Signalling and phagocytosis in the orchestration ... - Wiley Online Library

Cellular Microbiology (2007) 9(2), 290–299

doi:10.1111/j.1462-5822.2006.00864.x

Microreview Signalling and phagocytosis in the orchestration of host defence J. Magarian Blander* Center for Immunobiology, Mount Sinai School of Medicine, 1425 Madison Avenue, New York, NY 10029, USA. Summary Dendritic cells (DCs) orchestrate either tolerance or immunity. At the heart of this function lies phagocytosis, which allows DCs to sample the tissue microenvironment and deliver both its self and non-self constituents into endocytic compartments for clearance, degradation and presentation by major histocompatibility complex (MHC) molecules. Depending on the type of signalling pathways triggered during phagocytosis, DCs deliver appropriate signals to T cells that determine either their tolerance or activation and differentiation. Here I draw attention to the ability of DCs to read the contents of their phagosomes depending on the type of compartmentalized signalling pathways engaged during internalization. Toll-like receptors (TLRs) engaged during phagocytosis of microbial pathogens, but not syngeneic apoptotic cells exert phagosome autonomous control on both the kinetics and outcome of phagosome maturation. By bearing the assembly of signalling complexes on their membranes, individual phagosomes undergo separate programmes of maturation irrespective of the activation status of the DC carrying them. Phagosomes carrying microbial cargo are favoured for MHC class II presentation precluding phagosomes carrying self from contributing to the first signal delivered to T cells – the peptide–MHC complex. This mechanism prevents the potential presentation of peptides derived from self within the context of TLR-induced co-stimulatory signals. Introduction A key factor in the induction of immune responses is the Received 6 September, 2006; revised and accepted 25 October, 2006. *For correspondence. E-mail [email protected]; Tel. (+1) 212 659 9407; Fax (+1) 212 849 2525. © 2007 The Author Journal compilation © 2007 Blackwell Publishing Ltd

ability of dendritic cells (DCs) to internalize antigen and present antigen-derived peptides on major histocompatibility complex (MHC) molecules for recognition by T lymphocytes (Mellman, 2005). During infection, exogenous antigens are derived from phagocytosed microbial pathogens and presented by MHC class II molecules (MHC II), whereas endogenous antigens are those synthesized by infected cells and presented by MHC class I molecules (MHC I). This demarcation in antigen processing pathways ensures that CD4+ T cells target exogenous antigens to provide help to B cells that internalized these antigens through their B cell receptor (McHeyzer-Williams and McHeyzer-Williams, 2005), and help to CD8+ T cells recognizing their cognate ligand co-presented by DCs (Bevan, 2004). This demarcation also ensures that CD8+ T cells target endogenously synthesized antigens in infected cells for selective recognition and killing of these cells while sparing healthy neighbouring cells. DCs can also internalize exogenous antigens and present them within MHC I in a process called cross-presentation (Carbone et al., 1998; Rock and Shen, 2005). Following internalization, DCs migrate to lymph nodes delivering information from microbial pathogens and infected cells in the periphery to naïve T cells (Iwasaki and Medzhitov, 2004). Although there are many different modes of internalization into DCs, I focus my discussion on phagocytosis in order to highlight the existence of a subcellular level of regulation in DCs, which functions at the level of phagosomes and controls presentation of antigens within MHC II. I will first review the evidence arguing for the intersection of regulatory signalling pathways with the phagocytic pathway because it is central to understanding how and why phagosome maturation is controlled by signals in the first place. One route for internalization of self and non-self The same cellular components involved in internalization of cargo associated with a cell’s normal metabolic and housekeeping functions are used in internalization of microbial pathogens. A very good example of this is provided in phagocytosis where professional phagocytes like macrophages and DCs use phagocytosis to clear not only

Signalling and phagocytosis 291 apoptotic cells and other by-products of normal tissue turnover, but also microbial pathogens (Greenberg and Grinstein, 2002; Henson and Hume, 2006). Microbial pathogens in turn exploit the endocytic pathway in order to enter subcellular compartments. For example, although lipid rafts/caveolae have diverse cellular functions in signal transduction, polarized secretion and membrane transport, they are targeted as entry points by many viruses like SV40, HIV, measles and influenza, fimbriated bacteria, Mycobacterium species, and parasites like Plasmodium falciparum (Duncan et al., 2002; Marsh and Helenius, 2006). Clathrin-coated pits (CCPs) in turn offer a very efficient mode of internalization, and are targeted for example by Influenza A viruses, adenoviruses and the bacterial pathogen Listeria monocytogenes (Veiga and Cossart, 2005; Marsh and Helenius, 2006). In general, most viruses require endocytosis in order to productively infect cells where they use many properties of endocytic pathways to their own advantage. However, viruses like HIV, Sendai virus and herpes simplex virus 1 (HSV-1) can enter cells without relying on the endocytic pathway by directly fusing with the plasma membrane (Marsh and Helenius, 2006). Internalization pathways can thus directly deliver microbial pathogens to a protected intracellular niche away from extracellular mechanisms of host defence. With this loophole in internalization comes the necessity for immunosurveillance mechanisms that patrol the endocytic pathway. Patrolling the portals of entry into cells Among all the endocytic routes of entry into cells, immunosurveillance of the endosome–lysosome pathway is the best characterized. Internalized cargo is delivered to lysosomes where a combination of reactive oxygen and nitrogen intermediates, acidic pH and cathepsin hydrolytic enzyme activities all contribute to degradation (MacMicking et al., 1997; Nishi and Forgac, 2002; Cross and Segal, 2004; Trombetta and Mellman, 2005). Endocytosis efficiently achieves this by relying on an extensive network of endocytic vesicles that deliver cargo into lysosomes (Vieira et al., 2002). This endosome–lysosome pathway is so effective that microbial pathogens have devised many strategies to avoid this hurdle either by seeking any of the other routes of internalization into cells, or by preventing their endocytic home from fusing with lysosomes. Internalization through caveolae for example, avoids the endosome–lysosome pathway although a fraction of internalized material may still intersect this pathway (Duncan et al., 2002). Mycobacterium tuberculosis, Salmonella typhimurium, Legionella pneumophila and Francisella tularensis halt this endocytic pathway at different stages, and with the exception of mycobacteria use a sophisticated type III or type IV secretion system to do so (Duclos

