari MA, Paudel N, Chikoti L, Chandran B. 2013. Kaposi's sarcoma-associated herpesvirus latency in endothelial and B cells activates gamma interferon-inducible protein 16-mediated inflammasomes. J Virol 87:4417-4431. 13. Arbore G, West EE, Spolski R, Robertson AAB, Klos A, Rheinheimer C, Dutow P, Woodruff TM, Yu ZX, O'Neill LA, Coll RC, Sher A, Leonard WJ, Kohl J, Monk P, Cooper MA, Arno M, Afzali B, Lachmann HJ, Cope AP, Mayer-Barber KD, Kemper C. 2016. T helper 1 immunity requires complement-driven NLRP3 inflammasome activity in CD4(+) T cells. Science 352:aad1210. 14. Bradley N Martin CW, Cun-jin Zhang, Zizhen Kang, Muhammet Fatih Gulen, Jarod A Zepp, Junjie Zhao, Guanglin Bian, Jeong-su Do, Booki Min, Paul G Pavicic Jr, Caroline El- Sanadi, Paul L Fox, Aoi Akitsu, Yoichiro Iwakura, Anasuya Sarkar, Mark D Wewers, William J Kaiser, Edward S Mocarski, Marc E Rothenberg, Amy G Hise, George R Dubyak, Richard M Ransohoff & Xiaoxia Li. 2018. T cell–intrinsic ASC critically promotes TH17-mediated experimental autoimmune encephalomyelitis. Nature Immunology 17:583-592. 15. Doitsh G, Galloway NL, Geng X, Yang Z, Monroe KM, Zepeda O, Hunt PW, Hatano H, Sowinski S, Munoz-Arias I, Greene WC. 2014. Cell death by pyroptosis drives CD4 T-cell depletion in HIV-1 infection. Nature 505:509-514.

16. Tan MS, Tan L, Jiang T, Zhu XC, Wang HF, Jia CD, Yu JT. 2014. Amyloid-beta induces NLRP1-dependent neuronal pyroptosis in models of Alzheimer's disease. Cell Death Dis 5:e1382. 17. Yogarajah T, Ong KC, Perera D, Wong KT. 2017. AIM2 Inflammasome-Mediated Pyroptosis in Enterovirus A71-Infected Neuronal Cells Restricts Viral Replication. Sci Rep 7:5845. 18. Wu PJ, Liu HY, Huang TN, Hsueh YP. 2016. AIM 2 inflammasomes regulate neuronal morphology and influence anxiety and memory in mice. Sci Rep 6:32405. 19. Kaushal V, Dye R, Pakavathkumar P, Foveau B, Flores J, Hyman B, Ghetti B, Koller BH, LeBlanc AC. 2015. Neuronal NLRP1 inflammasome activation of Caspase-1 coordinately regulates inflammatory interleukin-1-beta production and axonal degeneration-associated Caspase-6 activation. Cell Death Differ 22:1676-1686. 20. Freeman L, Guo H, David CN, Brickey WJ, Jha S, Ting JP. 2017. NLR members NLRC4 and NLRP3 mediate sterile inflammasome activation in microglia and astrocytes. J Exp Med 214:1351-1370. 21. Minkiewicz J, de Rivero Vaccari JP, Keane RW. 2013. Human astrocytes express a novel NLRP2 inflammasome. Glia 61:1113-1121. 22. Halle A, Hornung V, Petzold GC, Stewart CR, Monks BG, Reinheckel T, Fitzgerald KA, Latz E, Moore KJ, Golenbock DT. 2008. The NALP3 inflammasome is involved in the innate immune response to amyloid-beta. Nat Immunol 9:857-865. 23. Hanamsagar R, Torres V, Kielian T. 2011. Inflammasome activation and IL-1beta/IL-18 processing are influenced by distinct pathways in microglia. J Neurochem 119:736-748. 24. Ising C, Venegas C, Zhang S, Scheiblich H, Schmidt SV, Vieira-Saecker A, Schwartz S, Albasset S, McManus RM, Tejera D, Griep A, Santarelli F, Brosseron F, Opitz S, Stunden J, Merten M, Kayed R, Golenbock DT, Blum D, Latz E, Buee L, Heneka MT. 2019. NLRP3 inflammasome activation drives tau pathology. Nature doi:10.1038/s41586-019-1769-z. 25. Csak T, Ganz M, Pespisa J, Kodys K, Dolganiuc A, Szabo G. 2011. Fatty acid and endotoxin activate inflammasomes in mouse hepatocytes that release danger signals to stimulate immune cells. Hepatology 54:133-144. 26. Zhong FL, Mamai O, Sborgi L, Boussofara L, Hopkins R, Robinson K, Szeverenyi I, Takeichi T, Balaji R, Lau A, Tye H, Roy K, Bonnard C, Ahl PJ, Jones LA, Baker PJ, Lacina L, Otsuka A, Fournie PR, Malecaze F, Lane EB, Akiyama M, Kabashima K, Connolly JE, Masters SL, Soler VJ, Omar SS, McGrath JA, Nedelcu R, Gribaa M, Denguezli M, Saad A, Hiller S, Reversade B. 2016. Germline NLRP1 Mutations Cause Skin Inflammatory and Cancer Susceptibility Syndromes via Inflammasome Activation. Cell 167:187- 202 e117. 27. Dai X, Tohyama M, Murakami M, Sayama K. 2017. Epidermal keratinocytes sense dsRNA via the NLRP3 inflammasome, mediating interleukin (IL)-1beta and IL-18 release. Exp Dermatol 26:904-911. 28. Fenini G, Grossi S, Contassot E, Biedermann T, Reichmann E, French LE, Beer HD. 2018. Genome Editing of Human Primary Keratinocytes by CRISPR/Cas9 Reveals an Essential Role of the NLRP1 Inflammasome in UVB Sensing. J Invest Dermatol 138:2644-2652. 29. Dombrowski Y, Peric M, Koglin S, Kammerbauer C, Goss C, Anz D, Simanski M, Glaser R, Harder J, Hornung V, Gallo RL, Ruzicka T, Besch R, Schauber J. 2011. Cytosolic DNA triggers inflammasome activation in keratinocytes in psoriatic lesions. Sci Transl Med 3:82ra38. 30. Hiruma J, Harada K, Motoyama A, Okubo Y, Maeda T, Yamamoto M, Miyai M, Hibino T, Tsuboi R. 2018. Key component of inflammasome, NLRC4, was identified in the lesional epidermis of psoriatic patients. J Dermatol 45:971-977. 31. Sand J, Haertel E, Biedermann T, Contassot E, Reichmann E, French LE, Werner S, Beer HD. 2018. Expression of inflammasome proteins and inflammasome activation occurs in human, but not in murine keratinocytes. Cell Death Dis 9:24. 97

32. Song-Zhao GX, Srinivasan N, Pott J, Baban D, Frankel G, Maloy KJ. 2014. Nlrp3 activation in the intestinal epithelium protects against a mucosal pathogen. Mucosal Immunol 7:763-774. 33. Rauch I, Deets KA, Ji DX, von Moltke J, Tenthorey JL, Lee AY, Philip NH, Ayres JS, Brodsky IE, Gronert K, Vance RE. 2017. NAIP-NLRC4 Inflammasomes Coordinate Intestinal Epithelial Cell Expulsion with Eicosanoid and IL-18 Release via Activation of Caspase-1 and -8. Immunity 46:649-659. 34. Nordlander S, Pott J, Maloy KJ. 2014. NLRC4 expression in intestinal epithelial cells mediates protection against an enteric pathogen. Mucosal Immunol 7:775-785. 35. Ratsimandresy RA, Indramohan M, Dorfleutner A, Stehlik C. 2017. The AIM2 inflammasome is a central regulator of intestinal homeostasis through the IL-18/IL-22/STAT3 pathway. Cell Mol Immunol 14:127-142. 36. Sellin ME, Muller AA, Felmy B, Dolowschiak T, Diard M, Tardivel A, Maslowski KM, Hardt WD. 2014. Epithelium-intrinsic NAIP/NLRC4 inflammasome drives infected enterocyte expulsion to restrict Salmonella replication in the intestinal mucosa. Cell Host Microbe 16:237- 248. 37. Knodler LA, Crowley SM, Sham HP, Yang H, Wrande M, Ma C, Ernst RK, Steele- Mortimer O, Celli J, Vallance BA. 2014. Noncanonical inflammasome activation of caspase- 4/caspase-11 mediates epithelial defenses against enteric bacterial pathogens. Cell Host Microbe 16:249-256. 38. Hu B, Jin C, Li HB, Tong J, Ouyang X, Cetinbas NM, Zhu S, Strowig T, Lam FC, Zhao C, Henao-Mejia J, Yilmaz O, Fitzgerald KA, Eisenbarth SC, Elinav E, Flavell RA. 2016. The DNA-sensing AIM2 inflammasome controls radiation-induced cell death and tissue injury. Science 354:765-768. 39. Hausmann A, Bock D, Geiser P, Berthold DL, Fattinger SA, Furter M, Bouman JA, Barthel-Scherrer M, Lang CM, Bakkeren E, Kolinko I, Diard M, Bumann D, Slack E, Regoes RR, Pilhofer M, Sellin ME, Hardt WD. 2020. Intestinal epithelial NAIP/NLRC4 restricts systemic dissemination of the adapted pathogen Salmonella Typhimurium due to site- specific bacterial PAMP expression. Mucosal Immunol doi:10.1038/s41385-019-0247-0. 40. Duncan JA, Canna SW. 2018. The NLRC4 Inflammasome. Immunol Rev 281:115-123. 41. Moreira LO, Zamboni DS. 2012. NOD1 and NOD2 Signaling in Infection and Inflammation. Front Immunol 3:328. 42. Connolly DJ, Bowie AG. 2014. The emerging role of human PYHIN proteins in innate immunity: implications for health and disease. Biochem Pharmacol 92:405-414. 43. Heilig R, Broz P. 2018. Function and mechanism of the pyrin inflammasome. Eur J Immunol 48:230-238. 44. Sborgi L, Ravotti F, Dandey VP, Dick MS, Mazur A, Reckel S, Chami M, Scherer S, Huber M, Bockmann A, Egelman EH, Stahlberg H, Broz P, Meier BH, Hiller S. 2015. Structure and assembly of the mouse ASC inflammasome by combined NMR spectroscopy and cryo-electron microscopy. Proc Natl Acad Sci U S A 112:13237-13242. 45. Lu A, Magupalli VG, Ruan J, Yin Q, Atianand MK, Vos MR, Schroder GF, Fitzgerald KA, Wu H, Egelman EH. 2014. Unified polymerization mechanism for the assembly of ASC- dependent inflammasomes. Cell 156:1193-1206. 46. Franklin BS, Bossaller L, De Nardo D, Ratter JM, Stutz A, Engels G, Brenker C, Nordhoff M, Mirandola SR, Al-Amoudi A, Mangan MS, Zimmer S, Monks BG, Fricke M, Schmidt RE, Espevik T, Jones B, Jarnicki AG, Hansbro PM, Busto P, Marshak-Rothstein A, Hornemann S, Aguzzi A, Kastenmuller W, Latz E. 2014. The adaptor ASC has extracellular and 'prionoid' activities that propagate inflammation. Nat Immunol 15:727-737. 47. Baroja-Mazo A, Martin-Sanchez F, Gomez AI, Martinez CM, Amores-Iniesta J, Compan V, Barbera-Cremades M, Yague J, Ruiz-Ortiz E, Anton J, Bujan S, Couillin I, Brough D, Arostegui JI, Pelegrin P. 2014. The NLRP3 inflammasome is released as a particulate danger signal that amplifies the inflammatory response. Nat Immunol 15:738-748. 98

