Bacteria Fighting Back – How Pathogens Target and Subvert the Host Innate Immune System

The First Line of Defense: Bacterial Subversion of Antimicrobial Peptides

While recent work on innate immunity has been primarily focused on elucidating signaling systems down stream of PRRs, antimicrobial peptides (AMPs) are one of the most ancient forms of protection against bacterial infection. For example, epithelial cells of the skin and lung, and Paneth cells in the gut, secrete defensins, gramicidins, and cathelicidins as a tissue protective mechanism against the continuous challenges of bacteria and fungi. In addition, circulating neutrophils also secrete AMPs at sites of acute microbial infection. Whether in the lung, intestine, or at the surface of the skin, most AMPs disrupt membrane integrity through pore formation, causing an efflux of essential molecules and ions from invading pathogens. Importantly however, we now recognize that bacterial pathogens have evolved specific strategies to avoid or abrogate the lytic activity of AMPs during infection.

Perhaps the simplest mechanism bacteria use to combat the anti-microbial activity of AMPs is by physical avoidance through electrostatic repulsion. For example, recent studies suggest that the multiple peptide resistance factor protein, or MprF, provides Staphylococcus aureus with resistance to defensin-like cationic peptides (Figure 1) (Ernst et al, 2009). Here, the hydrophilic C-terminal domain of MprF is responsible for adding an L-Lysine to the phosphatidylglycerol lipid while in the inner leaflet of the bacterial membrane. Afterward, the membrane-bound N-terminal portion of MprF acts to flip the lipid to the outer leaflet so that the positively charged lysine residue can act as an electrostatic buffer that provides limited resistance to the cationic defensin molecules, primarily secreted by neutrophils, present in the host lumen.

In addition to this cloaking mechanism, many bacteria secrete proteases such as the caseinolytic enzymes that can eliminate soluble AMPs. One such protease, ClpXP, and more specifically, the ATPase subunit ClpX from Bacillus anthracis, was shown to have extracellular protease activity toward cathelicidin and alpha-defensin antimicrobial peptides (Figure 1) (McGillivray et al, 2009). Functional ClpX was essential for B. anthracis to disrupt host innate immune clearance mechanisms. In a similar manner, the M1 protein from S. pyogenes was found to contribute to overall resistance of Group A Streptococcus to the cathelicidin AMP LL-37 (Lauth et al, 2009). Interestingly, instead of repelling LL-37, the M1 protein was shown to bind and sequester the AMP, thus rendering it harmless against the invading pathogen (Figure 1). Importantly, these recent examples indicate that AMP inhibition is a critical anti-innate immune mechanism employed by several pathogens, and that this area of investigation is ripe for new discovery.

Stealth and Evasion: Bacterial Pathogens Disguise Their PAMPs

If a bacterial pathogen is able to successfully evade destruction by antimicrobial peptides, most host organisms have evolved a second line of defense centered on microbial recognition of PAMPs by PRRs and the subsequent production of cell-intrinsic immune mechanisms and/or recruitment of immune cells. In response to these challenges, many bacterial pathogens have modified the molecular structure of their PAMPs, thereby avoiding immune detection through stealth and evasion. For example, lipopolysaccharide (LPS) is a ubiquitous component of Gram-negative bacterial cell wall, and is composed of diverse O-antigen side chains that are anchored to the outer leaflet of the bacterial envelope by Lipid A. Importantly, Lipid A is directly recognized by the mammalian TLR4-MD2-CD14 PRR complex to activate innate immune signaling pathways (Figure 1). Some bacterial species have evolved another form of LPS that has undergone modification of the fatty acid side chain of Lipid A from 12-16 to over 20 carbons in length, thereby augmenting its detectability by TLRs (Matsuura, 2013). Yersinia pestis, the causative agent of the Bubonic plague of the 14^th century, possess the ability to reduce the acylation status of Lipid A from hexa-acylated to tetra-acylated in response to an increase in temperature from 21°C (flea temperature) to 37°C (human temperature) (Montminy et al, 2006; Rebeil et al, 2004). This temperature-dependent hypo-acylation activity affords Yersinia the ability to evade host defenses by stealth, as tetra-acylated Lipid A is significantly less recognizable than hexa-acylated LPS by TLR4. Similarly, Shigella that have been internalized and multiply within the host cytoplasm hypo-acylate Lipid A such that upon escape from infected cells, these pathogens possessing the poorly immunogenic form do not alert the innate immune system, and thus stealthily spread and disseminate within the gut microenvironment (Paciello et al, 2013).

