Vying for the control of inflammasomes: The cytosolic frontier of enteric bacterial pathogen–host interactions

Abstract Enteric pathogen–host interactions occur at multiple interfaces, including the intestinal epithelium and deeper organs of the immune system. Microbial ligands and activities are detected by host sensors that elicit a range of immune responses. Membrane‐bound toll‐like receptors and cytosolic inflammasome pathways are key signal transducers that trigger the production of pro‐inflammatory molecules, such as cytokines and chemokines, and regulate cell death in response to infection. In recent years, the inflammasomes have emerged as a key frontier in the tussle between bacterial pathogens and the host. Inflammasomes are complexes that activate caspase‐1 and are regulated by related caspases, such as caspase‐11, ‐4, ‐5 and ‐8. Importantly, enteric bacterial pathogens can actively engage or evade inflammasome signalling systems. Extracellular, vacuolar and cytosolic bacteria have developed divergent strategies to subvert inflammasomes. While some pathogens take advantage of inflammasome activation (e.g. Listeria monocytogenes, Helicobacter pylori), others (e.g. E. coli, Salmonella, Shigella, Yersinia sp.) deploy a range of virulence factors, mainly type 3 secretion system effectors, that subvert or inhibit inflammasomes. In this review we focus on inflammasome pathways and their immune functions, and discuss how enteric bacterial pathogens interact with them. These studies have not only shed light on inflammasome‐mediated immunity, but also the exciting area of mammalian cytosolic immune surveillance.


| INTRODUCTION
Acute gastroenteritis is caused by infection of the stomach or intestinal mucosa with enteric pathogens, which are typically transmitted via contaminated food or water. It is characterised by damage to the mucosa and loss of mucosal barrier integrity, leading to malabsorption, diarrhoea and consequent dehydration (DuPont, 2009). Infectious gastroenteritis can affect people of all ages, thus having a substantial economic impact worldwide. In the USA, 179 million people suffer from acute gastroenteritis per year, leading to over 5,000 deaths (DuPont, 2009;Shane et al., 2017); in the UK the estimate is of 17 million cases per year (Tam et al., 2012). The most common causes of bacterial intestinal infections include the Gram-negative pathogens Salmonella enterica, Shigella sp., Escherichia coli and Yersinia and the Gram-positive pathogen Listeria monocytogenes (Kirk et al., 2015;Shane et al., 2017). While its mode of transmission is not currently established, Helicobacter pylori is the most prevalent aetiological agent of bacterial gastritis and is a major risk factor for the development of gastric malignancies (Chey, Leontiadis, Howden, & Moss, 2017). The gastrointestinal immune system, which encompasses both immune and intestinal epithelial cells (IECs) lining the mucosa, must recognise and be activated by pathogenic insults, while remaining anergic to the presence of the endogenous microbiota. One of the mechanisms involved in this distinction is the multiprotein cytosolic complex known as the inflammasome. The inflammasome acts as a molecular platform for caspase-1 activation and has been shown to have an increasingly important role in innate immunity since it was described in 2002 (Martinon, Burns, & Tschopp, 2002). It assembles in response to microbial or danger signals, triggering downstream signalling cascades that give rise to the release of pro-inflammatory factors, including cytokines (e.g. interleukin-1β  and IL-18) and alarmins (such as IL-1α and HMGB1), as well as pyroptotic cell death (Broz & Dixit, 2016;Hayward, Mathur, Ngo, & Man, 2018). In the intestine, inflammasome signalling is functional within myeloid cells, such as macrophages, dendritic cells (DCs) and neutrophils, as well as epithelial cells, pointing to their pivotal role in the early response to pathogens.