48. Balci-Peynircioglu B, Waite AL, Schaner P, Taskiran ZE, Richards N, Orhan D, Gucer S, Ozen S, Gumucio D, Yilmaz E. 2008. Expression of ASC in renal tissues of familial mediterranean fever patients with amyloidosis: postulating a role for ASC in AA type amyloid deposition. Exp Biol Med (Maywood) 233:1324-1333. 49. Boucher D, Monteleone M, Coll RC, Chen KW, Ross CM, Teo JL, Gomez GA, Holley CL, Bierschenk D, Stacey KJ, Yap AS, Bezbradica JS, Schroder K. 2018. Caspase-1 self-cleavage is an intrinsic mechanism to terminate inflammasome activity. J Exp Med 215:827-840. 50. Guo H, Callaway JB, Ting JP. 2015. Inflammasomes: mechanism of action, role in disease, and therapeutics. Nat Med 21:677-687. 51. Yu CH, Moecking J, Geyer M, Masters SL. 2018. Mechanisms of NLRP1-Mediated Autoinflammatory Disease in Humans and Mice. J Mol Biol 430:142-152. 52. Boyden ED, Dietrich WF. 2006. Nalp1b controls mouse macrophage susceptibility to anthrax lethal toxin. Nat Genet 38:240-244. 53. Fink SL, Bergsbaken T, Cookson BT. 2008. Anthrax lethal toxin and Salmonella elicit the common cell death pathway of caspase-1-dependent pyroptosis via distinct mechanisms. Proc Natl Acad Sci U S A 105:4312-4317. 54. Agrawal A, Pulendran B. 2004. Anthrax lethal toxin: a weapon of multisystem destruction. Cell Mol Life Sci 61:2859-2865. 55. Chavarria-Smith J, Mitchell PS, Ho AM, Daugherty MD, Vance RE. 2016. Functional and Evolutionary Analyses Identify Proteolysis as a General Mechanism for NLRP1 Inflammasome Activation. PLoS Pathog 12:e1006052. 56. Klimpel KR, Arora N, Leppla SH. 1994. Anthrax toxin lethal factor contains a zinc metalloprotease consensus sequence which is required for lethal toxin activity. Mol Microbiol 13:1093-1100. 57. Levinsohn JL, Newman ZL, Hellmich KA, Fattah R, Getz MA, Liu S, Sastalla I, Leppla SH, Moayeri M. 2012. Anthrax lethal factor cleavage of Nlrp1 is required for activation of the inflammasome. PLoS Pathog 8:e1002638. 58. Hellmich KA, Levinsohn JL, Fattah R, Newman ZL, Maier N, Sastalla I, Liu S, Leppla SH, Moayeri M. 2012. Anthrax lethal factor cleaves mouse nlrp1b in both toxin-sensitive and toxin- resistant macrophages. PLoS One 7:e49741. 59. Chavarria-Smith J, Vance RE. 2013. Direct proteolytic cleavage of NLRP1B is necessary and sufficient for inflammasome activation by anthrax lethal factor. PLoS Pathog 9:e1003452. 60. Frew BC, Joag VR, Mogridge J. 2012. Proteolytic processing of Nlrp1b is required for inflammasome activity. PLoS Pathog 8:e1002659. 61. Finger JN, Lich JD, Dare LC, Cook MN, Brown KK, Duraiswami C, Bertin J, Gough PJ. 2012. Autolytic proteolysis within the function to find domain (FIIND) is required for NLRP1 inflammasome activity. J Biol Chem 287:25030-25037. 62. D'Osualdo A, Weichenberger CX, Wagner RN, Godzik A, Wooley J, Reed JC. 2011. CARD8 and NLRP1 undergo autoproteolytic processing through a ZU5-like domain. PLoS One 6:e27396. 63. Okondo MC, Rao SD, Taabazuing CY, Chui AJ, Poplawski SE, Johnson DC, Bachovchin DA. 2018. Inhibition of Dpp8/9 Activates the Nlrp1b Inflammasome. Cell Chem Biol 25:262-267 e265. 64. Squires RC, Muehlbauer SM, Brojatsch J. 2007. Proteasomes control caspase-1 activation in anthrax lethal toxin-mediated cell killing. J Biol Chem 282:34260-34267. 65. Chui AJ, Okondo MC, Rao SD, Gai K, Griswold AR, Johnson DC, Ball DP, Taabazuing CY, Orth EL, Vittimberga BA, Bachovchin DA. 2019. N-terminal degradation activates the NLRP1B inflammasome. Science 364:82-85. 66. Sandstrom A, Mitchell PS, Goers L, Mu EW, Lesser CF, Vance RE. 2019. Functional degradation: A mechanism of NLRP1 inflammasome activation by diverse pathogen enzymes. Science 364. 99

67. Wickliffe KE, Leppla SH, Moayeri M. 2008. Killing of macrophages by anthrax lethal toxin: involvement of the N-end rule pathway. Cell Microbiol 10:1352-1362. 68. Mitchell PS, Sandstrom A, Vance RE. 2019. The NLRP1 inflammasome: new mechanistic insights and unresolved mysteries. Curr Opin Immunol 60:37-45. 69. Zhong FL, Robinson K, Teo DET, Tan KY, Lim C, Harapas CR, Yu CH, Xie WH, Sobota RM, Au VB, Hopkins R, D'Osualdo A, Reed JC, Connolly JE, Masters SL, Reversade B. 2018. Human DPP9 represses NLRP1 inflammasome and protects against autoinflammatory diseases via both peptidase activity and FIIND domain binding. J Biol Chem 293:18864-18878. 70. Neiman-Zenevich J, Stuart S, Abdel-Nour M, Girardin SE, Mogridge J. 2017. Listeria monocytogenes and Shigella flexneri Activate the NLRP1B Inflammasome. Infect Immun 85. 71. Ewald SE, Chavarria-Smith J, Boothroyd JC. 2014. NLRP1 is an inflammasome sensor for Toxoplasma gondii. Infect Immun 82:460-468. 72. Cirelli KM, Gorfu G, Hassan MA, Printz M, Crown D, Leppla SH, Grigg ME, Saeij JP, Moayeri M. 2014. Inflammasome sensor NLRP1 controls rat macrophage susceptibility to Toxoplasma gondii. PLoS Pathog 10:e1003927. 73. Grandemange S, Sanchez E, Louis-Plence P, Tran Mau-Them F, Bessis D, Coubes C, Frouin E, Seyger M, Girard M, Puechberty J, Costes V, Rodiere M, Carbasse A, Jeziorski E, Portales P, Sarrabay G, Mondain M, Jorgensen C, Apparailly F, Hoppenreijs E, Touitou I, Genevieve D. 2017. A new autoinflammatory and autoimmune syndrome associated with NLRP1 mutations: NAIAD (NLRP1-associated autoinflammation with arthritis and dyskeratosis). Ann Rheum Dis 76:1191-1198. 74. Kummer JA, Broekhuizen R, Everett H, Agostini L, Kuijk L, Martinon F, van Bruggen R, Tschopp J. 2007. Inflammasome components NALP 1 and 3 show distinct but separate expression profiles in human tissues suggesting a site-specific role in the inflammatory response. J Histochem Cytochem 55:443-452. 75. Zhou R, Yazdi AS, Menu P, Tschopp J. 2011. A role for mitochondria in NLRP3 inflammasome activation. Nature 469:221-225. 76. Dostert C, Petrilli V, Van Bruggen R, Steele C, Mossman BT, Tschopp J. 2008. Innate immune activation through Nalp3 inflammasome sensing of asbestos and silica. Science 320:674- 677. 77. Hornung V, Bauernfeind F, Halle A, Samstad EO, Kono H, Rock KL, Fitzgerald KA, Latz E. 2008. Silica crystals and aluminum salts activate the NALP3 inflammasome through phagosomal destabilization. Nat Immunol 9:847-856. 78. Murakami T, Ockinger J, Yu J, Byles V, McColl A, Hofer AM, Horng T. 2012. Critical role for calcium mobilization in activation of the NLRP3 inflammasome. Proc Natl Acad Sci U S A 109:11282-11287. 79. Munoz-Planillo R, Kuffa P, Martinez-Colon G, Smith BL, Rajendiran TM, Nunez G. 2013. K(+) efflux is the common trigger of NLRP3 inflammasome activation by bacterial toxins and particulate matter. Immunity 38:1142-1153. 80. Petrilli V, Papin S, Dostert C, Mayor A, Martinon F, Tschopp J. 2007. Activation of the NALP3 inflammasome is triggered by low intracellular potassium concentration. Cell Death Differ 14:1583-1589. 81. Mangan MSJ, Olhava EJ, Roush WR, Seidel HM, Glick GD, Latz E. 2018. Targeting the NLRP3 inflammasome in inflammatory diseases. Nat Rev Drug Discov 17:688. 82. Schmid-Burgk JL, Chauhan D, Schmidt T, Ebert TS, Reinhardt J, Endl E, Hornung V. 2016. A Genome-wide CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats) Screen Identifies NEK7 as an Essential Component of NLRP3 Inflammasome Activation. J Biol Chem 291:103-109.