and Desjardins, 2000; Vergne et al., 2004; Roy et al., 2006). L. monocytogenes, F. tularensis and numerous viruses escape the endocytic pathway and enter the cytosol (O’Riordan and Portnoy, 2002; Marsh and Helenius, 2006). Our increased understanding of innate immune defence mechanisms in recent years allows us to put together an updated view of immunosurveillance of the endocytic pathway and cytosol. Along the endocytic pathway, Toll-like receptors (TLRs) are expressed at optimal locations for detecting microbial ligands. TLRs recognize various microbial molecular structures, and initiate signalling pathways through the sorting/signalling adaptor protein pairs TIRAP/MyD88 and TRAM/TRIF that activate mitogen-activated protein kinases (MAPK) p38, ERK and JNK, and transcription of NF-kB and interferon regulatory factor (IRF)-responsive immune responses genes (Akira et al., 2006). TLR1, 2, 4, 5 and 6 are mostly expressed on the plasma membrane. TLR2 can be recruited to phagosomes forming around Saccharomyces cerevisiae (Underhill et al., 1999). TLR4 expression is restricted predominantly to early endosomes and the plasma membrane (Barton et al., 2006), and presumably all surface TLRs would recycle between these compartments. Signalling from these TLRs induces transcription of inflammatory cytokines like IL-12 and IL-6. TLR3, 7, 8 and 9, on the other hand, are confined to late endocytic compartments and recognize microbial constituents exposed only after these microorganisms have entered the endocytic compartment and have been subjected to degradation. Both enveloped and nonenveloped viral particles disassemble upon entry into cells releasing their viral nucleic acids (Marsh and Helenius, 2006). The expression of TLR 3, 7, 8 and 9 is thus ideally suited for the detection of viruses. Similar to viruses, delivery of bacteria to lysosomes results in the degradation of their DNA genome into shorter CpG containing sequences that engage TLR9. Unlike mammalian genomic DNA, bacterial genomic DNA contains abundant unmethylated CpG dinucleotides. However, mammalian DNA has been reported to stimulate TLR9 under pathological conditions (Leadbetter et al., 2002; Viglianti et al., 2003), and similarly single-stranded RNA of non-viral origin was found to stimulate TLR7 (Diebold et al., 2004), indicating that endosomal TLRs could recognize nucleic acids of both self and non-self origin. However, compartmentalized expression of both endosomal TLRs and selfRNA contributes to the preferential recognition of microbial rather than self nucleic acids (Diebold et al., 2004; Barton et al., 2006). Engagement of endosomal TLRs by nucleic acids induces type I interferons (IFN) that have potent antimicrobial activities. The cytosol is also equipped with its own immunosurveillance capabilities. There are currently two known families of proteins that function in this role, the NOD-LRR

© 2007 The Author Journal compilation © 2007 Blackwell Publishing Ltd, Cellular Microbiology, 9, 290–299

292 J. M. Blander proteins (nucleotide binding oligomerization domain, NOD1 and NOD2) and the CARD-helicase proteins (retinoic-acid-inducible protein I, RIG-I, and the melanoma differentiation associated gene 5, MDA-5) reviewed in the study by Akira et al. (2006) and will not be discussed further here. Thus, expression of many signalling receptors with critical roles in host defence is juxtaposed with the endocytic pathway, and even goes beyond it into the cytosol. Signal transduction pathways mobilized through TLRs ensure that microbial exploitation of internalization pathways into cells is met with a vigorous immune response. An emerging issue is whether signalling pathways besides executing their roles in the transcriptional initiation of immune response genes regulate the endocytic pathway itself to obtain optimal host defence. Because phagocytosis relies heavily on endocytic traffic, it is plausible that signalling from TLRs could regulate multiple aspects of phagocytosis. Let us first consider whether there is precedence for cross-talk between the endocytic pathway and defined signal transduction pathways. Integration of endocytosis with signal transduction There are many examples showing integration between signalling and endocytic traffic, reviewed in the studies by Cavalli et al. (2001a), Di Fiore and De Camilli (2001), McPherson et al. (2001) and Gonzalez-Gaitan (2003), as a result of which signalling can impact endocytosis and vice versa. I will consider only a very few examples here. Signalling from various receptor tyrosine kinases like EGFR and NGFR enhances endocytosis by increasing recruitment of clathrin and stimulating formation of CCPs. Signalling through EGFR induces Rab5 activation and tyrosine phosphorylation of the endocytic regulatory protein Eps15 that play essential roles in endocytosis of EGFR (Barbieri et al., 2000; Confalonieri et al., 2000). Ligand-dependent endocytosis of the opioid receptor occurs upon p38 phosphorylation resulting in subsequent phopshorylation of the Rab5 effectors EEA1 and Rabenosyn-5 (Mace et al., 2005). More recently, an RNAibased screen for potential regulators of endocytosis revealed that more than a third of the 590 kinases examined affected either SV40 and/or vesicular stomatitis virus (VSV) infection that enter cells via caveolae/rafts and CCPs, respectively (Pelkmans et al., 2005), suggesting that many signalling pathways where these kinases participate potentially have regulatory input on the endocytic pathway. In a reciprocal role, endocytosis in turn impacts signal transduction. Receptor endocytosis has classically been regarded as a means to terminate signalling but it is now well appreciated that many signalling complexes assemble on endocytic membranes to properly trigger signal transduction. Inhibition of endocytosis, for example,