As with the example of Lipid A, we now recognize that innate immune evasion strategies may depend on molecular adaptation of highly conserved bacterial cellular components. However, it is also clear that virulence factors, and particularly molecules that directly target host cell physiological systems, are detected by the innate immune system. As discussed in more detail below, many Gram-negative bacterial pathogens utilize protein secretory systems (e.g. Type III/IV/VI secretion systems) to deliver virulence factors, termed ‘effector’ proteins, directly into the host cell. Several groups have now demonstrated that the Type III translocon is a pro-inflammatory PAMP, capable of being recognized by both the NLRP3 and NLRC4 inflammasome (Brodsky et al, 2010; Miao et al, 2010; Zhao et al, 2011). Because bacterial secretory systems are among the most important virulence determinants, it is also reasonable to suspect that pathogens have evolved mechanisms to block immune recognition of these essential systems. Indeed, it was recently discovered that Yersinia spp. masks pore forming attachment of the Type III translocon to the host cell, thereby cloaking the major virulence system of Yersinia from immune recognition (Brodsky et al, 2010). After being secreted by the Type III needle apparatus, Brodsky et al. demonstrated that YopK interacts with the distal end of the Type III translocon to prevent host cellular recognition and inflammasome activation (Figure 1). This is a unique strategy, as it is normally thought that Type III effector proteins have specific eukaryotic target(s). It will be interesting to determine if YopK alters the structure of the needle apparatus directly, or alternatively, competitively inhibits inflammasome activation. In either case however, YopK targets and cloaks a newly recognized PAMP, highlighting the evolutionary arms race that takes place between host and pathogen. Surely as new host immune strategies are uncovered, we expect to find additional mechanisms of masking bacterial PAMPs as immune evasion strategies.

The Inside Story: Bacterial Pathogens Target Intracellular Signal Transduction Cascades

While our discussion has so far focused on initial points of interaction between bacterial pathogens and host cells, its is clear that the power of the immune systems is based on the complex and highly dynamic cellular signal transduction systems the relay information from PRRs to nuclear transcription of pro-inflammatory modulators. Thus, it stands to reason that signal transduction is a high value target of bacterial pathogens. Indeed, as the field of innate immunity has exploded over the past 15 years, the molecular details by which pathogens modulate host immune response systems have also come to light. A majority of our understanding has come from the various ways in which pathogens target the Mitogen Activated Protein Kinase (MAPK) signaling axis and the NEMO-IKK-NF-κB pathway, which we will explore in more detail here.

Effector Molecules Target the MAPK Cascade

As outlined in Figure 1, the MAPK signaling cascade involves sequential activation and amplification of downstream kinases that govern a bevy of transcriptional responses required for proper cellular metabolism, growth, and division. While MAPK signaling is essential for normal operation of the eukaryotic cell, some branches are critical for decoding information obtained from PRRs, and transmitting the appropriate response signal to induce inflammation as an antimicrobial defense mechanism. Because the MAPK pathway response to PAMPs or DAMPs is carefully tuned at multiple sites, this cascade represents a major pressure point that has been successfully exploited by bacterial toxins and effector proteins.