100

83. Shi H, Wang Y, Li X, Zhan X, Tang M, Fina M, Su L, Pratt D, Bu CH, Hildebrand S, Lyon S, Scott L, Quan J, Sun Q, Russell J, Arnett S, Jurek P, Chen D, Kravchenko VV, Mathison JC, Moresco EM, Monson NL, Ulevitch RJ, Beutler B. 2016. NLRP3 activation and mitosis are mutually exclusive events coordinated by NEK7, a new inflammasome component. Nat Immunol 17:250-258. 84. He Y, Zeng MY, Yang D, Motro B, Nunez G. 2016. NEK7 is an essential mediator of NLRP3 activation downstream of potassium efflux. Nature 530:354-357. 85. Ito M, Shichita T, Okada M, Komine R, Noguchi Y, Yoshimura A, Morita R. 2015. Bruton's tyrosine kinase is essential for NLRP3 inflammasome activation and contributes to ischaemic brain injury. Nat Commun 6:7360. 86. Samir P, Kesavardhana S, Patmore DM, Gingras S, Malireddi RKS, Karki R, Guy CS, Briard B, Place DE, Bhattacharya A, Sharma BR, Nourse A, King SV, Pitre A, Burton AR, Pelletier S, Gilbertson RJ, Kanneganti TD. 2019. DDX3X acts as a live-or-die checkpoint in stressed cells by regulating NLRP3 inflammasome. Nature 573:590-594. 87. de Brito OM, Scorrano L. 2008. Mitofusin 2 tethers endoplasmic reticulum to mitochondria. Nature 456:605-610. 88. Misawa T, Takahama M, Kozaki T, Lee H, Zou J, Saitoh T, Akira S. 2013. Microtubule- driven spatial arrangement of mitochondria promotes activation of the NLRP3 inflammasome. Nat Immunol 14:454-460. 89. Chen J, Chen ZJ. 2018. PtdIns4P on dispersed trans-Golgi network mediates NLRP3 inflammasome activation. Nature 564:71-76. 90. Chen KW, Boucher D, Broz P. 2019. Divide to conquer: NLRP3 is activated on dispersed trans- Golgi network. Cell Res 29:181-182. 91. Spalinger MR, Kasper S, Gottier C, Lang S, Atrott K, Vavricka SR, Scharl S, Raselli T, Frey-Wagner I, Gutte PM, Grutter MG, Beer HD, Contassot E, Chan AC, Dai X, Rawlings DJ, Mair F, Becher B, Falk W, Fried M, Rogler G, Scharl M. 2016. NLRP3 tyrosine phosphorylation is controlled by protein tyrosine phosphatase PTPN22. J Clin Invest 126:1783- 1800. 92. Stutz A, Kolbe CC, Stahl R, Horvath GL, Franklin BS, van Ray O, Brinkschulte R, Geyer M, Meissner F, Latz E. 2017. NLRP3 inflammasome assembly is regulated by phosphorylation of the pyrin domain. J Exp Med 214:1725-1736. 93. Song N, Liu ZS, Xue W, Bai ZF, Wang QY, Dai J, Liu X, Huang YJ, Cai H, Zhan XY, Han QY, Wang H, Chen Y, Li HY, Li AL, Zhang XM, Zhou T, Li T. 2017. NLRP3 Phosphorylation Is an Essential Priming Event for Inflammasome Activation. Mol Cell 68:185- 197 e186. 94. Py BF, Kim MS, Vakifahmetoglu-Norberg H, Yuan J. 2013. Deubiquitination of NLRP3 by BRCC3 critically regulates inflammasome activity. Mol Cell 49:331-338. 95. Juliana C, Fernandes-Alnemri T, Kang S, Farias A, Qin F, Alnemri ES. 2012. Non- transcriptional priming and deubiquitination regulate NLRP3 inflammasome activation. J Biol Chem 287:36617-36622. 96. Zhang Z, Meszaros G, He WT, Xu Y, de Fatima Magliarelli H, Mailly L, Mihlan M, Liu Y, Puig Gamez M, Goginashvili A, Pasquier A, Bielska O, Neven B, Quartier P, Aebersold R, Baumert TF, Georgel P, Han J, Ricci R. 2017. Protein kinase D at the Golgi controls NLRP3 inflammasome activation. J Exp Med 214:2671-2693. 97. Han S, Lear TB, Jerome JA, Rajbhandari S, Snavely CA, Gulick DL, Gibson KF, Zou C, Chen BB, Mallampalli RK. 2015. Lipopolysaccharide Primes the NALP3 Inflammasome by Inhibiting Its Ubiquitination and Degradation Mediated by the SCFFBXL2 E3 Ligase. J Biol Chem 290:18124-18133. 98. Song H, Liu B, Huai W, Yu Z, Wang W, Zhao J, Han L, Jiang G, Zhang L, Gao C, Zhao W. 2016. The E3 ubiquitin ligase TRIM31 attenuates NLRP3 inflammasome activation by promoting proteasomal degradation of NLRP3. Nat Commun 7:13727. 101

99. Yan Y, Jiang W, Liu L, Wang X, Ding C, Tian Z, Zhou R. 2015. Dopamine controls systemic inflammation through inhibition of NLRP3 inflammasome. Cell 160:62-73. 100. Booshehri LM, Hoffman HM. 2019. CAPS and NLRP3. J Clin Immunol 39:277-286. 101. Aksentijevich I, Putnam CD, Remmers EF, Mueller JL, Le J, Kolodner RD, Moak Z, Chuang M, Austin F, Goldbach-Mansky R, Hoffman HM, Kastner DL. 2007. The clinical continuum of cryopyrinopathies: novel CIAS1 mutations in North American patients and a new cryopyrin model. Arthritis Rheum 56:1273-1285. 102. Goldbach-Mansky R, Dailey NJ, Canna SW, Gelabert A, Jones J, Rubin BI, Kim HJ, Brewer C, Zalewski C, Wiggs E, Hill S, Turner ML, Karp BI, Aksentijevich I, Pucino F, Penzak SR, Haverkamp MH, Stein L, Adams BS, Moore TL, Fuhlbrigge RC, Shaham B, Jarvis JN, O'Neil K, Vehe RK, Beitz LO, Gardner G, Hannan WP, Warren RW, Horn W, Cole JL, Paul SM, Hawkins PN, Pham TH, Snyder C, Wesley RA, Hoffmann SC, Holland SM, Butman JA, Kastner DL. 2006. Neonatal-onset multisystem inflammatory disease responsive to interleukin-1beta inhibition. N Engl J Med 355:581-592. 103. Hawkins PN, Lachmann HJ, McDermott MF. 2003. Interleukin-1-receptor antagonist in the Muckle-Wells syndrome. N Engl J Med 348:2583-2584. 104. Hoffman HM, Rosengren S, Boyle DL, Cho JY, Nayar J, Mueller JL, Anderson JP, Wanderer AA, Firestein GS. 2004. Prevention of cold-associated acute inflammation in familial cold autoinflammatory syndrome by interleukin-1 receptor antagonist. Lancet 364:1779-1785. 105. Masters SL, Simon A, Aksentijevich I, Kastner DL. 2009. Horror autoinflammaticus: the molecular pathophysiology of autoinflammatory disease (*). Annu Rev Immunol 27:621-668. 106. Coll RC, Robertson AA, Chae JJ, Higgins SC, Munoz-Planillo R, Inserra MC, Vetter I, Dungan LS, Monks BG, Stutz A, Croker DE, Butler MS, Haneklaus M, Sutton CE, Nunez G, Latz E, Kastner DL, Mills KH, Masters SL, Schroder K, Cooper MA, O'Neill LA. 2015. A small-molecule inhibitor of the NLRP3 inflammasome for the treatment of inflammatory diseases. Nat Med 21:248-255. 107. Coll RC, Hill JR, Day CJ, Zamoshnikova A, Boucher D, Massey NL, Chitty JL, Fraser JA, Jennings MP, Robertson AAB, Schroder K. 2019. MCC950 directly targets the NLRP3 ATP- hydrolysis motif for inflammasome inhibition. Nat Chem Biol 15:556-559. 108. Duncan JA, Bergstralh DT, Wang Y, Willingham SB, Ye Z, Zimmermann AG, Ting JP. 2007. Cryopyrin/NALP3 binds ATP/dATP, is an ATPase, and requires ATP binding to mediate inflammatory signaling. Proc Natl Acad Sci U S A 104:8041-8046. 109. Jiang H, He H, Chen Y, Huang W, Cheng J, Ye J, Wang A, Tao J, Wang C, Liu Q, Jin T, Jiang W, Deng X, Zhou R. 2017. Identification of a selective and direct NLRP3 inhibitor to treat inflammatory disorders. J Exp Med 214:3219-3238. 110. Huang Y, Jiang H, Chen Y, Wang X, Yang Y, Tao J, Deng X, Liang G, Zhang H, Jiang W, Zhou R. 2018. Tranilast directly targets NLRP3 to treat inflammasome-driven diseases. EMBO Mol Med 10. 111. He H, Jiang H, Chen Y, Ye J, Wang A, Wang C, Liu Q, Liang G, Deng X, Jiang W, Zhou R. 2018. Oridonin is a covalent NLRP3 inhibitor with strong anti-inflammasome activity. Nat Commun 9:2550. 112. Youm YH, Nguyen KY, Grant RW, Goldberg EL, Bodogai M, Kim D, D'Agostino D, Planavsky N, Lupfer C, Kanneganti TD, Kang S, Horvath TL, Fahmy TM, Crawford PA, Biragyn A, Alnemri E, Dixit VD. 2015. The ketone metabolite beta-hydroxybutyrate blocks NLRP3 inflammasome-mediated inflammatory disease. Nat Med 21:263-269. 113. Poyet JL, Srinivasula SM, Tnani M, Razmara M, Fernandes-Alnemri T, Alnemri ES. 2001. Identification of Ipaf, a human caspase-1-activating protein related to Apaf-1. J Biol Chem 276:28309-28313. 114. Kofoed EM, Vance RE. 2011. Innate immune recognition of bacterial ligands by NAIPs determines inflammasome specificity. Nature 477:592-595. 102