| THE INFLAMMASOME COMPONENTS
The assembly of inflammasomes is triggered by the recognition of a signal by a cytosolic sensor, which can be a member of the NLR family (e.g. NLRP3 or NLRC4), an ALR, or the TRIM protein PYRIN (Broz & Dixit, 2016) (Figure 1). The NLRs are further divided based on their Nterminal protein-protein interaction domains, for example NAIPs (also known as NLRBs) contain BIR, NLRCs contain CARD and NLRPs contain pyrin motifs (except NLRP1, which contains a CARD). The pyrin domain is also present in ALRs and PYRIN. The pyrin domain mediates interactions with the adaptor ASC, a small protein that itself consists of a pyrin domain and a CARD, which promotes the recruitment of pro-caspase-1 to oligomerised inflammasomes. This leads to caspase-1 oligomerisation and proximity-induced activation via autoproteolysis ( Figure 1). While non-CARD-containing sensors require ASC to recruit pro-caspase-1, NLRCs can interact via CARD and directly activate full-length pro-caspase-1. This leads to ASC-independent pyroptosis, but ASC is still required for caspase-1 autoproteolysis and cytokine processing (Broz, von Moltke, Jones, Vance, & Monack, 2010). NAIPs can only activate inflammasomes by stimulating NLRC4, which in turn activates pro-caspase-1. Proteolytically activated caspase-1 cleaves pro-IL-1β and pro-IL-18 into their bioactive forms.

| SIGNALS FOR INFLAMMASOME ACTIVATION
It is commonly accepted that inflammasome activation in myeloid cells and the resultant IL-1β release require two signals. The 'priming' signals (signal 1) activate the transcription and/or posttranslational regulation of inflammasome-associated genes/proteins and are best understood for the activation of the NLRP3 inflammasome in macrophages. These include: (i) the transcription of pro-IL-1β, (ii) the transcriptional and translational licensing of NLRP3 via toll-like receptors (TLRs) and NF-κB Fernandes-Alnemri et al., 2013;Lin et al., 2014), (iii) interferondependent up-regulation of murine caspase-11 Benaoudia et al., 2019) and mouse and human GBPs  and (iv) IRF2-driven expression of human caspase-4 (Benaoudia et al., 2019) and mouse GSDMD (Kayagaki et al., 2019).
Importantly, bacterial ligands such as cell wall components and nucleic acids can serve as signal 1, and pathogen-associated virulence factors and/or activities, as discussed below, serve as a second signal specific to the NLR/ALR/PYRIN inflammasome sensors.
Thus, during infection by bacterial pathogens, inflammasome signalling is a fast, inflammatory process that cooperates with other innate immune signalling pathways such as TLRs and interferons.
Cytosolic double-stranded DNA (dsDNA) is recognised by AIM2 via its HIN domain (Hornung et al., 2009). While PYRIN is autoinhibited through interactions with 14-3-3 proteins, it can be activated upon the inactivation of cellular RhoA GTPases by various bacterial toxins (Xu et al., 2014;Masters et al., 2016). The mouse NLRP1B sensor is also basally autoinhibited through its N-terminal regions, which can undergo proteasomal degradation, for example in response to anthrax lethal toxin, leading to the release of the CARD-F I G U R E 1 Inflammasome-forming sensors and their known activators. Inflammasomes are multiprotein complexes that function as platforms to activate caspase-1. Some inflammasome sensors, such as NLRP3, PYRIN and NLRP1B, are activated following perturbations of cellular homeostasis triggered by damage or microbial associated molecular patterns. For example, mitochondrial or lysosomal disruption will lead to NLRP3 activation, while inhibition of host Rho-GTPases will allow PYRIN inflammasome assembly and degradation of the NLRP1B N-terminal will lead to nucleation of the free CARD-containing NLRP1B C-terminus. Other inflammasome sensors, exemplified by AIM2, NAIP-NLRC4 and caspase-11 (caspase-4 and 5 in humans), are activated in response to direct detection of their ligands: DNA is recognised by the AIM2 HIN200 domain, NAIP proteins bind flagellin and type 3 secretion system (T3SS) needles and rods, and the caspase-11 CARD domain interacts with LPS. Active caspase-11/4/5 cleaves Gasdermin D (GSDMD), leading to pore formation and subsequent potassium efflux, which can trigger noncanonical activation of the NLRP3 inflammasome, and pyroptosis. NLRP6 functions as a direct sensor of lipoteichoic acid (LTA), but can also be activated by changes in the microbiota and has additionally been shown to perform inflammasome-independent functions. *Human NLRP1 has an N-terminal PYD domain. Domain compositions are colour coded and abbreviated as follows: CARD, caspase-activation and recruitment domain; p20 and p10, large and small catalytic subunits; PYD, pyrin domain; NOD, nucleotide binding and oligomerisation domain; LRR, leucine rich repeat; BIR, baculovirus inhibitor of apoptosis domain; HIN200, haematopoietic expression, interferon inducible, nuclear localised (HIN) DNA binding domain of 200 residues; C-C, coiled-coil; FIIND, function to find domain containing C-terminal fragment that oligomerises to recruit caspase-1 and activate the NLRP1 inflammasome (Chui et al., 2019;Sandstrom et al., 2019;Xu et al., 2019). Other than the inhibitors of the serine proteases DPP8/9 (Okondo et al., 2017;Zhong et al., 2018), physiological activators of human NLRP1 inflammasome remain to be discovered.