One of the earliest reported direct inhibition of MAPKs by a bacterial pathogen was the discovery that Anthrax lethal toxin (lethal factor; LF) of Bacillus anthracis cleaves the amino terminus of MAPKK1 and MAPKK2 (Figure 1) (Duesbery et al, 1998). This seminal finding was shortly followed by the discovery that a Type III secreted effector protein, YopJ of Yersinia pestis (YopP in Y. enterocolotica and YopY in Y. pseudotuberculosis, hereafter referred to as YopJ), also inhibits the activation of MAPK signaling pathways (Orth et al, 1999). Although some discrepancies exist as to the exact enzymatic activity of YopJ and its homologs, it is clear that several isoforms, including YopJ itself, functions as an acetyl transferase that covalently modifies key serine and threonine residues of MAP kinases (Mukherjee et al, 2006). This acetylation reaction results in the inability of upstream kinases to trans-phosphorylate the down stream kinase, thus preventing the MAPK-mediated cytokine transcriptional response to Yersinia infection (Mukherjee et al, 2006; Mukherjee et al, 2008). Importantly, there is now good evidence that YopJ targets several MAP kinases, including MAPK 1, 3, 4, and 5 and TGF-β-activated kinase 1 (TAK1) (Figure 1) (Mukherjee et al, 2006; Paquette et al, 2012). It is also important to note that this general catalytic mechanism has been evolved for specific pathogen purposes. Recently, the YopJ homologue from Salmonella, AvrA, was found to harbor acetyl transferase activity and to inhibit MAPK4 and MAPK7 of the c-Jun NH2-terminal kinase (JNK) signaling pathway (Figure 1), thereby altering the transcription of pro survival genes during infection (Jones et al, 2008; Wu et al, 2012).

Historically, the identification and molecular characterization of LF and YopJ ushered in a new era of anti-innate immune signaling research. From numerous studies over the past 5-7 years, it is now clear that the irreversible modification of host signaling enzymes represents an effective and widely used strategy for manipulating the immune response system. Because the post-translational modifications of host enzymes are unique to a given pathogen, and often represent completely new catalytic mechanisms, they have been difficult to discover. However, the widespread development of mass spectrometry as tool for biological discovery has aided in this effort. For example, the Shigella flexneri Type III effector protein OspF (which is homologous to the Salmonella SpvC and Pseudomonas syringae HopAI1) was initially described to possess phosphatase-like activity toward activated MAPK. Subsequent mass spectrometry analysis shed light on this reaction, as it was discovered that OspF catalyzes a β-elimination reaction of phosphothreonine to a β-methyldehydroalanine, thus rendering the MAP kinase cascade permanently inactive (Figure 1) (Li et al, 2007; Zhu et al, 2007). Phosphorylation and dephosphorylation are pervasive signaling mechanisms in biology, and the phosphothreonine lyase activity of the OspF family represents the only known enzyme class capable of carrying out this permanent and irreversible modification. Therefore, OspF defines a novel phosphothreonine lyase capable of permanently dephosphorylating host MAPK enzymes. Perhaps equally striking is that OspF, and its homologues in Salmonella and Pseudomonas, take advantage of the a conserved MAPK docking motif (D motif) in mammals in order to specifically recognize and target host substrates for covalent modification (Wei et al, 2012). This level of target specificity by a class of bacterial effector proteins demonstrates the elegance by which pathogens inhibit inflammatory activation.

Effector Molecules Target the NF-κB Pathway

The canonical NF-κB pathway has been the focus of intense investigation for nearly three decades. Significant progress has been made, and although full understanding of particular nuances remains, the function of NF-κB and its transcriptional regulation of immunomodulatory genes have been determined with great detail. In general, the NF-κB pathway involves recognition of PAMPs via TLRs (and other PRRs) and subsequent signal transduction through TIRAP and MyD88 that activates the kinase activity of the NEMO/IKKα/IKKβ complex. This, in turn, leads to the phosphorylation and subsequent ubiquitinylation of IκBα, freeing the NF-κB components (represented in Figure 1 as p65 and p50) to migrate to the nucleus and begin transcription of the various NF-κB-regulated genes. Because this summary does not justify the amazing complexity of this system, we recommend several excellent reviews that provide in-depth perspective on this critical pathway (Dev et al, 2011; Hayden & Ghosh, 2008; Kovalenko & Wallach, 2006). Importantly, the NF-κB signal transduction pathway is a target-rich environment for many pathogens that successfully invade, colonize and disseminate within the host organism.