115. Zhao Y, Yang J, Shi J, Gong YN, Lu Q, Xu H, Liu L, Shao F. 2011. The NLRC4 inflammasome receptors for bacterial flagellin and type III secretion apparatus. Nature 477:596- 600. 116. Yang J, Zhao Y, Shi J, Shao F. 2013. Human NAIP and mouse NAIP1 recognize bacterial type III secretion needle protein for inflammasome activation. Proc Natl Acad Sci U S A 110:14408- 14413. 117. Kortmann J, Brubaker SW, Monack DM. 2015. Cutting Edge: Inflammasome Activation in Primary Human Macrophages Is Dependent on Flagellin. J Immunol 195:815-819. 118. Karki R, Lee E, Place D, Samir P, Mavuluri J, Sharma BR, Balakrishnan A, Malireddi RKS, Geiger R, Zhu Q, Neale G, Kanneganti TD. 2018. IRF8 Regulates Transcription of Naips for NLRC4 Inflammasome Activation. Cell 173:920-933 e913. 119. Hu Z, Yan C, Liu P, Huang Z, Ma R, Zhang C, Wang R, Zhang Y, Martinon F, Miao D, Deng H, Wang J, Chang J, Chai J. 2013. Crystal structure of NLRC4 reveals its autoinhibition mechanism. Science 341:172-175. 120. Zhang L, Chen S, Ruan J, Wu J, Tong AB, Yin Q, Li Y, David L, Lu A, Wang WL, Marks C, Ouyang Q, Zhang X, Mao Y, Wu H. 2015. Cryo-EM structure of the activated NAIP2- NLRC4 inflammasome reveals nucleated polymerization. Science 350:404-409. 121. Hu Z, Zhou Q, Zhang C, Fan S, Cheng W, Zhao Y, Shao F, Wang HW, Sui SF, Chai J. 2015. Structural and biochemical basis for induced self-propagation of NLRC4. Science 350:399- 404. 122. Liu Z, Xiao TS. 2015. STRUCTURAL BIOLOGY. Assembling the wheel of death. Science 350:376-377. 123. Case CL, Shin S, Roy CR. 2009. Asc and Ipaf Inflammasomes direct distinct pathways for caspase-1 activation in response to Legionella pneumophila. Infect Immun 77:1981-1991. 124. Man SM, Hopkins LJ, Nugent E, Cox S, Gluck IM, Tourlomousis P, Wright JA, Cicuta P, Monie TP, Bryant CE. 2014. Inflammasome activation causes dual recruitment of NLRC4 and NLRP3 to the same macromolecular complex. Proc Natl Acad Sci U S A 111:7403-7408. 125. Mascarenhas DPA, Cerqueira DM, Pereira MSF, Castanheira FVS, Fernandes TD, Manin GZ, Cunha LD, Zamboni DS. 2017. Inhibition of caspase-1 or gasdermin-D enable caspase-8 activation in the Naip5/NLRC4/ASC inflammasome. PLoS Pathog 13:e1006502. 126. Qu Y, Misaghi S, Izrael-Tomasevic A, Newton K, Gilmour LL, Lamkanfi M, Louie S, Kayagaki N, Liu J, Komuves L, Cupp JE, Arnott D, Monack D, Dixit VM. 2012. Phosphorylation of NLRC4 is critical for inflammasome activation. Nature 490:539-542. 127. Liu W, Liu X, Li Y, Zhao J, Liu Z, Hu Z, Wang Y, Yao Y, Miller AW, Su B, Cookson MR, Li X, Kang Z. 2017. LRRK2 promotes the activation of NLRC4 inflammasome during Salmonella Typhimurium infection. J Exp Med 214:3051-3066. 128. Matusiak M, Van Opdenbosch N, Vande Walle L, Sirard JC, Kanneganti TD, Lamkanfi M. 2015. Flagellin-induced NLRC4 phosphorylation primes the inflammasome for activation by NAIP5. Proc Natl Acad Sci U S A 112:1541-1546. 129. Lee E, Karki R, Kanneganti TD. 2018. The NLRC4 inflammasome requires IRF8-dependent production of NAIPs. Cell Stress 2:144-146. 130. Vanaja SK, Rathinam VAK. 2018. (IR)Factor for NAIP Expression. Cell 173:817-819. 131. Canna SW, de Jesus AA, Gouni S, Brooks SR, Marrero B, Liu Y, DiMattia MA, Zaal KJ, Sanchez GA, Kim H, Chapelle D, Plass N, Huang Y, Villarino AV, Biancotto A, Fleisher TA, Duncan JA, O'Shea JJ, Benseler S, Grom A, Deng Z, Laxer RM, Goldbach-Mansky R. 2014. An activating NLRC4 inflammasome mutation causes autoinflammation with recurrent macrophage activation syndrome. Nat Genet 46:1140-1146. 132. Romberg N, Al Moussawi K, Nelson-Williams C, Stiegler AL, Loring E, Choi M, Overton J, Meffre E, Khokha MK, Huttner AJ, West B, Podoltsev NA, Boggon TJ, Kazmierczak BI, Lifton RP. 2014. Mutation of NLRC4 causes a syndrome of enterocolitis and autoinflammation. Nat Genet 46:1135-1139. 103

133. Volker-Touw CM, de Koning HD, Giltay JC, de Kovel CG, van Kempen TS, Oberndorff KM, Boes ML, van Steensel MA, van Well GT, Blokx WA, Schalkwijk J, Simon A, Frenkel J, van Gijn ME. 2017. Erythematous nodes, urticarial rash and arthralgias in a large pedigree with NLRC4-related autoinflammatory disease, expansion of the phenotype. Br J Dermatol 176:244-248. 134. Kitamura A, Sasaki Y, Abe T, Kano H, Yasutomo K. 2014. An inherited mutation in NLRC4 causes autoinflammation in human and mice. J Exp Med 211:2385-2396. 135. Muruve DA, Petrilli V, Zaiss AK, White LR, Clark SA, Ross PJ, Parks RJ, Tschopp J. 2008. The inflammasome recognizes cytosolic microbial and host DNA and triggers an innate immune response. Nature 452:103-107. 136. Roberts TL, Idris A, Dunn JA, Kelly GM, Burnton CM, Hodgson S, Hardy LL, Garceau V, Sweet MJ, Ross IL, Hume DA, Stacey KJ. 2009. HIN-200 proteins regulate caspase activation in response to foreign cytoplasmic DNA. Science 323:1057-1060. 137. Hornung V, Ablasser A, Charrel-Dennis M, Bauernfeind F, Horvath G, Caffrey DR, Latz E, Fitzgerald KA. 2009. AIM2 recognizes cytosolic dsDNA and forms a caspase-1-activating inflammasome with ASC. Nature 458:514-518. 138. Fernandes-Alnemri T, Yu JW, Datta P, Wu J, Alnemri ES. 2009. AIM2 activates the inflammasome and cell death in response to cytoplasmic DNA. Nature 458:509-513. 139. Burckstummer T, Baumann C, Bluml S, Dixit E, Durnberger G, Jahn H, Planyavsky M, Bilban M, Colinge J, Bennett KL, Superti-Furga G. 2009. An orthogonal proteomic-genomic screen identifies AIM2 as a cytoplasmic DNA sensor for the inflammasome. Nat Immunol 10:266-272. 140. Rathinam VA, Jiang Z, Waggoner SN, Sharma S, Cole LE, Waggoner L, Vanaja SK, Monks BG, Ganesan S, Latz E, Hornung V, Vogel SN, Szomolanyi-Tsuda E, Fitzgerald KA. 2010. The AIM2 inflammasome is essential for host defense against cytosolic bacteria and DNA viruses. Nat Immunol 11:395-402. 141. Fernandes-Alnemri T, Yu JW, Juliana C, Solorzano L, Kang S, Wu J, Datta P, McCormick M, Huang L, McDermott E, Eisenlohr L, Landel CP, Alnemri ES. 2010. The AIM2 inflammasome is critical for innate immunity to Francisella tularensis. Nat Immunol 11:385-393. 142. Jones JW, Kayagaki N, Broz P, Henry T, Newton K, O'Rourke K, Chan S, Dong J, Qu Y, Roose-Girma M, Dixit VM, Monack DM. 2010. Absent in melanoma 2 is required for innate immune recognition of Francisella tularensis. Proc Natl Acad Sci U S A 107:9771-9776. 143. Jin T, Perry A, Jiang J, Smith P, Curry JA, Unterholzner L, Jiang Z, Horvath G, Rathinam VA, Johnstone RW, Hornung V, Latz E, Bowie AG, Fitzgerald KA, Xiao TS. 2012. Structures of the HIN domain:DNA complexes reveal ligand binding and activation mechanisms of the AIM2 inflammasome and IFI16 receptor. Immunity 36:561-571. 144. Sauer JD, Witte CE, Zemansky J, Hanson B, Lauer P, Portnoy DA. 2010. Listeria monocytogenes triggers AIM2-mediated pyroptosis upon infrequent bacteriolysis in the macrophage cytosol. Cell Host Microbe 7:412-419. 145. Rathinam VA, Vanaja SK, Waggoner L, Sokolovska A, Becker C, Stuart LM, Leong JM, Fitzgerald KA. 2012. TRIF licenses caspase-11-dependent NLRP3 inflammasome activation by gram-negative bacteria. Cell 150:606-619. 146. Broz P, Ruby T, Belhocine K, Bouley DM, Kayagaki N, Dixit VM, Monack DM. 2012. Caspase-11 increases susceptibility to Salmonella infection in the absence of caspase-1. Nature 490:288-291. 147. Meunier E, Dick MS, Dreier RF, Schurmann N, Kenzelmann Broz D, Warming S, Roose- Girma M, Bumann D, Kayagaki N, Takeda K, Yamamoto M, Broz P. 2014. Caspase-11 activation requires lysis of pathogen-containing vacuoles by IFN-induced GTPases. Nature 509:366-370. 104