Activation of caspase-11 has also been linked to modulation of the actin cytoskeleton, phagosome maturation and leukocyte migration (Li et al., 2007;Akhter et al., 2012), showing that this caspase also has non-inflammatory roles.
Other less well-studied NLRs include NLRP6, NLRP12, NLRC3 and NLRC5. In particular, NLRP6 is important in the gastrointestinal tract as it is expressed in myeloid cells, IECs and goblet cells, where it contributes to intestinal homeostasis (Elinav et al., 2011;Wlodarska et al., 2014). Notably, NLRC3, NLRP6 and NLRP12 are implicated in the suppression of the inflammatory response, and deficiency of these proteins can confer resistance to infection in vivo (Anand et al., 2012;Zaki, Man, Vogel, Lamkanfi, & Kanneganti, 2014;Zhang et al., 2014).
A/E lesions are characterised by effacement of the brush border microvilli and intimate bacterial attachment to the apical membrane of IECs. Intimate attachment is mediated by strong interactions between intimin on the bacterial surface and Tir (translocated intimin receptor), which is injected into IECs by the T3SS (Frankel & Phillips, 2008;Slater, Sågfors, Pollard, Ruano-Gallego, & Frankel, 2018), all of which are encoded within the locus for enterocyte effacement (LEE) common to A/E pathogens (Frankel et al., 1998). Intimin:Tir interactions lead to Tir clustering and actin polymerisation underneath attached bacteria, which facilitates delivery of other T3SS effectors that subvert mammalian cell processes (Shenoy, Furniss, Goddard, & Clements, 2018). Strain to strain variability in both the repertoire of T3SS effectors and other virulence factors affects the outcome of pathogen-host interactions; haemolysins (such as EhxA) and the Shiga toxins (Stx1 and 2) mentioned herein are exclusively expressed by EHEC.
In addition to T3SS effectors, structural T3SS components also provoke inflammasome activation. The inner rod protein EscI of the LEE-encoded T3SS, the inner rod protein EprJ and the needle protein EprI of the E. coli type III secretion system 2 (ETT2) activate murine NLRC4 inflammasomes in vitro when introduced directly into the macrophage cytosol Zhao et al., 2011;Yang et al., 2013;Wu et al., 2019); however, the ETT2 has accumulated considerable mutational attrition and is not believed to form a functional T3SS (Ren et al., 2004). Critically, the EPEC T3SS needle protein EscF is one of the needle proteins that are not detected in human THP-1 macrophages (Yang et al., 2013). Similarly, the EPEC/EHEC flagellin (FliC) is not recognised by murine NAIP5/human NAIP (Zhao et al., 2011), whereas its E. coli K12 counterpart is readily detected . Taken together, murine NLRC4 may detect EPEC/ EHEC expressing the LEE T3SS however, these bacteria likely evade this inflammasome in human macrophages.
Conversely, A/E pathogens encode T3SS effectors which specifically target inflammasome components. NleA (also named EspI), interacts with the pyrin and leucine-rich repeat domains of ubiquitylated NLRP3, impairing its deubiquitylation and effectively inhibiting recruitment of caspase-1 to the inflammasome (Yen, Sugimoto, & Tobe, 2015). Importantly, deletion of the gene encoding NleA/EspI results in severe attenuation of C. rodentium in vivo (Mundy et al., 2004). The T3SS effector NleF inhibits caspase-4, as well as caspase-8 and 9 (Blasche et al., 2013;Pallett et al., 2017). Inhibition of caspase-4 by NleF blocks secretion of processed IL-18 following infection of Caco-2 human IEC-like cells with EPEC at 4 days post-infection with C. rodentium in mice, which results in reduced neutrophil recruitment (Pallett et al., 2017) (Table 1). Notably, inflammasome responses to the colonic, non-invasive, pathogen C. rodentium, differ significantly from those triggered by Salmonella, an invasive pathogen that causes inflammation of the small intestine, as discussed next.  Zychlinsky, 2006). In addition to reduced IL-1β and IL-18 that are required for an optimal Th1 immune response for bacterial clearance, the loss of Casp1 also reduces bacterial uptake by neutrophils. As a result, in Casp1 −/− mice macrophage pyroptosis through caspase-11 leads to higher extracellular bacterial load, resulting in greater susceptibility to infection (Broz et al., 2012). Casp1/Casp11-double knockout mice, whose cells are resistant to pyroptotic lysis, have fewer extracellular bacteria and are therefore less susceptible than Casp1 −/− single knockouts (Broz et al., 2012). These findings point towards a role of inflammasomes and caspase-1/11 in pyroptosis, neutrophil function and adaptive immune responses during S. Typhimurium infection in vivo. Furthermore, the effector SlrP (Salmonella leucine-rich repeat protein), secreted through both T3SSs, inhibits inflammasome activation and IL-1β release in the small intestine; higher IL-1β levels during infection with a ΔslrP strain promote anorexia and increases disease severity (Rao et al., 2017) (Table 1).