In addition to manipulating the innate immune responses of epithelial cells, bacterial pathogen inhibition of NF-κB signaling pathways greatly reduces macrophages’ ability to induce inflammation, undergo apoptosis, and communicate with other immune cells. One of the best characterized pathogens, Shigella flexneri, harbors several unique mechanism for NF-κB signaling inhibition which, together, allow them to propagate within macrophages and to reduce acute inflammation during invasion. This is accomplished in part by the Type III effector OspI, which suppresses the tumor-necrosis factor receptor-associated factor 6 (TRAF6)-mediated signaling pathway (Sanada et al, 2012). TRAF6 auto-polyubiquitinylation, subsequent to the activity of an E2 ligase, is responsible for initiating a cascade that culminates in the NF-κB dependent transcription of pro-inflammatory genes. OspI was found to deamidate glutamine-100 of the E2 ubiquitin ligase UBC13 to glutamic acid thus rendering it inactive, which prohibits TRAF6 auto-polyubiquitinylation. This ingeniously simple modification leads to a reduction of the NF-κB proinflammatory pathway to prolonged infection by Shigella (Figure 1).

Manipulation of ubiquitinylation pathways within the host is not an isolated mechanism of NF-κB subversion. Shigella also secretes IpaH9.8, which functions as an E3 ubiquitin ligase to dampen the host immune response by targeting the NEMO complex for proteasomal degradation (Ashida et al, 2010). As mentioned previously, NEMO is critical for downstream activation of NF-κB. IpaH9.8 activity induces Lys27-mediated polyubiquitylation of NEMO, which in turn promotes its proteasome-dependent degradation, and results in reduced NF-κB activation during infection (Figure 1). The IpaH family is of particular interest since many bacterial species that target vertebrate and plant hosts contain proteins that share sequence similarity within this family (Rohde et al, 2007).

Clearly, ubiquitinylation represents an important signaling and regulatory mechanism of NF-κB signaling, and it is therefore no surprise that pathogens have developed multiple mechanisms to exploit this post-translation modification. The Shigella effector OspG was shown to bind ubiquitin and ubiquitinylated E2s to reduce E3-mediated ubiquitinylation of IκBα, thereby preventing the dissociation of IκBα/p50/RelA and inhibiting NF-κB-mediated transcription (Figure 1) (Kim et al, 2005; Zhou et al, 2013). Interestingly, OspG possess kinase activity in addition to that of ubiquitin binding activity (Kim et al, 2005). Recent studies indicate that the binding to ubiquitin activates kinase activity through an unusual allosteric mechanism involving and E2~Ub conjugate (Kim et al, 2005; Pruneda et al, 2014; Zhou et al, 2013). In fact, the crystal structure of OspG bound to UbcH5c~Ub revealed that OspG binds to the ubiquitinylated form of UbcH5c and that this binding was sufficient to reorganize the kinase active site into a catalytically competent state. Because these studies provide significant insight into the kinase regulatory mechanism at the molecular level, we are now poised to discover the host substrate of the kinase-active form of OspG that regulates NF-κB signal transduction.