148. Pilla DM, Hagar JA, Haldar AK, Mason AK, Degrandi D, Pfeffer K, Ernst RK, Yamamoto M, Miao EA, Coers J. 2014. Guanylate binding proteins promote caspase-11-dependent pyroptosis in response to cytoplasmic LPS. Proc Natl Acad Sci U S A 111:6046-6051. 149. Man SM, Karki R, Sasai M, Place DE, Kesavardhana S, Temirov J, Frase S, Zhu Q, Malireddi RKS, Kuriakose T, Peters JL, Neale G, Brown SA, Yamamoto M, Kanneganti TD. 2016. IRGB10 Liberates Bacterial Ligands for Sensing by the AIM2 and Caspase-11-NLRP3 Inflammasomes. Cell 167:382-396 e317. 150. Meunier E, Wallet P, Dreier RF, Costanzo S, Anton L, Ruhl S, Dussurgey S, Dick MS, Kistner A, Rigard M, Degrandi D, Pfeffer K, Yamamoto M, Henry T, Broz P. 2015. Guanylate-binding proteins promote activation of the AIM2 inflammasome during infection with Francisella novicida. Nat Immunol 16:476-484. 151. Man SM, Karki R, Malireddi RK, Neale G, Vogel P, Yamamoto M, Lamkanfi M, Kanneganti TD. 2015. The transcription factor IRF1 and guanylate-binding proteins target activation of the AIM2 inflammasome by Francisella infection. Nat Immunol 16:467-475. 152. Schattgen SA, Gao G, Kurt-Jones EA, Fitzgerald KA. 2016. Cutting Edge: DNA in the Lung Microenvironment during Influenza Virus Infection Tempers Inflammation by Engaging the DNA Sensor AIM2. J Immunol 196:29-33. 153. Zhang H, Luo J, Alcorn JF, Chen K, Fan S, Pilewski J, Liu A, Chen W, Kolls JK, Wang J. 2017. AIM2 Inflammasome Is Critical for Influenza-Induced Lung Injury and Mortality. J Immunol 198:4383-4393. 154. Di Micco A, Frera G, Lugrin J, Jamilloux Y, Hsu ET, Tardivel A, De Gassart A, Zaffalon L, Bujisic B, Siegert S, Quadroni M, Broz P, Henry T, Hrycyna CA, Martinon F. 2016. AIM2 inflammasome is activated by pharmacological disruption of nuclear envelope integrity. Proc Natl Acad Sci U S A 113:E4671-4680. 155. Lugrin J, Martinon F. 2018. The AIM2 inflammasome: Sensor of pathogens and cellular perturbations. Immunol Rev 281:99-114. 156. Caneparo V, Landolfo S, Gariglio M, De Andrea M. 2018. The Absent in Melanoma 2-Like Receptor IFN-Inducible Protein 16 as an Inflammasome Regulator in Systemic Lupus Erythematosus: The Dark Side of Sensing Microbes. Front Immunol 9:1180. 157. Ru H, Ni X, Zhao L, Crowley C, Ding W, Hung LW, Shaw N, Cheng G, Liu ZJ. 2013. Structural basis for termination of AIM2-mediated signaling by p202. Cell Res 23:855-858. 158. Yin Q, Sester DP, Tian Y, Hsiao YS, Lu A, Cridland JA, Sagulenko V, Thygesen SJ, Choubey D, Hornung V, Walz T, Stacey KJ, Wu H. 2013. Molecular mechanism for p202- mediated specific inhibition of AIM2 inflammasome activation. Cell Rep 4:327-339. 159. Komada T, Chung H, Lau A, Platnich JM, Beck PL, Benediktsson H, Duff HJ, Jenne CN, Muruve DA. 2018. Macrophage Uptake of Necrotic Cell DNA Activates the AIM2 Inflammasome to Regulate a Proinflammatory Phenotype in CKD. J Am Soc Nephrol 29:1165- 1181. 160. de Koning HD, Bergboer JG, van den Bogaard EH, van Vlijmen-Willems IM, Rodijk- Olthuis D, Simon A, Zeeuwen PL, Schalkwijk J. 2012. Strong induction of AIM2 expression in human epidermis in acute and chronic inflammatory skin conditions. Exp Dermatol 21:961-964. 161. Denes A, Coutts G, Lenart N, Cruickshank SM, Pelegrin P, Skinner J, Rothwell N, Allan SM, Brough D. 2015. AIM2 and NLRC4 inflammasomes contribute with ASC to acute brain injury independently of NLRP3. Proc Natl Acad Sci U S A 112:4050-4055. 162. Adamczak SE, de Rivero Vaccari JP, Dale G, Brand FJ, 3rd, Nonner D, Bullock MR, Dahl GP, Dietrich WD, Keane RW. 2014. Pyroptotic neuronal cell death mediated by the AIM2 inflammasome. J Cereb Blood Flow Metab 34:621-629. 163. Wang PH, Ye ZW, Deng JJ, Siu KL, Gao WW, Chaudhary V, Cheng Y, Fung SY, Yuen KS, Ho TH, Chan CP, Zhang Y, Kok KH, Yang W, Chan CP, Jin DY. 2018. Inhibition of AIM2 inflammasome activation by a novel transcript isoform of IFI16. EMBO Rep 19. 105

164. Liu T, Tang Q, Liu K, Xie W, Liu X, Wang H, Wang RF, Cui J. 2016. TRIM11 Suppresses AIM2 Inflammasome by Degrading AIM2 via p62-Dependent Selective Autophagy. Cell Rep 16:1988-2002. 165. Kaminski JJ, Schattgen SA, Tzeng TC, Bode C, Klinman DM, Fitzgerald KA. 2013. Synthetic oligodeoxynucleotides containing suppressive TTAGGG motifs inhibit AIM2 inflammasome activation. J Immunol 191:3876-3883. 166. Chae JJ, Cho YH, Lee GS, Cheng J, Liu PP, Feigenbaum L, Katz SI, Kastner DL. 2011. Gain-of-function Pyrin mutations induce NLRP3 protein-independent interleukin-1beta activation and severe autoinflammation in mice. Immunity 34:755-768. 167. Yu JW, Fernandes-Alnemri T, Datta P, Wu J, Juliana C, Solorzano L, McCormick M, Zhang Z, Alnemri ES. 2007. Pyrin activates the ASC pyroptosome in response to engagement by autoinflammatory PSTPIP1 mutants. Mol Cell 28:214-227. 168. Xu H, Yang J, Gao W, Li L, Li P, Zhang L, Gong YN, Peng X, Xi JJ, Chen S, Wang F, Shao F. 2014. Innate immune sensing of bacterial modifications of Rho GTPases by the Pyrin inflammasome. Nature 513:237-241. 169. Aktories K. 2011. Bacterial protein toxins that modify host regulatory GTPases. Nat Rev Microbiol 9:487-498. 170. Hall A. 1998. Rho GTPases and the actin cytoskeleton. Science 279:509-514. 171. Park YH, Wood G, Kastner DL, Chae JJ. 2016. Pyrin inflammasome activation and RhoA signaling in the autoinflammatory diseases FMF and HIDS. Nat Immunol 17:914-921. 172. Gao W, Yang J, Liu W, Wang Y, Shao F. 2016. Site-specific phosphorylation and microtubule dynamics control Pyrin inflammasome activation. Proc Natl Acad Sci U S A 113:E4857-4866. 173. Van Gorp H, Saavedra PH, de Vasconcelos NM, Van Opdenbosch N, Vande Walle L, Matusiak M, Prencipe G, Insalaco A, Van Hauwermeiren F, Demon D, Bogaert DJ, Dullaers M, De Baere E, Hochepied T, Dehoorne J, Vermaelen KY, Haerynck F, De Benedetti F, Lamkanfi M. 2016. Familial Mediterranean fever mutations lift the obligatory requirement for microtubules in Pyrin inflammasome activation. Proc Natl Acad Sci U S A 113:14384-14389. 174. van Gent M, Sparrer KMJ, Gack MU. 2018. TRIM Proteins and Their Roles in Antiviral Host Defenses. Annu Rev Virol 5:385-405. 175. Kimura T, Jain A, Choi SW, Mandell MA, Schroder K, Johansen T, Deretic V. 2015. TRIM- mediated precision autophagy targets cytoplasmic regulators of innate immunity. J Cell Biol 210:973-989. 176. Papin S, Cuenin S, Agostini L, Martinon F, Werner S, Beer HD, Grutter C, Grutter M, Tschopp J. 2007. The SPRY domain of Pyrin, mutated in familial Mediterranean fever patients, interacts with inflammasome components and inhibits proIL-1beta processing. Cell Death Differ 14:1457-1466. 177. Chae JJ, Wood G, Masters SL, Richard K, Park G, Smith BJ, Kastner DL. 2006. The B30.2 domain of pyrin, the familial Mediterranean fever protein, interacts directly with caspase-1 to modulate IL-1beta production. Proc Natl Acad Sci U S A 103:9982-9987. 178. Schnappauf O, Chae JJ, Kastner DL, Aksentijevich I. 2019. The Pyrin Inflammasome in Health and Disease. Front Immunol 10:1745. 179. Masters SL, Lagou V, Jeru I, Baker PJ, Van Eyck L, Parry DA, Lawless D, De Nardo D, Garcia-Perez JE, Dagley LF, Holley CL, Dooley J, Moghaddas F, Pasciuto E, Jeandel PY, Sciot R, Lyras D, Webb AI, Nicholson SE, De Somer L, van Nieuwenhove E, Ruuth-Praz J, Copin B, Cochet E, Medlej-Hashim M, Megarbane A, Schroder K, Savic S, Goris A, Amselem S, Wouters C, Liston A. 2016. Familial autoinflammation with neutrophilic dermatosis reveals a regulatory mechanism of pyrin activation. Sci Transl Med 8:332ra345.