results in decreased signalling downstream of EGFR or the b2-adrenergic receptor (Miaczynska et al., 2004). Endocytosis also delivers critical signalling adaptors to membrane receptors as in the case of ADP ribosylation factor (ARF) 6-dependent delivery of the adaptor TIRAP to TLR4 to initiate signalling (Kagan and Medzhitov, 2006). In another example, endocytosis of Notch is a crucial step for signal transduction that precedes a critical cleavage event necessary for releasing the Notch intracellular domain for transport to the nucleus (Le Borgne et al., 2005). Inflammatory versus non-inflammatory phagocytosis Phagocytosis relies on a network of endocytic vesicles to deliver cargo from nascent phagosomes to lysosomes for degradation (Vieira et al., 2002). In as much as endocytosis can be regulated by signal transduction, it becomes conceivable that the endocytic traffic that greets incoming phagosomes can itself be regulated by signals from those receptors engaged during phagocytosis. The question I thus now pose is whether phagocytosis can be regulated by signals from receptors, and if so how can this occur, and what would the consequences of such regulation be? The primary step where signal transduction occurs during phagocytosis is at the point of cargo internalization when a particle binds to any of a number of phagocytic receptors and initiates internalization (Niedergang and Chavrier, 2004). Let us consider what these receptors may be. Logically, the nature of the ligands on the internalized cargo directly dictates the types of receptors engaged. For example, a particle opsonized with immunoglobulin will engage Fc-receptors (FcR); one opsonized with complement will engage complement receptors (CR). Similarly, mannose residues and phosphatidyl serine will engage distinct receptors. In addition to these examples, there are many other types of ligands that engage multiple receptors including C-type lectins, like DEC-205, Dectin-1, Langerin, and scavenger receptors, like SR-AI/II and MARCO (Gordon, 2002). Importantly, these receptors generally do not induce inflammatory responses on their own (with the exception of Dectin-1), and a few have been reported to cooperate with TLRs in the induction of some of these responses (Gordon, 2002; Brown et al., 2003; Gantner et al., 2003; Jozefowski et al., 2006). What does this mean in a broader sense? The choice of receptor–ligand engagement underlies the classification of phagocytosis into a distinct category. This could be opsonic versus non-opsonic phagaocytosis depending on whether or not the particle is opsonized by immunoglobulin or complement, both of which are opsonins (from Greek opsOnein: to provide a meal). Phagocytosis can also be either inflammatory versus non-inflammatory


Signalling and phagocytosis 293 depending on whether particle internalization induces an inflammatory immune response or not (Niedergang and Chavrier, 2004). These classifications are not merely descriptive but rather have important consequences on the immune system, determined in turn by decisions made at the level of receptor–ligand interactions engaged during phagocytosis. The critical determinant here is whether the receptors engaged trigger activation of immune defence functions. Thus, a distinction has to be made between receptors that trigger mobilization of the necessary cellular components driving particle internalization, and those that induce immune response effectors and transcription of host defence genes. Particle internalization involves the focal delivery of VAMP3/cellubrevin containing endomembranes to newly forming phagosomes through the ARF6, ‘zippering’ of membranes along the internalized particle, actin polymerization following the activation of Rho-GTPases, and particle engulfment associated with PI3K activity (Niedergang and Chavrier, 2004). On the other hand, engagement of other types of receptors like FcR, mannose receptor and Dectin-1 for example, additionally triggers an inflammatory immune response associated with the production of inflammatory cytokines like IL-1b, IL-6, TNF-a and IL-12 (Aderem and Underhill, 1999; Swanson and Hoppe, 2004; Brown, 2006; Gross et al., 2006). Toll-like receptors lead the pack of receptors that induce inflammatory responses (Akira et al., 2006). Professional phagocytes like macrophages and DCs express TLRs, which do not act as phagocytic receptors but rather, their engagement by microbial structures during phagocytosis of microbial pathogens induces a strong inflammatory response. In contrast, the phagocytosis of apoptotic cells is not associated with inflammation but is rather considered to be anti-inflammatory. The difference in immune response in this case is determined primarily by the engagement of TLRs exclusively during the inflammatory phagocytosis of microbial pathogens, and not during noninflammatory phagocytosis of apoptotic cells. TLR signalling from phagosomes Phagosomes are organelles that form de novo at the plasma membrane during phagocytosis, and maintain an asymmetric lipid bilayer with associated proteins and lipids. Once inside cells, the composition of these membranes gains a unique character determined first and foremost through the types of receptors engaged during phagocytosis. With time, phagosomes acquire new components through sequential fusions with the endocytic pathway, a process described as phagosome maturation (Vieira et al., 2002). Phagosome membranes can then become quite complex. Indeed, proteomic analysis of phagosomes containing latex beads has illustrated that at one given time