While Shigella has evolved several unique mechanisms to dampen the host innate immune response through NF-κB signaling inhibition, extracellular pathogens such as members of the Attaching and Effacing (A/E) pathogen group that include enteropathogenic E. coli (EPEC), enterohemorrhagic E. coli O157:H7 (EHEC), and Citrobacter rodentium, inhibit innate immune responses while maintaining intimate contact with the host plasma membrane. One of the first Type III effector proteins discovered in A/E pathogens was the Translocated Intimin Receptor (Tir), which integrates into the host plasma membrane and binds directly to the bacterial cell surface protein intimin (Kenny et al, 1997). Although Tir is essential for the strong bacterial attachment to host cells, and is a key molecule involved in the attaching and effacing lesion observed during infection, it has since been discovered to suppress innate immune signaling pathways. Yan and colleagues demonstrated that Tir shares sequence similarities with the immunoreceptor tyrosine-based inhibition motifs (ITIMs), conserved sequences often found in the cytoplasmic tails of inhibitory receptors of the immune system (Yan et al, 2012). After secretion into the host, ITIM-like motifs present in the Tir effector promote recruitment of Src homology region 2 domain-containing phosphatase 1 (SHP-1), which is known to have a suppressive role in cellular immune responses. SHP-1 has been demonstrated to inhibit the activation of NF-κB and MAPK by directly dephosphorylating them (Nandan et al, 1999), and when in complex with TRAF6, prevents its polyubiquitinylation and subsequent activation of TAK1, thereby further inhibiting the MAPK and NF-κB pathways (Zhang et al, 2003). Here, Tir directly associates with SHP-1 and nucleates a Tir-SHP-1-TRAF6 complex to suppress host cell innate immune responses to A/E lesions, thus providing a striking example of a bacterial effector protein that mimics an endogenous innate immune regulatory mechanism (Figure 1).

Beyond Tir, A/E pathogens have evolved an arsenal of effector proteins that target the NF-κB pathway. NleB, which was first reported to suppress p65 nuclear localization (Newton et al, 2010), was subsequently shown to reduce TNFα-dependent polyubiquitinylation of TRAF2 and stabilize IκB in response to TNFα stimulation. Gao and colleagues first identified Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as an interacting partner of NleB, although the affinity was somewhat modest (K_d of 36 μM). Nonetheless, this interaction disrupted the GAPDH-TRAF2 interaction, and thus suppressed TRAF2-mediated polyubiquitinylation and subsequent NF-κB activation (Gao et al, 2013). More recently, two back-to-back reports indicated that NleB also targets the death domain (DD) containing adaptor protein TRADD in order to inhibit NF-κB activation (Li et al, 2013; Pearson et al, 2013). TRADD was identified as a host interacting partner of NleB, and further analysis showed that NleB was capable of disrupting TRADD death domain oligomerization. Mass spectrometry analysis revealed that NleB catalyzes the GlcNacylation of the TRADD DD at Arginine 235, and this post-translational modification was responsible for inhibiting functional oligomerization of the DDs necessary for the host cell to mount an apoptotic response. This was surprising, since most known GlcNacylation occurs on serine or threonine residues (Hurtado-Guerrero et al, 2008). In addition, TRADD was not the only NleB target identified in these studies, as this bacterial effector is capable of targeting FAS-associated protein with death domain (FADD) and the receptor-interacting serine/threonine-protein kinase 1 (RIPK1) proteins. GlcNacylation of TRADD and FADD arginines prevented the maturation of Caspase-8, a necessary component of the host apoptotic pathway, thus affording A/E pathogens the ability to thwart innate immune defense mechanisms of host cells.