106

180. Moghaddas F, Llamas R, De Nardo D, Martinez-Banaclocha H, Martinez-Garcia JJ, Mesa- Del-Castillo P, Baker PJ, Gargallo V, Mensa-Vilaro A, Canna S, Wicks IP, Pelegrin P, Arostegui JI, Masters SL. 2017. A novel Pyrin-Associated Autoinflammation with Neutrophilic Dermatosis mutation further defines 14-3-3 binding of pyrin and distinction to Familial Mediterranean Fever. Ann Rheum Dis 76:2085-2094. 181. Stoffels M, Szperl A, Simon A, Netea MG, Plantinga TS, van Deuren M, Kamphuis S, Lachmann HJ, Cuppen E, Kloosterman WP, Frenkel J, van Diemen CC, Wijmenga C, van Gijn M, van der Meer JW. 2014. MEFV mutations affecting pyrin amino acid 577 cause autosomal dominant autoinflammatory disease. Ann Rheum Dis 73:455-461. 182. Nakaseko H, Iwata N, Izawa K, Shibata H, Yasuoka R, Kohagura T, Abe N, Kawabe S, Nishikomori R. 2019. Expanding clinical spectrum of autosomal dominant pyrin-associated autoinflammatory disorder caused by the heterozygous MEFV p.Thr577Asn variant. Rheumatology (Oxford) 58:182-184. 183. Cookson BT, Brennan MA. 2001. Pro-inflammatory programmed cell death. Trends Microbiol 9:113-114. 184. Boise LH, Collins CM. 2001. Salmonella-induced cell death: apoptosis, necrosis or programmed cell death? Trends Microbiol 9:64-67. 185. Aachoui Y, Leaf IA, Hagar JA, Fontana MF, Campos CG, Zak DE, Tan MH, Cotter PA, Vance RE, Aderem A, Miao EA. 2013. Caspase-11 protects against bacteria that escape the vacuole. Science 339:975-978. 186. Vanaja SK, Russo AJ, Behl B, Banerjee I, Yankova M, Deshmukh SD, Rathinam VAK. 2016. Bacterial Outer Membrane Vesicles Mediate Cytosolic Localization of LPS and Caspase- 11 Activation. Cell 165:1106-1119. 187. Santos JC, Dick MS, Lagrange B, Degrandi D, Pfeffer K, Yamamoto M, Meunier E, Pelczar P, Henry T, Broz P. 2018. LPS targets host guanylate-binding proteins to the bacterial outer membrane for non-canonical inflammasome activation. EMBO J 37. 188. Deng M, Tang Y, Li W, Wang X, Zhang R, Zhang X, Zhao X, Liu J, Tang C, Liu Z, Huang Y, Peng H, Xiao L, Tang D, Scott MJ, Wang Q, Liu J, Xiao X, Watkins S, Li J, Yang H, Wang H, Chen F, Tracey KJ, Billiar TR, Lu B. 2018. The Endotoxin Delivery Protein HMGB1 Mediates Caspase-11-Dependent Lethality in Sepsis. Immunity 49:740-753 e747. 189. Rivers-Auty J, Brough D. 2015. Potassium efflux fires the canon: Potassium efflux as a common trigger for canonical and noncanonical NLRP3 pathways. Eur J Immunol 45:2758-2761. 190. Ruhl S, Broz P. 2015. Caspase-11 activates a canonical NLRP3 inflammasome by promoting K(+) efflux. Eur J Immunol 45:2927-2936. 191. de Carvalho RVH, Andrade WA, Lima-Junior DS, Dilucca M, de Oliveira CV, Wang K, Nogueira PM, Rugani JN, Soares RP, Beverley SM, Shao F, Zamboni DS. 2019. Leishmania Lipophosphoglycan Triggers Caspase-11 and the Non-canonical Activation of the NLRP3 Inflammasome. Cell Rep 26:429-437 e425. 192. Shi J, Zhao Y, Wang K, Shi X, Wang Y, Huang H, Zhuang Y, Cai T, Wang F, Shao F. 2015. Cleavage of GSDMD by inflammatory caspases determines pyroptotic cell death. Nature 526:660-665. 193. Kayagaki N, Stowe IB, Lee BL, O'Rourke K, Anderson K, Warming S, Cuellar T, Haley B, Roose-Girma M, Phung QT, Liu PS, Lill JR, Li H, Wu J, Kummerfeld S, Zhang J, Lee WP, Snipas SJ, Salvesen GS, Morris LX, Fitzgerald L, Zhang Y, Bertram EM, Goodnow CC, Dixit VM. 2015. Caspase-11 cleaves gasdermin D for non-canonical inflammasome signalling. Nature 526:666-671. 194. Liu X, Zhang Z, Ruan J, Pan Y, Magupalli VG, Wu H, Lieberman J. 2016. Inflammasome- activated gasdermin D causes pyroptosis by forming membrane pores. Nature 535:153-158. 195. Sborgi L, Ruhl S, Mulvihill E, Pipercevic J, Heilig R, Stahlberg H, Farady CJ, Muller DJ, Broz P, Hiller S. 2016. GSDMD membrane pore formation constitutes the mechanism of pyroptotic cell death. EMBO J 35:1766-1778. 107

196. Evavold CL, Ruan J, Tan Y, Xia S, Wu H, Kagan JC. 2018. The Pore-Forming Protein Gasdermin D Regulates Interleukin-1 Secretion from Living Macrophages. Immunity 48:35-44 e36. 197. Kuriakose T, Kanneganti TD. 2018. Gasdermin D Flashes an Exit Signal for IL-1. Immunity 48:1-3. 198. Chen KW, Gross CJ, Sotomayor FV, Stacey KJ, Tschopp J, Sweet MJ, Schroder K. 2014. The neutrophil NLRC4 inflammasome selectively promotes IL-1beta maturation without pyroptosis during acute Salmonella challenge. Cell Rep 8:570-582. 199. Diamond CE, Leong KWK, Vacca M, Rivers-Auty J, Brough D, Mortellaro A. 2017. Salmonella typhimurium-induced IL-1 release from primary human monocytes requires NLRP3 and can occur in the absence of pyroptosis. Sci Rep 7:6861. 200. Vigano E, Diamond CE, Spreafico R, Balachander A, Sobota RM, Mortellaro A. 2015. Human caspase-4 and caspase-5 regulate the one-step non-canonical inflammasome activation in monocytes. Nat Commun 6:8761. 201. Monteleone M, Stanley AC, Chen KW, Brown DL, Bezbradica JS, von Pein JB, Holley CL, Boucher D, Shakespear MR, Kapetanovic R, Rolfes V, Sweet MJ, Stow JL, Schroder K. 2018. Interleukin-1beta Maturation Triggers Its Relocation to the Plasma Membrane for Gasdermin-D-Dependent and -Independent Secretion. Cell Rep 24:1425-1433. 202. Kimura T, Jia J, Kumar S, Choi SW, Gu Y, Mudd M, Dupont N, Jiang S, Peters R, Farzam F, Jain A, Lidke KA, Adams CM, Johansen T, Deretic V. 2017. Dedicated SNAREs and specialized TRIM cargo receptors mediate secretory autophagy. EMBO J 36:42-60. 203. Black RA, Kronheim SR, Merriam JE, March CJ, Hopp TP. 1989. A pre-aspartate-specific protease from human leukocytes that cleaves pro-interleukin-1 beta. J Biol Chem 264:5323-5326. 204. Kostura MJ, Tocci MJ, Limjuco G, Chin J, Cameron P, Hillman AG, Chartrain NA, Schmidt JA. 1989. Identification of a monocyte specific pre-interleukin 1 beta convertase activity. Proc Natl Acad Sci U S A 86:5227-5231. 205. Thornberry NA, Bull HG, Calaycay JR, Chapman KT, Howard AD, Kostura MJ, Miller DK, Molineaux SM, Weidner JR, Aunins J, et al. 1992. A novel heterodimeric cysteine protease is required for interleukin-1 beta processing in monocytes. Nature 356:768-774. 206. Walsh JG, Logue SE, Luthi AU, Martin SJ. 2011. Caspase-1 promiscuity is counterbalanced by rapid inactivation of processed enzyme. J Biol Chem 286:32513-32524. 207. Broz P, von Moltke J, Jones JW, Vance RE, Monack DM. 2010. Differential requirement for Caspase-1 autoproteolysis in pathogen-induced cell death and cytokine processing. Cell Host Microbe 8:471-483. 208. Sandstrom A, Vance RE. 2018. Defusing inflammasomes. J Exp Med 215:723-724. 209. Agard NJ, Maltby D, Wells JA. 2010. Inflammatory stimuli regulate caspase substrate profiles. Mol Cell Proteomics 9:880-893. 210. Van Opdenbosch N, Lamkanfi M. 2019. Caspases in Cell Death, Inflammation, and Disease. Immunity 50:1352-1364. 211. Lagrange B, Benaoudia S, Wallet P, Magnotti F, Provost A, Michal F, Martin A, Di Lorenzo F, Py BF, Molinaro A, Henry T. 2018. Human caspase-4 detects tetra-acylated LPS and cytosolic Francisella and functions differently from murine caspase-11. Nat Commun 9:242. 212. Bekpen C, Hunn JP, Rohde C, Parvanova I, Guethlein L, Dunn DM, Glowalla E, Leptin M, Howard JC. 2005. The interferon-inducible p47 (IRG) GTPases in vertebrates: loss of the cell autonomous resistance mechanism in the human lineage. Genome Biol 6:R92. 213. Baker PJ, Boucher D, Bierschenk D, Tebartz C, Whitney PG, D'Silva DB, Tanzer MC, Monteleone M, Robertson AA, Cooper MA, Alvarez-Diaz S, Herold MJ, Bedoui S, Schroder K, Masters SL. 2015. NLRP3 inflammasome activation downstream of cytoplasmic LPS recognition by both caspase-4 and caspase-5. Eur J Immunol 45:2918-2926. 214. Afonina IS, Muller C, Martin SJ, Beyaert R. 2015. Proteolytic Processing of Interleukin-1 Family Cytokines: Variations on a Common Theme. Immunity 42:991-1004. 108