point, one thousand distinct proteins can be found, of which only 500 have been identified (Brunet et al., 2003). How do professional phagocytes discern the nature of the contents of an incoming phagosome? The most likely mechanism is through the recruitment of TLRs to the nascent phagosome (Underhill et al., 1999; Ozinsky et al., 2000). This recruitment appears to occur constitutively regardless of whether TLR ligands are engaged or not, as the enrichment of TLRs on phagosomes does not depend on signalling (Underhill et al., 1999). When microbial ligands are present, recruited TLRs are engaged and a signalling complex is assembled. Indeed, upon phagocytosis of S. cerevisiae, the TLR sorting adaptor TIRAP is recruited to phagosomes in a phosphatidylinositol 4,5bisphosphate (PI(4,5)P2)-dependent manner (Kagan and Medzhitov, 2006; and J.C. Kagan, pers. comm.). The assembly of TLR signalling complexes specifically on phagosomes that contain microbial cargo can thus potentially endow these phagosomes with a distinct proteomic profile that directly dictates the fate of that phagosome, its contents, as well as its associated functions. TLR control of phagosome maturation As I discussed earlier, the type of receptor–ligand interactions that occur during the early steps of phagocytosis determines whether phagocytosis is inflammatory or noninflammatory. At the subcellular level, however, these first receptor–ligand interactions have a different consequence on the phagosome itself. Upon phagocytosis of microbial pathogens, TLRs present on the plasma membrane like TLR4 and TLR2 are engaged by microbial structures. In contrast, because apoptotic cells do not contain TLR ligands, TLR signalling is not engaged. This is not to say that no signalling whatsoever occurs upon phagocytosis of apoptotic cells (Henson and Hume, 2006). Distinct signalling pathways are indeed very likely engaged that in turn impact the apoptotic cell phagosome proteome. The nature of this phagosome proteome is currently unclear. Phagosome maturation is controlled by signals from TLRs through the adaptor protein MyD88 and the MAPK p38 (Blander and Medzhitov, 2004). When TLR signalling is engaged, an inducible mode of phagosome maturation is observed with distinct kinetics and functional consequences. Microbial pathogens like Escherichia coli, Staphylococcus aureus and S. typhimurium, engage TLRs during internalization and are delivered to lysosomes at an inducible rate manifested by increased clearance and phagolysosomal fusion. These studies revealed the existence of two modes of phagocytosis, constitutive and inducible. Although TLR signalling pathways were shown to be responsible for the inducible mode, it is likely that other signalling pathways downstream of receptors


294 J. M. Blander that elicit inflammatory responses in APCs, like mannose receptor and FcR, also trigger the inducible mode of phagocytosis (Blander and Medzhitov, 2006a). How can signalling from inflammatory receptors during phagocytosis influence phagosome maturation? Candidate targets are as follows. Target phagosome proteomes If one envisions a particular phagosome proteome in its basal state, the imposition of a TLR signal (or maybe a signal from FcR or mannose receptor) could mobilize a set of biochemical changes in the state of protein phosphorylation, for example, or the GTP/GDP-bound forms of small GTPases like Rhos and Rabs. How can a single MAPK like p38 mobilize such changes? Gareth Griffiths has recently described a model where the activation of a few key components like MAPKs, or protein kinase C on an individual phagosome could switch its proteome from one idle state to a second active state, based on the idea of a ‘scale-free’ network (Griffiths, 2004). Barabasi originally proposed that such a network has a tendency to self-evolve because an extensive number of interconnections are assigned to a select few nodes, whereas a small number of connections are assigned to the majority of remaining nodes. This pattern is mathematically referred to as ‘power law’ (Barabasi and Albert, 1999). Barabasi presents an effective argument that cellular signalling systems predominantly use this type of network for selforganization (Jeong et al., 2000). A simple event such as phosphorylation of p38 molecules specifically by signals from TLRs can thus set into motion a constellation of dynamic changes in the composition of the entire phagosome proteome. For example, stress-induced activation of p38 results in p38-dependent phosphorylation of the guanine nucleotide dissociation inhibitor (GDI) (Cavalli et al., 2001b). Phosphorylated GDI acquires an increased affinity for the GDP-bound form of Rab5 present on target endocytic membranes, and shuttles GDP-Rab5 through the cytosol back to donor membranes where nucleotide exchange takes place. This allows prompt recycling of Rab5 thereby overcoming what is considered to be a rate-limiting component in endocytic transport. Therefore, a distinct signalling complex may assemble around the TLR-activated p38 scaffold, the components of which may vary when TLRs are engaged than when they are not engaged. This TLR-initiated complex can then endow on that phagosome, a TLR-based molecular signature that dictates the immediate fate of the cargo and the immune responses tailored to that cargo. Target phagosome motility Another potential way in which inflammatory receptor signalling can trigger inducible phagosome maturation is to