While NleB functions upstream of NF-κB and modulates death receptor signaling pathways, EPEC secretes another effector protein, NleC, which possesses zinc protease activity, to delay host immune detection (Figure 1) (Yen et al, 2010). NleC catalyzed the cleavage of p65 between the 10^th and 11^th amino acids and therefore rendered it unable to partner with p50, and was subsequently degraded through normal host clearance mechanisms. As a result, NleC efficiently suppresses the host pro-inflammatory response to infection and provides A/E bacterial pathogens additional time and space to multiply. In addition to proteolytic inhibition, A/E pathogens also block nuclear trafficking of NF-κB as innate immunity inhibitory mechanism. The EPEC effector NleE, which shares significant sequence and functional homology to Shigella OspZ, plays a key role in this reduction of host immune response. Upon transfection of cells with NleE and subsequent stimulation by IL-1β or TNF-α, the variable NF-κB components p65 and c-Rel were prevented from translocating into the nucleus, and release of IL-8 was greatly reduced as a result (Figure 1) (Newton et al, 2010). Therefore, it seems that NleE is specific for arresting the traffic of p65 and c-Rel, as the nuclear translocation of p50 was still permitted (Newton et al, 2010). Interestingly, it was recently shown that NleE possesses S-adenosyl-L-methionine-dependent methyltransferase activity against TAB2 and TAB3, ubiquitin-chain sensory proteins involved in the NF-κB pathway. NleE specifically modified the zinc-coordination site in the zinc-finger domains of TAB2 and TAB3, causing these proteins to lose the zinc ion as well as ubiquitin-chain binding ability (Zhang et al, 2012). Disruption of the ubiquitin-chain binding by TAB2 and TAB3 results in their inability to activate TAK1 and IKK, thereby preventing the requisite NF-κB components from being released to translocate to the nucleus and mount an inflammatory response (Figure 1).

Because direct molecular mechanisms have been elucidated in detail, our analysis of the NF-κB pathway has largely focused on the Shigella and A/E pathogen group. Importantly however, NF-κB is certainly targeted by many pathogens, alternative mechanisms of signaling inhibition are likely to have evolved, and we suspect that many new and exciting details will emerge in the coming years.

Organelle Dysfunction: The Host Secretory Pathway as an Immune Modulatory Target of Bacterial Pathogens

As we have described so far, bacterial pathogens harbor a variety of elegant mechanisms to inhibit cellular signal transduction pathways that dampen the host inflammatory response. However, recent evidence reveals that pathogens have evolved alternative strategies to inhibit innate immunity by blocking the secretion of cytokines and trafficking of cell surface receptors. Here we will discuss the recent advancements in understanding how bacterial effectors target, engage, and inhibit molecules that regulate the host general secretory pathway (GSP).

Initiating at the Endoplasmic reticulum (ER), the GSP traffics molecular cargo (e.g. cell surface receptors and cytokines) through a series of membrane-bound compartments, including ER-Golgi intermediate compartment (ERGIC) and the Golgi proper, and releasing this cargo through vesicle fusion with the plasma membrane (Figure 2). Vesicular trafficking between these compartments is supervised in large part by small GTPases including Arf1 and Sar1 that nucleate coat protein complex I and II (COPI and COPII) on vesicles involved in retrograde or anterograde transport, respectively. One of the most prolific GSP inhibitors are the A/E pathogens (EHEC, EPEC, and Citrobacter) that secrete the Type III effector proteins NleA/EspI, NleF, and EspG. These effector proteins specifically target the early steps of ER to Golgi trafficking through distinct molecular mechanisms. The first step of cargo packaging and vesicle budding from the ER is regulated by NleA/EspI, which acts to prevent normal COPII mediated vesicle trafficking, thereby inhibiting ER to Golgi transport (Figure 2) (Kim et al, 2007). Although this mechanism of regulation is not fully understood, it is clear that NleA/EspI binds to Sec23/24, a heterodimeric subunit of the CopII coat, in a Sar1 and lipid-dependent manner and decreases the rate of GTP hydrolysis by Sar1 (Thanabalasuriar et al, 2012). This interaction has been shown to stabilize COPII, which presumably leads to a defect in recycling and thus an abnormal buildup of COPII coated vesicles that slows proper sorting of anterograde trafficked vesicular cargo. In contrast to this mechanism, the GSP is also inhibited by the effector protein NleF which impedes the activity of COPI by binding to the transmembrane protein Tmp21, a COPI docking partner (Figure 2) (Olsen et al, 2013). Blocking COPI type vesicles would inhibit retrograde transport, which is essential for Golgi function. In both cases however, additional biochemical and structural analyses will be needed to fully understand these important host-pathogen interactions at the molecular level.