| THE INFLAMMASOME AND
Additionally, AIM2 has been reported to have a role in preserving epithelial integrity of the intestinal barrier during Salmonella infection, but it is not clear whether this defence mechanism involves inflammasome or non-inflammasome-dependent functions of AIM2 (Hu et al., 2016). Conversely, NLRP12 plays an undefined role in dampening the immune response to S. Typhimurium, and deficiency of this NLR leads to resistance to infection (Vladimer et al., 2012;Zaki et al., 2014), independently of caspase-1 (Zaki et al., 2014). In summary, tissue-and cell type-specific inflammasome signalling is protective against Salmonella infection.

| SHIGELLA -INFLAMMASOME INTERACTIONS ARE CELL-TYPE SPECIFIC
Infection by Shigella is a leading cause of diarrhoea in children in lowand middle-income countries, where it is associated with high mortality rates (Tickell et al., 2017). induces rapid pyroptosis (Hilbi et al., 1998), whereas in epithelial cells

S. flexneri avoids inflammasome activation and pyroptosis and instead
triggers a delayed calpain-dependent necrotic cell death mediated by the effector VirA (Bergounioux et al., 2012).
Compared to studies with Salmonella, much less is currently known about the roles inflammasomes play during Shigella infection.
S. flexneri LPS can be detected by caspase-4 in vitro and in epithelial cells (Kobayashi et al., 2013;Shi et al., 2014) (Figure 2c; Table 1). To counteract this and prolong epithelial cell survival, S. flexneri delivers the T3SS effector OspC3. OspC3 interacts with the cleaved caspase-4 subunit p19 and inhibits its activation by preventing heterodimerisation of the caspase-4 p19 and p10 subunits (Kobayashi et al., 2013). The importance of OspC3 was confirmed in a synthetic 'bottom-up' approach to identify effectors that inhibit epithelial cell death. This study also identified the effectors OspD2 and IpaH1.4 (Mou, Souter, Du, Reeves, & Lesser, 2018 In macrophages, S. flexneri activates a number of inflammasome pathways leading to rapid cell death. The T3SS needle and rod proteins (MxiH and MxiI) are recognised by human NAIP (and mouse Naip1 and 2 respectively) to induce NLRC4 inflammasome activation (Kofoed & Vance, 2011;Yang et al., 2013;Suzuki, Franchi, et al., 2014;Zhao et al., 2016). Cellular damage resulting from pore formation by the T3SS effector IpaB may also contribute to NLRC4 inflammasome activation (Senerovic et al., 2012). In addition, IpaH7.8 promotes NLRC4 and NLRP3 inflammasome activation by ubiquitinating glomulin (GLMN) and eliciting its degradation . GLMN is an inflammasome repressor which specifically targets cellular inhibitor of apoptosis proteins 1 and 2 (cIAP1 and cIAP2), members of the inhibitor of apoptosis family of RING-E3 ligases, resulting in reduced cIAP E3 ligase activity and consequently diminished cIAP-mediated inflammasome activation.  (Table 1). Notably, conflicting findings have occasionally been reported, which may have arisen from the use of different L. monocytogenes strains, cell-types (monocytes vs. macrophages), species (human vs. mouse) and/or pre-treatment with TLR ligands (e.g. LPS) or type I or type II interferons (IFNα/β or IFNγ respectively).