215. Antonopoulos C, El Sanadi C, Kaiser WJ, Mocarski ES, Dubyak GR. 2013. Proapoptotic chemotherapeutic drugs induce noncanonical processing and release of IL-1beta via caspase-8 in dendritic cells. J Immunol 191:4789-4803. 216. Maelfait J, Vercammen E, Janssens S, Schotte P, Haegman M, Magez S, Beyaert R. 2008. Stimulation of Toll-like receptor 3 and 4 induces interleukin-1beta maturation by caspase-8. J Exp Med 205:1967-1973. 217. Shenderov K, Riteau N, Yip R, Mayer-Barber KD, Oland S, Hieny S, Fitzgerald P, Oberst A, Dillon CP, Green DR, Cerundolo V, Sher A. 2014. Cutting edge: Endoplasmic reticulum stress licenses macrophages to produce mature IL-1beta in response to TLR4 stimulation through a caspase-8- and TRIF-dependent pathway. J Immunol 192:2029-2033. 218. Bossaller L, Chiang PI, Schmidt-Lauber C, Ganesan S, Kaiser WJ, Rathinam VA, Mocarski ES, Subramanian D, Green DR, Silverman N, Fitzgerald KA, Marshak-Rothstein A, Latz E. 2012. Cutting edge: FAS (CD95) mediates noncanonical IL-1beta and IL-18 maturation via caspase-8 in an RIP3-independent manner. J Immunol 189:5508-5512. 219. Antonopoulos C, Russo HM, El Sanadi C, Martin BN, Li X, Kaiser WJ, Mocarski ES, Dubyak GR. 2015. Caspase-8 as an Effector and Regulator of NLRP3 Inflammasome Signaling. J Biol Chem 290:20167-20184. 220. Gringhuis SI, Kaptein TM, Wevers BA, Theelen B, van der Vlist M, Boekhout T, Geijtenbeek TB. 2012. Dectin-1 is an extracellular pathogen sensor for the induction and processing of IL-1beta via a noncanonical caspase-8 inflammasome. Nat Immunol 13:246-254. 221. Goncalves AV, Margolis SR, Quirino GFS, Mascarenhas DPA, Rauch I, Nichols RD, Ansaldo E, Fontana MF, Vance RE, Zamboni DS. 2019. Gasdermin-D and Caspase-7 are the key Caspase-1/8 substrates downstream of the NAIP5/NLRC4 inflammasome required for restriction of Legionella pneumophila. PLoS Pathog 15:e1007886. 222. Schneider KS, Gross CJ, Dreier RF, Saller BS, Mishra R, Gorka O, Heilig R, Meunier E, Dick MS, Cikovic T, Sodenkamp J, Medard G, Naumann R, Ruland J, Kuster B, Broz P, Gross O. 2017. The Inflammasome Drives GSDMD-Independent Secondary Pyroptosis and IL-1 Release in the Absence of Caspase-1 Protease Activity. Cell Rep 21:3846-3859. 223. Rogers C, Fernandes-Alnemri T, Mayes L, Alnemri D, Cingolani G, Alnemri ES. 2017. Cleavage of DFNA5 by caspase-3 during apoptosis mediates progression to secondary necrotic/pyroptotic cell death. Nat Commun 8:14128. 224. Wang Y, Gao W, Shi X, Ding J, Liu W, He H, Wang K, Shao F. 2017. Chemotherapy drugs induce pyroptosis through caspase-3 cleavage of a gasdermin. Nature 547:99-103. 225. Lee BL, Mirrashidi KM, Stowe IB, Kummerfeld SK, Watanabe C, Haley B, Cuellar TL, Reichelt M, Kayagaki N. 2018. ASC- and caspase-8-dependent apoptotic pathway diverges from the NLRC4 inflammasome in macrophages. Sci Rep 8:3788. 226. Compan V, Baroja-Mazo A, Bragg L, Verkhratsky A, Perroy J, Pelegrin P. 2012. A genetically encoded IL-1beta bioluminescence resonance energy transfer sensor to monitor inflammasome activity. J Immunol 189:2131-2137. 227. Bartok E, Bauernfeind F, Khaminets MG, Jakobs C, Monks B, Fitzgerald KA, Latz E, Hornung V. 2013. iGLuc: a luciferase-based inflammasome and protease activity reporter. Nat Methods 10:147-154. 228. O'Brien M, Moehring D, Munoz-Planillo R, Nunez G, Callaway J, Ting J, Scurria M, Ugo T, Bernad L, Cali J, Lazar D. 2017. A bioluminescent caspase-1 activity assay rapidly monitors inflammasome activation in cells. J Immunol Methods 447:1-13. 229. Ko YJ, Lee JW, Yang EJ, Jang N, Park J, Jeon YK, Yu JW, Cho NH, Kim HS, Chan Kwon I. 2020. Non-invasive in vivo imaging of caspase-1 activation enables rapid and spatiotemporal detection of acute and chronic inflammatory disorders. Biomaterials 226:119543.

109

230. Galban S, Jeon YH, Bowman BM, Stevenson J, Sebolt KA, Sharkey LM, Lafferty M, Hoff BA, Butler BL, Wigdal SS, Binkowski BF, Otto P, Zimmerman K, Vidugiris G, Encell LP, Fan F, Wood KV, Galban CJ, Ross BD, Rehemtulla A. 2013. Imaging proteolytic activity in live cells and animal models. PLoS One 8:e66248. 231. Fan F BB, Butler BL, Stecha PF, Lewis MK, Wood KV. 2008. Novel genetically encoded biosensors using firefly luciferase. ACS Chem Biol 3:346-351. 232. Susan S. Wigdal JLA, Gediminas J. Vidugiris, John Shultz, Keith V. Wood and Frank Fan. 2008. A Novel Bioluminescent Protease Assay Using Engineered Firefly Luciferase. Current Chemical Genomics 2:16-28. 233. Andy Kilianski AMM, Xufang Deng and Susan Baker. 2013. Assessing Activity and Inhibition of Middle East Respiratory Syndrome Coronavirus Papain-Like and 3C-Like Proteases Using Luciferase-Based Biosensors. J Virol 87. 234. O'Doherty U, Swiggard WJ, Malim MH. 2000. Human immunodeficiency virus type 1 spinoculation enhances infection through virus binding. J Virol 74:10074-10080. 235. Helaine S, Thompson JA, Watson KG, Liu M, Boyle C, Holden DW. 2010. Dynamics of intracellular bacterial replication at the single cell level. Proc Natl Acad Sci U S A 107:3746- 3751. 236. Howard AD KM, Thornberry N, Ding GJ, Limjuco G, Weidner J, Salley JP, Hogquist KA, Chaplin DD, Mumford RA, et al. 1991. IL-1-converting enzyme requires aspartic acid residues for processing of the IL-1 beta precursor at two distinct sites and does not cleave 31-kDa IL-1 alpha. J Immunol 147:2964-2969. 237. Kenji Akita TO, Yoshiyuki Nukada, Tadao Tanimoto, Motoshi Namba, Takanori Okura, Rohko Takakura-Yamamoto, Kakuji Torigoe, Yong Gu, Michael S.-S. Su, Mitsukiyo Fujii, Michiyo Satoh-Itoh, Kouzo Yamamoto, Keizo Kohno, Masao Ikeda, and Masashi Kurimoto. 1997. Involvement of Caspase-1 and Caspase-3 in the Production and Processing of Mature Human Interleukin 18 in Monocytic THP.1 Cells. The Journal of Biochemical Chemistry 272:26595-26603. 238. Lamkanfi M, Kanneganti TD. 2010. Caspase-7: a protease involved in apoptosis and inflammation. Int J Biochem Cell Biol 42:21-24. 239. Boucher D, Blais V, Drag M, Denault JB. 2011. Molecular determinants involved in activation of caspase 7. Biosci Rep 31:283-294. 240. Mohamed Lamkanfi T-DK, Petra Van Damme, Tom Vanden Berghe, Isabel Vanoverberghe, Joël Vandekerckhove, Peter Vandenabeele, Kris Gevaert, and Gabriel Núñez. 2008. Targeted Peptidecentric Proteomics Reveals Caspase-7 as a Substrate of the Caspase-1 Inflammasomes Mol Cell Proteomics 7:2350-2363. 241. Shao W, Yeretssian G, Doiron K, Hussain SN, Saleh M. 2007. The caspase-1 digestome identifies the glycolysis pathway as a target during infection and septic shock. J Biol Chem 282:36321-36329. 242. Thomas A Ranol TT, Erin P Peterson, Jennifer Rotonda, Donald W Nicholson, Joseph W Becker, Kevin T Chapman and Nancy A Thornberry. 1997. A combinatorial approach for determining protease specificities: application to interleukin-1B converting enzyme (ICE). Chemistry & Biology 4. 243. Franchi L, Eigenbrod T, Munoz-Planillo R, Ozkurede U, Kim YG, Arindam C, Gale M, Jr., Silverman RH, Colonna M, Akira S, Nunez G. 2014. Cytosolic double-stranded RNA activates the NLRP3 inflammasome via MAVS-induced membrane permeabilization and K+ efflux. J Immunol 193:4214-4222. 244. Franchi L. 2011. Role of inflammasomes in salmonella infection. Front Microbiol 2:8. 245. Sanman LE, Qian Y, Eisele NA, Ng TM, van der Linden WA, Monack DM, Weerapana E, Bogyo M. 2016. Disruption of glycolytic flux is a signal for inflammasome signaling and pyroptotic cell death. Elife 5:e13663. 110

246. Crowley SM, Knodler LA, Vallance BA. 2016. Salmonella and the Inflammasome: Battle for Intracellular Dominance. Curr Top Microbiol Immunol 397:43-67. 247. Broz P, Newton K, Lamkanfi M, Mariathasan S, Dixit VM, Monack DM. 2010. Redundant roles for inflammasome receptors NLRP3 and NLRC4 in host defense against Salmonella. J Exp Med 207:1745-1755. 248. Hrecka K, Hao C, Gierszewska M, Swanson SK, Kesik-Brodacka M, Srivastava S, Florens L, Washburn MP, Skowronski J. 2011. Vpx relieves inhibition of HIV-1 infection of macrophages mediated by the SAMHD1 protein. Nature 474:658-661. 249. Amie SM, Noble E, Kim B. 2013. Intracellular nucleotide levels and the control of retroviral infections. Virology 436:247-254. 250. Liu C. 2013. Strategies for designing transgenic DNA constructs. Methods Mol Biol 1027:183- 201. 251. Rivera-Soto SKPaR. 2015. Viral expression Cassette Elements to Enhance Transgene Target Specificity and Expression in Gene Therapy Discov Med 19:49-57. 252. Abe T, Fujimori T. 2013. Reporter mouse lines for fluorescence imaging. Dev Growth Differ 55:390-405. 253. Dzhagalov IL, Chen KG, Herzmark P, Robey EA. 2013. Elimination of self-reactive T cells in the thymus: a timeline for negative selection. PLoS Biol 11:e1001566. 254. Elliott MR, Ravichandran KS. 2016. The Dynamics of Apoptotic Cell Clearance. Dev Cell 38:147-160. 255. Martinon F, Petrilli V, Mayor A, Tardivel A, Tschopp J. 2006. Gout-associated uric acid crystals activate the NALP3 inflammasome. Nature 440:237-241. 256. Piancone F, Saresella M, Marventano I, La Rosa F, Santangelo MA, Caputo D, Mendozzi L, Rovaris M, Clerici M. 2018. Monosodium Urate Crystals Activate the Inflammasome in Primary Progressive Multiple Sclerosis. Front Immunol 9:983. 257. Kankkunen P, Teirila L, Rintahaka J, Alenius H, Wolff H, Matikainen S. 2010. (1,3)-beta- glucans activate both dectin-1 and NLRP3 inflammasome in human macrophages. J Immunol 184:6335-6342. 258. Paik W, Alonzo F, 3rd, Knight KL. 2019. Probiotic Exopolysaccharide Protects against Systemic Staphylococcus aureus Infection, Inducing Dual-Functioning Macrophages That Restrict Bacterial Growth and Limit Inflammation. Infect Immun 87. 259. Grayczyk JP, Harvey CJ, Laczkovich I, Alonzo F, 3rd. 2017. A Lipoylated Metabolic Protein Released by Staphylococcus aureus Suppresses Macrophage Activation. Cell Host Microbe 22:678-687 e679. 260. Alonzo F, 3rd, Benson MA, Chen J, Novick RP, Shopsin B, Torres VJ. 2012. Staphylococcus aureus leucocidin ED contributes to systemic infection by targeting neutrophils and promoting bacterial growth in vivo. Mol Microbiol 83:423-435. 261. Abigail R. Cannon PVK, Adrienne N. Cobb, Xianzhong Ding, Anai N. Kothari, Paul C. Kuo, Joshua M. Eberhardt, Adam M. Hammer, Niya L. Morris, Xiaoling Li, Mashkoor A. Choudhry. 2018. Alcohol Enhances Symptoms and Propensity for Infection in Inflammatory Bowel Disease Patients and a Murine Model of DSS-Induced Colitis. J Leukoc Biol 104:543-555. 262. Prevention CfDCa. 2011. Staphyloccocus aureus in Healthcare Settings. 263. Powers ME, Bubeck Wardenburg J. 2014. Igniting the fire: Staphylococcus aureus virulence factors in the pathogenesis of sepsis. PLoS Pathog 10:e1003871. 264. Alonzo F, 3rd, Torres VJ. 2013. Bacterial survival amidst an immune onslaught: the contribution of the Staphylococcus aureus leukotoxins. PLoS Pathog 9:e1003143. 265. Miller LS CJ. 2011. Immunity against Staphylococcus aureus cutaneous infections. Nat Rev Immunol 11:505-518. 266. Melehani JH, Duncan JA. 2016. Inflammasome Activation Can Mediate Tissue-Specific Pathogenesis or Protection in Staphylococcus aureus Infection. Curr Top Microbiol Immunol 397:257-282. 111