While originally thought to alter host microtubule dynamics, it was later revealed that EspG acts as a scaffold to bring together GTP-bound Arf GTPases and active p21-associated kinase (PAK) resulting in the subsequent fragmentation of the Golgi apparatus (Figure 2) (Selyunin et al, 2011). Indeed, crystal structures show that EspG binds to Arf6 in the GTP-bound state and protects the bound GTP-binding protein from inactivation by ArfGAP (ASAP1). In addition, EspG directly activates PAK by disrupting the autoinhibitory interactions in this multi-domain kinase. While these studied indicated that EspG is a multifaceted effector molecule capable of bridging critical signaling pathways within cells, more recent studies have provided additional levels of complexity. For example, Dong et al. found that EspG (and its Shigella homolog VirA) function as Rab1-specific GTPase activating protein (GAP) (Dong et al, 2012). Further crystallographic evidence revealed that EspG harbors a catalytic TBC (Tre-2/Bub2/Cdc16) GAP catalytic core, yet the sequence and structural fold of this bacterial GAP domain has almost no similarity to host GAPs that target Rab family GTPases. Perhaps most remarkably, the binding site of Rab1 on EspG perfectly overlaps that of PAK, suggesting that this is a site of molecular evolution toward new effector functions (Selyunin et al, 2014, Cell Reports, in press). In addition, EspG binds and protects activated Arf1 leading to Golgi disruption while at the same time acting as a Rab1GAP thereby greatly reducing ER to Golgi vesicular trafficking (Figure 2) (Selyunin et al, 2014, Cell Reports, in press). Together, these multivalent interactions inhibit cytokine secretion during EPEC infection (Dong et al, 2012)

EHEC/EPEC is not the only pathogen that has evolved elegant mechanisms to inhibit the host secretory system. Shigella flexneri was recently found to use two different strategies for Golgi destruction and GSP inhibition. First, the effector VirA, which is an EspG homolog, was shown to also have TBC-mediated Rab1 GAP activity (Dong et al, 2012). This activity was sufficient to disrupt intra-Golgi vesicular movements and also confer protection against the host autophagic pathway, thus enabling Shigella to promote its intracellular lifestyle (Figure 2). Interestingly however, VirA is sufficient, but not necessary, to disrupt Golgi architecture during bacterial infection. Experiments carried out using a virA strain of Shigella were shown to cause Golgi disruption similar to wild type bacteria (Mounier et al, 2012). Recently, the missing component was identified through a gain-of-function screen for bacterial inhibitors of the GSP. Burnaevskiy et al. found that the Shigella Type III effector IpaJ was both necessary and sufficient to inhibit cytokine secretion and disrupt Golgi morphology during infection (Burnaevskiy et al, 2013). Bioinformatics analysis revealed that IpaJ shared secondary structural similarity with members of the C39-peptidase-like family, and contained the invariant catalytic triad consisting of Cys64, His206, and Asp218. Using a combination of yeast genetics and mass spectrometry, IpaJ was shown to specifically cleave the myristoylated glycine from Arf1, as well as numerous lipid-modified substrates, thus preventing host signal transduction from specific membrane microdomains (Figure 2). Together, these studies reveal an alternative mechanism of innate immune pathway inhibition based on global cellular disruption of host organelle structure and function. It will be important to determine if these bacterial effector proteins also block non-canonical secretion systems (such as the release of IL-1) that are not dependent on the GSP. In addition, further studies will be needed to elucidate the similarities and differences in VirA and IpaJ cellular function, as well as their role in immune function during infection.