increase the rate at which phagosomes move along microtubules. The mobility of endocytic organelles, including phagosomes, along microtubules occurs in both directions guided by motor protein complexes, kinesin/ kinectin and dynein/dynactin (Blocker et al., 1998; Karcher et al., 2002). The presence of dynactin on phagosome membranes recruits the cytosolic dynein and this appears to require a post-translational event such as phosphorylation by a dynactin-associated kinase (HammAlvarez et al., 1993; Blocker et al., 1997). Adenoviruses increase minus-end-directed motility mediated the dynein/dynactin motor complex (Suomalainen et al., 2001) to reach the nucleus where they replicate. They do so by stimulating both the cAMP-dependent protein kinase A (PKA) and the MAPK p38 (Suomalainen et al., 2001). Similarly, microtubule-associated proteins (MAPs) like CLIP-115 (De Zeeuw et al., 1997) and CLIP-170 (Pierre et al., 1992) may increase the rate of phagosome motility by dislodging from microtubules upon phosphorylation (Rickard and Kreis, 1996). Therefore, the regulation of motor protein activity by signal-dependent phosphorylation for example, may be another mechanism of regulating endocytic traffic. TLR control of MHC class II presentation How does control of phagosome maturation relate to MHC presentation and DC function? Processing of phagocytosed cargo within lysosomes results in the degradation of cargo proteins into smaller peptides that are assembled with MHC II and transported to the plasma membrane (Trombetta and Mellman, 2005). These newly synthesized peptide : MHC II complexes are then recognized by CD4+ T cells having the corresponding T cell receptor specificity. This process is an important immunological consequence of phagocytosis, and it is likely controlled by TLRs at various steps (Chow and Mellman, 2005). It has been known for some time that the stimulation of DCs with TLR ligands results in the movement of MHC II from lysosomal compartments to the plasma membrane (Turley et al., 2000). As such, one of the hallmarks of DC differentiation (maturation) is the upregulation of MHC II. The appearance of these molecules on the plasma membrane may be due to TLR signals that regulate the endocytic machinery mediating retrograde transport of MHC II to the plasma membrane, or it may directly be related to TLR regulation of earlier steps involving either antigen or invariant chain (Ii) processing. Ii is a type II integral membrane protein that chaperones newly synthesized MHC II from the endoplasmic reticulum (ER) to the Golgi targeting it to the endocytic pathway (Cresswell, 1996). The hydrolytic compartment in maturing phagosomes is important not only for processing of the cargo, but also for processing


Signalling and phagocytosis 295 of Ii through a series of proteolytic steps such that an N-terminal fragment called the class II associated invariant chain-derived peptide (CLIP) remains occupying the peptide-binding groove. CLIP is then exchanged for peptides derived from cargo proteins through a reaction catalysed by the non-classical MHC molecules called H2-DM and DO (mouse) or HLA-DM and DO (human). Many of these steps can be subject to regulation. TLR signals regulate the processing of Ii (Blander and Medzhitov, 2006b). Phagosomes that do not engage TLR signalling fail to process Ii. The endosome– lysosome targeting sequence within the Ii cytoplasmic domain could thus target its associated MHC II to lysosomes where all is degraded together with the phagocytosed cargo. Thus, phagosomes that do not engage TLR signalling mature into terminal lysosomes and no presentation of cargo-derived peptides within MHC II takes place. On the other hand, phagosomes where TLR signalling is engaged are met with a different fate. Here, TLR signals result in processing of Ii, likely through TLRmediated activation of proteolytic enzyme activity, as discussed above and reviewed in the study by Watts (2004). How can signalling from TLRs regulate antigen processing and MHC II presentation? A few possible candidate targets are as follows. Target phagolysosome acidification Within most cell types, including non-professional APCs such as fibroblasts, fusion of phagosomes with late endosomes and lysosomes during phagosome maturation is by default accompanied by a rapid reduction in phagosomal pH. Despite this default mechanism, this process can be regulated by external signals like the engagement of TLRs such that it can proceed at an increased inducible rate. This is most evident in immature DCs, which seem to actively maintain a more alkaline pH within their phagosomes, through recruitment of the NADPH oxidase NOX2 (Savina et al., 2006), and upon stimulation by TLR ligands lower the pH of their phagolysosomes by approximately one pH unit from ª5.5 to 4.5 (Trombetta et al., 2003). In DCs, this decrease in pH is essential not only for the proteolyic activity of degradative enzymes, but also for many steps in MHC II presentation including the dissociation of Ii [reviewed in the study by Trombetta and Mellman (2005)]. Although the effect of TLR engagement on NOX2 recruitment to phagosomes has not been studied directly, one study has reported increases in NOX2 activity by TLR ligands (Vulcano et al., 2004). In alveolar macrophages, acidification of lysosomes also involves the cystic fibrosis transmembrane conductance regulator (CFTR), which is a specific Cl– ion channel that promotes luminal H+ accumulation such that acidification was impaired in both