Here we summarise key findings on inflammasome activation by L. monocytogenes and refer readers to a detailed discussion elsewhere (Theisen & Sauer, 2016).
In mouse macrophages, AIM2 is the major inflammasome sensor L. monocytogenes is a poor activator of NLRC4 (Sauer et al., 2010;Warren et al., 2010;Sauer et al., 2011), but can activate NLRP6, caspase-11 (Hara et al., 2018) andNLRP1B (Neiman-Zenevich et al., 2017) in mouse macrophages, and NLRP3 in mouse bone-marrow derived DCs (Clark, Schmidt, McDermott, & Lenz, 2018). Most L. monocytogenes strains turn down flagellin expression at 37 C and thus evade detection by TLR5 and the NAIP5-NLRC4 pathway (Theisen & Sauer, 2016). However, due to a mutation in MogR (lmo0674), 10403S exhibits low basal flagellin expression which weakly activates the NLRC4 inflammasome (Gründling, Burrack, Bouwer, & Higgins, 2004). L. monocytogenes engineered to overexpress flagellins results in severe attenuation in vivo in an NLRC4-dependent manner and these strains were also found to be poor vaccine candidates (Sauer et al., 2011). Surprisingly, bacterial lipoteichoic acid activates NLRP6 and caspase-11 in mouse macrophages infected with L. monocytogenes strain EGD (Hara et al., 2018) ( Table 1). NLRP6-ASC complexes recruited both caspase-1 and caspase-11, where the role of active caspase-11 was to promote caspase-1 activation, which in turn processed IL-18 and IL-1β cytokines ( Figure 2d). Importantly, in human cells, NLRP6-silencing does not affect caspase-1 activation by L. monocytogenes EGD (Meixenberger et al., 2010), which points towards species-specific differences. Furthermore, genome sequencing studies performed by P. Cossart have confirmed that the EGD strain is markedly different from 10403S and EGDe strains and has a mutation in the master transcriptional regulator PrfA (PrfA*) which results in constitutive expression of various virulence genes (Bécavin et al., 2014). It is plausible that these differences also contribute to strain-specific responses in host cells.
Altogether, these studies are consistent with the recently identified AIM2-independent, NLRP3-dependent detection of cytosolic DNA in human, but not mouse, myeloid cells (Gaidt et al., 2017). This new DNA-sensing pathway in human cells also requires lysosomal damage (Gaidt et al., 2017). Therefore, L. monocytogenes, plausibly via both LLO and release of DNA in the cytosol, activates the human NLRP3 inflammasome and the murine AIM2 inflammasome (Figure 2d).
Early in vivo studies on L. monocytogenes pointed towards a protective role of inflammasome-related genes (Hirsch, Irikura, Paul, & Hirsh, 1996;Labow et al., 1997;Tsuji et al., 2004); however, these studies were carried out in mice with a mixed C57BL/6 and 129/S background which also carry an inactivating passenger mutation in Casp11. More recent studies suggest that loss of Il18, Nlrp3 or Asc increases resistance to L. monocytogenes, and loss of Casp1/11 −/− does not markedly affect survival or adaptive immune responses (Lochner et al., 2008;Sauer et al., 2011;Tsuchiya et al., 2014;Clark et al., 2018). Intriguingly, Nlrp6 −/− and Casp11 −/− mice, which would not recognise Listeria LTA, are also more resistant to L. monocytogenes infection (Hara et al., 2018). However, increased resistance of Nlrp6 −/− mice to L. monocytogenes has also been attributed to increased canonical NF-κB signalling and myeloid cell responses that better restrict bacterial growth (Anand et al., 2012). As inflammasome-driven inflammation increases neutrophil influx, these findings suggest that a heightened neutrophil-response is detrimental in the defence against L. monocytogenes in vivo. Taken together, most in vivo studies in mice suggest that inflammasome activation benefits Listeria and is detrimental to the host.