267. Dirk Holzinger LG, Vijayashree Mysore, Nadine Nippe, Debra J. Taxman, Joseph A. Duncan, Peter M. Broglie Kristina Marketon,Judith Austermann,Thomas Vogl,Dirk Foell,Silke Niemann,Georg Peters, Johannes Roth,and Bettina Löffler. 2012. Staphylococcus aureus Panton-Valentine leukocidin induces an inflammatory response in human phagocytes via the NLRP3 inflammasome. J Leukoc Biol 92:1069-1081. 268. Craven RR, Gao X, Allen IC, Gris D, Bubeck Wardenburg J, McElvania-Tekippe E, Ting JP, Duncan JA. 2009. Staphylococcus aureus alpha-hemolysin activates the NLRP3- inflammasome in human and mouse monocytic cells. PLoS One 4:e7446. 269. Kebaier C, Chamberland RR, Allen IC, Gao X, Broglie PM, Hall JD, Jania C, Doerschuk CM, Tilley SL, Duncan JA. 2012. Staphylococcus aureus alpha-hemolysin mediates virulence in a murine model of severe pneumonia through activation of the NLRP3 inflammasome. J Infect Dis 205:807-817. 270. Munoz-Planillo R, Franchi L, Miller LS, Nunez G. 2009. A critical role for hemolysins and bacterial lipoproteins in Staphylococcus aureus-induced activation of the Nlrp3 inflammasome. J Immunol 183:3942-3948. 271. Mariathasan S, Weiss DS, Newton K, McBride J, O'Rourke K, Roose-Girma M, Lee WP, Weinrauch Y, Monack DM, Dixit VM. 2006. Cryopyrin activates the inflammasome in response to toxins and ATP. Nature 440:228-232. 272. Leech JM, Lacey KA, Mulcahy ME, Medina E, McLoughlin RM. 2017. IL-10 Plays Opposing Roles during Staphylococcus aureus Systemic and Localized Infections. J Immunol 198:2352-2365. 273. Blazejewski AJ, Thiemann S, Schenk A, Pils MC, Galvez EJC, Roy U, Heise U, de Zoete MR, Flavell RA, Strowig T. 2017. Microbiota Normalization Reveals that Canonical Caspase-1 Activation Exacerbates Chemically Induced Intestinal Inflammation. Cell Rep 19:2319-2330. 274. Siegmund B, Lehr HA, Fantuzzi G, Dinarello CA. 2001. IL-1 beta -converting enzyme (caspase-1) in intestinal inflammation. Proc Natl Acad Sci U S A 98:13249-13254. 275. Gagliani N, Palm NW, de Zoete MR, Flavell RA. 2014. Inflammasomes and intestinal homeostasis: regulating and connecting infection, inflammation and the microbiota. Int Immunol 26:495-499. 276. Zhao J, Wang H, Dai C, Wang H, Zhang H, Huang Y, Wang S, Gaskin F, Yang N, Fu SM. 2013. P2X7 blockade attenuates murine lupus nephritis by inhibiting activation of the NLRP3/ASC/caspase 1 pathway. Arthritis Rheum 65:3176-3185. 277. Zhao C, Gu Y, Zeng X, Wang J. 2018. NLRP3 inflammasome regulates Th17 differentiation in rheumatoid arthritis. Clin Immunol 197:154-160. 278. Zeiser R, Blazar BR. 2017. Acute Graft-versus-Host Disease - Biologic Process, Prevention, and Therapy. N Engl J Med 377:2167-2179. 279. Malard F, Gaugler B, Lamarthee B, Mohty M. 2016. Translational opportunities for targeting the Th17 axis in acute graft-vs.-host disease. Mucosal Immunol 9:299-308. 280. Holtan SG, Pasquini M, Weisdorf DJ. 2014. Acute graft-versus-host disease: a bench-to- bedside update. Blood 124:363-373. 281. Edinger M, Hoffmann P, Ermann J, Drago K, Fathman CG, Strober S, Negrin RS. 2003. CD4+CD25+ regulatory T cells preserve graft-versus-tumor activity while inhibiting graft- versus-host disease after bone marrow transplantation. Nat Med 9:1144-1150. 282. Schroeder MA, Ashami K, Staser K. 2018. Modeling Chronic Graft Versus Host Disease in Mice Using Allogeneic Bone Marrow and Splenocyte Transfer. Curr Protoc Pharmacol 83:e47. 283. Cohen JL, Boyer O, Salomon B, Onclercq R, Charlotte F, Bruel S, Boisserie G, Klatzmann D. 1997. Prevention of graft-versus-host disease in mice using a suicide gene expressed in T lymphocytes. Blood 89:4636-4645. 284. Paynich ML, Jones-Burrage SE, Knight KL. 2017. Exopolysaccharide from Bacillus subtilis Induces Anti-Inflammatory M2 Macrophages That Prevent T Cell-Mediated Disease. J Immunol 198:2689-2698. 112

285. Jones SE, Knight KL. 2012. Bacillus subtilis-mediated protection from Citrobacter rodentium- associated enteric disease requires espH and functional flagella. Infect Immun 80:710-719. 286. Swartzendruber JA, Incrocci RW, Wolf SA, Jung A, Knight KL. 2019. Bacillus subtilis exopolysaccharide prevents allergic eosinophilia. Allergy 74:819-821. 287. Omoto Y, Yamanaka K, Tokime K, Kitano S, Kakeda M, Akeda T, Kurokawa I, Gabazza EC, Tsutsui H, Katayama N, Yamanishi K, Nakanishi K, Mizutani H. 2010. Granzyme B is a novel interleukin-18 converting enzyme. J Dermatol Sci 59:129-135. 288. Sarhan J, Liu BC, Muendlein HI, Li P, Nilson R, Tang AY, Rongvaux A, Bunnell SC, Shao F, Green DR, Poltorak A. 2018. Caspase-8 induces cleavage of gasdermin D to elicit pyroptosis during Yersinia infection. Proc Natl Acad Sci U S A 115:E10888-E10897. 289. Orning P, Weng D, Starheim K, Ratner D, Best Z, Lee B, Brooks A, Xia S, Wu H, Kelliher MA, Berger SB, Gough PJ, Bertin J, Proulx MM, Goguen JD, Kayagaki N, Fitzgerald KA, Lien E. 2018. Pathogen blockade of TAK1 triggers caspase-8-dependent cleavage of gasdermin D and cell death. Science 362:1064-1069. 290. Chen KW, Demarco B, Heilig R, Shkarina K, Boettcher A, Farady CJ, Pelczar P, Broz P. 2019. Extrinsic and intrinsic apoptosis activate pannexin-1 to drive NLRP3 inflammasome assembly. EMBO J 38. 291. Creagh EM. 2014. Caspase crosstalk: integration of apoptotic and innate immune signalling pathways. Trends Immunol 35:631-640. 292. Man SM, Kanneganti TD. 2016. Converging roles of caspases in inflammasome activation, cell death and innate immunity. Nat Rev Immunol 16:7-21. 293. Taabazuing CY, Okondo MC, Bachovchin DA. 2017. Pyroptosis and Apoptosis Pathways Engage in Bidirectional Crosstalk in Monocytes and Macrophages. Cell Chem Biol 24:507-514 e504. 294. Tsuchiya K, Nakajima S, Hosojima S, Thi Nguyen D, Hattori T, Manh Le T, Hori O, Mahib MR, Yamaguchi Y, Miura M, Kinoshita T, Kushiyama H, Sakurai M, Shiroishi T, Suda T. 2019. Caspase-1 initiates apoptosis in the absence of gasdermin D. Nat Commun 10:2091. 295. Ignowski JM, Schaffer DV. 2004. Kinetic analysis and modeling of firefly luciferase as a quantitative reporter gene in live mammalian cells. Biotechnol Bioeng 86:827-834. 296. Troy T, Jekic-McMullen D, Sambucetti L, Rice B. 2004. Quantitative comparison of the sensitivity of detection of fluorescent and bioluminescent reporters in animal models. Mol Imaging 3:9-23. 297. Han Y, Zhao T, Cheng X, Zhao M, Gong SH, Zhao YQ, Wu HT, Fan M, Zhu LL. 2018. Cortical Inflammation is Increased in a DSS-Induced Colitis Mouse Model. Neurosci Bull 34:1058-1066.

VITA

Sarah Talley was born in New Orleans, Louisiana to Pat and Sharon Talley. She attended Rhodes

College in Memphis, TN, where she earned a Bachelor of Science degree in Chemistry in 2012. After graduation, Sarah worked as a research technician at St. Jude Children’s Research hospital in the

Infectious Disease department, where she worked on a study investigating respiratory infections in patients with ALL and a number of clinical trials for adolescents infected with HIV. She spent the rest of the year volunteering in at an Infectious Disease clinic in Arusha, Tanzania, before entering the INDII MS program at Loyola University. In 2014, Sarah joined the laboratory of Dr. Edward Campbell, where she studied HIV latency, viral entry and restriction. Sarah matriculated into the PhD program in 2015 and joined the Integrative Cell Biology program.

Sarah’s dissertation work focused on developing a reporter mouse model to study inflammation in animals. This research was supported by the T32 Immunology Training Grant awarded to Dr.

Katherine Knight and an Arthur J. Schmitt Fellowship in Leadership and Service. After completion of her doctoral work, Sarah plans to continue her research as a Postdoctoral Fellow.

113