lysosomes and phagosomes in macrophages from Cftr–/– mice (Di et al., 2006). Phagolysosomal acidification is primarily mediated through the assembly of the vacuolar proton-adenosine triphosphatase (V-ATPase). This is a multisubunit complex composed of peripheral (V1) and integral (V0) domains each consisting of individual polypeptides (Nishi and Forgac, 2002). In the absence of signals from TLRs, although phagosomes containing E. coli and S. typhimurium do colocalize with the acidophilic dye LysoTracker, they do so at a slower rate suggesting that these phagosomes fail to undergo the inducible rate of acidification in the absence of signals from TLRs (Blander and Medzhitov, 2004). The presence of multiple mechanisms to regulate phagolysosomal acidification suggests that this process can be regulated and is not merely a default mechanism occurring during phagolysosomal fusion. Both V-ATPase and NOX2 activities are likely regulated by signals from membrane receptors like TLRs either by controlling the density of these complexes on the phagosomal membrane, the recruited assembly of their cytosolic components to phagosomal membranes, or by regulating the activity of inhibitory and activating associated proteins (Nishi and Forgac, 2002). Microbial pathogens exploit these regulatory mechanisms to control acidification of phagolysosomes in their favour. Mycobacterium avium, for example, inhibits acidification of its phagosomes by selectively inhibiting the recruitment of the V-ATPase (SturgillKoszycki et al., 1994). Target distinct classes of proteases Two broad classes of cathepsins or hydrolytic enzymes exist within late endocytic and lysosomal compartments (Watts, 2004). TLRs may control the activity of these hydrolytic enzymes either directly or indirectly by controlling phagosomal pH. Those enzymes that directly control processing of Ii, like the asparagine-specific endopeptidase AEP, cathepsins S, B, D, L and F would be positively regulated by TLRs and these would be concerned mainly with the preparation of MHC II into peptide loading competent molecules. The second would not be positively regulated by TLRs but activated as part of the constitutive mode of phagosome maturation. These cathepsins and hydrolytic enzymes are concerned with processing cargoderived proteins in preparation for their loading onto MHC II. The IFN-g-inducible lysosomal thiol reductase GILT and other exopeptidases could be possible candidates within this second class (Watts, 2004). It remains entirely likely that TLRs may partially inhibit the activity of this latter group of enzymes in order to prevent complete destructive degradation of cargo-derived proteins. In this way, by exerting positive and negative regulation on distinct subsets of hydrolytic enzymes, signalling from TLRs may


296 J. M. Blander help achieve the proper balance in creating the optimal environment for antigen processing such that antigenic peptides can be rescued from complete degradation and become available to bind MHC II. Phagosome autonomous behaviour directed by TLRs Notably, TLR control of the kinetics and function of phagosome maturation is phagosome-autonomous. In this way within the same cell, maturation of a phagosome containing apoptotic cells, for example, is not influenced by TLR signalling at the plasma membrane or at the membrane of a different phagosome containing bacteria (Blander and Medzhitov, 2004). The compartmentalized assembly of TLR signalling complexes specifically along phagosomal membranes containing cargo that carries TLR ligands can easily achieve this. Thus, when a single cell phagocytoses both bacteria and apoptotic cells at the same time, each cargo is dealt with separately within the confines of distinct phagosomal membranes suggesting that phagosomes achieve and retain individual complexity within a single cellular space. This is aided by the distinct nature of receptor–ligand interactions that form at the outset of internalization as I discussed earlier. Distinct signalling complexes can thus dock on these phagosomal membranes and form individual signalling networks imparting each phagosome with its own signature molecular profile. In this way, each phagosome could undergo its individually tailored programme of maturation regardless of the activation status of the DC. This intriguing concept also suggests that beneath the plasma membrane lies another level of immunosurveillance, one that deals with new organelles (phagosomes) carrying elements of the extracellular environment. One important consequence of phagosome individuality is the uncoupling of processing and presentation from TLR-induced expression of co-stimulatory molecules (Blander and Medzhitov, 2006b). This may play an important physiological role in tissues during infection where DCs can phagocytose both microbial pathogens and tissue apoptotic cells. In such a scenario, it is expected that MHC II will present immunogenic peptides from both types of phagocytosed cargo to naïve CD4+ T cells in the context of T cell co-stimulatory molecules. Although this is ideal for the activation of T cells against microbial derived antigens, the activation of T cells against apoptotic cellderived antigens can lead to adverse autoimmune responses. Using TLR-dependent selection of antigens, only antigenic peptides derived from phagosomes that engage TLR signalling are presented by MHC II and travel to the plasma membrane (Blander and Medzhitov, 2006b). Apoptotic cells do not engage TLRs and their antigens are not presented regardless of the engagement

of TLR signalling in the same cell at the plasma membrane by soluble TLR ligands or from phagosomes containing bacteria. Based on these findings, a model emerges where upon simultaneous phagocytosis of microbial and apoptotic cells, phagosomes that contain microbial cargo that engage TLRs are favoured for MHC II presentation (Fig. 1). These phagosomes mature with enhanced kinetics of fusion with the endocytic pathway, and activate hydrolytic enzymes that process Ii making MHC II receptive to binding of antigenic peptides. Apoptotic cells, on the other hand, do not engage TLR signalling and their phagosomes mature into terminal lysosomes where apoptotic cell proteins are completely degraded. Here, specific proteases that cleave Ii are not activated and instead MHC II bound to unprocessed Ii is targeted to lysosomes for degradation. Phagosome autonomous control by TLRs thus ensures that processing of Ii occurs only within phagosomes containing microbial cells where ligands on the cargo engage compartmentalized TLR signalling pathways. The resultant peptide : MHC complexes that form specifically within these phagosomes are then transported to the plasma membrane and presented to T cells in the context of TLR-induced co-stimulation. Apoptotic cell-derived peptides are excluded from these MHC II and are rather terminally degraded. Perspectives Phagosome autonomous control by compartmentalized signalling pathways is a powerful mechanism that launches new possibilities. For designing vaccine formulations, rather than mixing TLR ligands and antigens together, physically linking the antigen with the TLR ligand can ensure delivery into a phagosome that by the very nature of its cargo is supremely tailored for MHC II presentation. DCs gain a new level of subcellular surveillance that scans the molecular profile on individual phagosomal membranes. In this way, DCs can internalize different cargos and execute appropriate fates for these cargos while maintaining one particular activation state dominated by the prevailing signalling pathway. But new questions arise. How are infected apoptotic cells that deliver both self and microbial constituents into one phagosome handled? Is cross-presentation of phagocytosed antigens subject to TLR control? Is the presentation of antigens internalized through pathways of internalization other than phagocytosis, also subject to regulatory control by TLRs or other inflammatory receptors? Are there differences in the control of antigen presentation among different DC subsets? What are the characteristics of an apoptotic cell phagosome? How is tolerance to self antigens derived from apoptotic cells ever induced?