| HELICOBACTER PYLORI, INFLAMMASOME STIMULATION AND PERSISTENT INFECTION
H. pylori is found in the gastric mucosa of around 50% of the world population but only a minority of infected individuals will develop long-term clinical symptoms (Abadi & Kusters, 2016). Individuals with an active chronic infection can develop gastric ulcers, chronic gastritis, gastric mucosa-associated lymphoid tissue (MALT) lymphoma and gastric cancers (Chey et al., 2017). Inflammasome-mediated recognition of H. pylori in myeloid cells is mainly attributed to NLRP3 through a not well-defined mechanism requiring expression of the bacterial vacuolating cytotoxin A (VacA) and the cytotoxin-associated genes pathogenicity island, which encodes the cytotoxin associated gene A (CagA) and the type IV secretion system (T4SS) (Kim, Park, Franchi, Backert, & Núñez, 2013;Semper et al., 2014;Kameoka et al., 2016). It is tempting to speculate that VacA, which forms pores in the plasma membrane, causes mitochondrial damage, increases ROS levels and induces cell apoptosis, may likewise contribute to NLRP3 activation.
Caspase-1, IL-1β and IL-18 play an important role in H. pylori pathogenesis in vivo, but whether inflammasome activation benefits the host or the pathogen remains to be defined (Hitzler et al., 2012;Kim et al., 2013). The role of NLRP3 in promoting (Hitzler et al., 2012;Koch et al., 2015;Arnold et al., 2017) versus restricting (Semper et al., 2014) H. pylori infection also remains unclear. These differences may be due to the use of mice of different ages, differences in the microbiota or the H. pylori strains, or to variations in the methodologies used in the month-long infection model. However, supporting the role of inflammasomes in promoting H. pylori pathogenesis, the protective role of Mucin1 expression in the gastric mucosa has been linked to the down-regulation of NLRP3 expression and consequent IL-1β release   YopD (Brodsky et al., 2010;Zwack et al., 2015). The PYRIN inflammasome is activated through the RhoA GTPase-inhibiting activity of YopE and YopT (Chung et al., 2016;Ratner et al., 2016) ( Figure 2f). To counteract detection by the inflammasomes, Yersinia encodes the effector YopK, which interacts with the T3SS and regulates translocation of effectors (Brodsky et al., 2010;Zwack et al., 2015), and YopM, a homologue of the The acetyltransferase YopJ (YopP in Y. enterocolitica) is a homologue of Salmonella AvrA that inhibits transforming growth factor beta-activated kinase 1 (TAK1), IκB kinase β (IKKβ) and mitogenactivated protein kinase (MAPK) kinases, inhibiting pro-inflammatory signalling in response to TLR and/or TNF signalling (Peterson et al., 2016;Pinaud et al., 2018). TAK1 inhibition promotes non-canonical RIPK1-FADD-caspase-8-dependent GSDMD cleavage, pore formation and cell death in mouse macrophages. The resulting potassium efflux triggers NLRP3 and caspase-1 activation, and release of IL-1β Orning et al., 2018) (Figure 2f). In addition, blockade of TNF-mediated pro-survival NF-κB and MAPK signalling by YopJ can also lead to RIPK1-dependent apoptosis (Weng et al., 2014;Peterson et al., 2016;Peterson et al., 2017). Further, YopJ-mediated inhibition of the Nod2 pathway has also been linked to Nod2-dependent activation of caspase-1 and IL-1β release from Peyer's patches, and loss of intestinal barrier function in mice (Meinzer et al., 2012). During Y. pseudotuberculosis infection in vivo, YopJ-driven cell death pathways promote bacterial clearance and host survival, seemingly counteracting the pathogen-mediated blockade of inflammatory signalling (Meinzer et al., 2012;Philip et al., 2014;Peterson et al., 2017).
In addition to secreted effectors, Yersinia LPS can also activate inflammasomes. However, Yersinia evades detection by mouse caspase-11 in vivo by deacylating its lipid A to four lipid chains at 37 C (Hagar, Powell, Aachoui, Ernst, & Miao, 2013). It is possible that human caspase-4, which can recognise tetra-acylated LPS from Francisella novicida, might also recognise tetra-acylated Yersinia lipid A . While its role during infection with enteric Yersinia species in vivo is yet to be defined, mouse NLRP12, whose activation mechanisms remain poorly understood, also contributes to IL-1β release from bone marrow-derived macrophages infected with Y. pseudotuberculosis or Y. enterocolitica (Vladimer et al., 2012). Taken together, inflammasome activation during Yersinia infection is beneficial for the host as it promotes bacterial clearance (Brodsky et al., 2010;Meinzer et al., 2012;Philip et al., 2014;Peterson et al., 2017). trafficking systems (Cossart, Boquet, Normark, & Rappuoli, 1996).
Studies from her group and others also ushered the phase of discov-