Signalling and phagocytosis 297

Fig. 1. Model for TLR-dependent discrimination of self from non-self at the level of vesicular traffic. By virtue of engaging compartmentalized TLR signals, phagosomes containing microbial cargo mature autonomously with enhanced kinetics of fusion with the endocytic pathway. The pool of p38 MAPK species specifically phosphorylated by signals from TLRs could increase the rate of endocytic traffic by phosphorylating the cytosolic guanine-nucleotide dissociation inhibitor GDI, which in phosphorylated form acquires an increased affinity for exhausted GDP-bound Rabs (Rab5 : GDP), and recycles them through the cytosolic phase of their activation back to donor vesicles for reloading with GTP (Cavalli et al., 2001a). Phagosome maturation is accompanied by a progressive decrease in pH, as indicated in the coloured boxes, a process that ensures processing and degradation of the cargo. However, the presence of a TLR signal additionally activates Ii proteolysis within these compartments making MHC II receptive to binding of antigenic peptides with the help of H2-DM. Apoptotic cells, on the other hand, do not engage TLR signalling and their phagosomes mature into distinct compartments where MHC II bound to unprocessed Ii is targeted to lysosomes for degradation. The nature of apoptotic cell phagosomes is not known. Upon the receipt of a TLR signal, the pool of MHC II that travels to the surface derives selectively from phagosomes where TLR signalling is engaged. Phagosome autonomous control by TLRs thus ensures that microbial antigens are preferentially presented to T cells in the context of TLR-induced co-stimulatory molecule expression (shown here as CD80/CD86 on mature DC engaging CD28 on naïve CD4+ T cells).

Acknowledgements I am grateful to all my past and present colleagues. I thank the Arthritis Foundation for continued support of my work. I extend my regrets to those authors whose relevant studies I could not cite because of space limitations.

References Aderem, A., and Underhill, D.M. (1999) Mechanisms of phagocytosis in macrophages. Annu Rev Immunol 17: 593–623. Akira, S., Uematsu, S., and Takeuchi, O. (2006) Pathogen recognition and innate immunity. Cell 124: 783–801. Barabasi, A.L., and Albert, R. (1999) Emergence of scaling in random networks. Science 286: 509–512. Barbieri, M.A., Roberts, R.L., Gumusboga, A., Highfield, H., Alvarez-Dominguez, C., Wells, A., and Stahl, P.D. (2000) Epidermal growth factor and membrane trafficking. EGF receptor activation of endocytosis requires Rab5a. J Cell Biol 151: 539–550.

Barton, G.M., Kagan, J.C., and Medzhitov, R. (2006) Intracellular localization of Toll-like receptor 9 prevents recognition of self DNA but facilitates access to viral DNA. Nat Immunol 7: 49–56. Bevan, M.J. (2004) Helping the CD8(+) T-cell response. Nat Rev Immunol 4: 595–602. Blander, J.M., and Medzhitov, R. (2004) Regulation of phagosome maturation by signals from toll-like receptors. Science 304: 1014–1018. Blander, J.M., and Medzhitov, R. (2006a) On regulation of phagosome maturation and antigen presentation. Nat Immunol 7: 1029–1035. Blander, J.M., and Medzhitov, R. (2006b) Toll-dependent selection of microbial antigens for presentation by dendritic cells. Nature 440: 808–812. Blocker, A., Severin, F.F., Burkhardt, J.K., Bingham, J.B., Yu, H., Olivo, J.C., et al. (1997) Molecular requirements for bi-directional movement of phagosomes along microtubules. J Cell Biol 137: 113–129. Blocker, A., Griffiths, G., Olivo, J.C., Hyman, A.A., and Severin, F.F. (1998) A role for microtubule dynamics in


298 J. M. Blander phagosome movement. J Cell Sci 111 (Part 3): 303– 312. Brown, G.D. (2006) Dectin-1: a signalling non-TLR patternrecognition receptor. Nat Rev Immunol 6: 33–43. Brown, G.D., Herre, J., Williams, D.L., Willment, J.A., Marshall, A.S., and Gordon, S. (2003) Dectin-1 mediates the biological effects of beta-glucans. J Exp Med 197: 1119– 1124. Brunet, S., Thibault, P., Gagnon, E., Kearney, P., Bergeron, J.J., and Desjardins, M. (2003) Organelle proteomics: looking at less to see more. Trends Cell Biol 13: 629– 638. Carbone, F.R., Kurts, C., Bennett, S.R., Miller, J.F., and Heath, W.R. (1998) Cross-presentation: a general mechanism for CTL immunity and tolerance. Immunol Today 19: 368–373. Cavalli, V., Corti, M., and Gruenberg, J. (2001a) Endocytosis and signaling cascades: a close encounter. FEBS Lett 498: 190–196. Cavalli, V., Vilbois, F., Corti, M., Marcote, M.J., Tamura, K., Karin, M., et al. (2001b) The stress-induced MAP kinase p38 regulates endocytic trafficking via the GDI : Rab5 complex. Mol Cell 7: 421–432. Chow, A.Y., and Mellman, I. (2005) Old lysosomes, new tricks: MHC II dynamics in DCs. Trends Immunol 26: 72–78. Confalonieri, S., Salcini, A.E., Puri, C., Tacchetti, C., and Di Fiore, P.P. (2000) Tyrosine phosphorylation of Eps15 is required for ligand-regulated, but not constitutive, endocytosis. J Cell Biol 150: 905–912. Cresswell, P. (1996) Invariant chain structure and MHC class II function. Cell 84: 505–507. Cross, A.R., and Segal, A.W. (2004) The NADPH oxidase of professional phagocytes – prototype of the NOX electron transport chain systems. Biochim Biophys Acta 1657: 1–22. De Zeeuw, C.I., Hoogenraad, C.C., Goedknegt, E., Hertzberg, E., Neubauer, A., Grosveld, F., and Galjart, N. (1997) CLIP-115, a novel brain-specific cytoplasmic linker protein, mediates the localization of dendritic lamellar bodies. Neuron 19: 1187–1199. Di, A., Brown, M.E., Deriy, L.V., Li, C., Szeto, F.L., Chen, Y., et al. (2006) CFTR regulates phagosome acidification in macrophages and alters bactericidal activity. Nat Cell Biol 8: 933–944. Di Fiore, P.P., and De Camilli, P. (2001) Endocytosis and signaling: an inseparable partnership. Cell 106: 1–4. Diebold, S.S., Kaisho, T., Hemmi, H., Akira, S., and Reis e Sousa, C. (2004) Innate antiviral responses by means of TLR7-mediated recognition of single-stranded RNA. Science 303: 1529–1531. Duclos, S., and Desjardins, M. (2000) Subversion of a young phagosome: the survival strategies of intracellular pathogens. Cell Microbiol 2: 365–377. Duncan, M.J., Shin, J.S., and Abraham, S.N. (2002) Microbial entry through caveolae: variations on a theme. Cell Microbiol 4: 783–791. Gantner, B.N., Simmons, R.M., Canavera, S.J., Akira, S., and Underhill, D.M. (2003) Collaborative induction of inflammatory responses by dectin-1 and Toll-like receptor 2. J Exp Med 197: 1107–1117.

Gonzalez-Gaitan, M. (2003) Signal dispersal and transduction through the endocytic pathway. Nat Rev Mol Cell Biol 4: 213–224. Gordon, S. (2002) Pattern recognition receptors: doubling up for the innate immune response. Cell 111: 927–930. Greenberg, S., and Grinstein, S. (2002) Phagocytosis and innate immunity. Curr Opin Immunol 14: 136–145. Griffiths, G. (2004) On phagosome individuality and membrane signalling networks. Trends Cell Biol 14: 343–351. Gross, O., Gewies, A., Finger, K., Schafer, M., Sparwasser, T., Peschel, C., et al. (2006) Card9 controls a non-TLR signalling pathway for innate anti-fungal immunity. Nature 442: 651–656. Hamm-Alvarez, S.F., Kim, P.Y., and Sheetz, M.P. (1993) Regulation of vesicle transport in CV-1 cells and extracts. J Cell Sci 106 (Part 3): 955–966. Henson, P.M., and Hume, D.A. (2006) Apoptotic cell removal in development and tissue homeostasis. Trends Immunol 27: 244–250. Iwasaki, A., and Medzhitov, R. (2004) Toll-like receptor control of the adaptive immune responses. Nat Immunol 5: 987–995. Jeong, H., Tombor, B., Albert, R., Oltvai, Z.N., and Barabasi, A.L. (2000) The large-scale organization of metabolic networks. Nature 407: 651–654. Jozefowski, S., Sulahian, T.H., Arredouani, M., and Kobzik, L. (2006) Role of scavenger receptor MARCO in macrophage responses to CpG oligodeoxynucleotides. J Leukoc Biol 80: 870–879. Kagan, J.C., and Medzhitov, R. (2006) Phosphoinositidemediated adaptor recruitment controls Toll-like receptor signaling. Cell 125: 943–955. Karcher, R.L., Deacon, S.W., and Gelfand, V.I. (2002) Motor– cargo interactions: the key to transport specificity. Trends Cell Biol 12: 21–27. Le Borgne, R., Bardin, A., and Schweisguth, F. (2005) The roles of receptor and ligand endocytosis in regulating Notch signaling. Development 132: 1751–1762. Leadbetter, E.A., Rifkin, I.R., Hohlbaum, A.M., Beaudette, B.C., Shlomchik, M.J., and Marshak-Rothstein, A. (2002) Chromatin–IgG complexes activate B cells by dual engagement of IgM and Toll-like receptors. Nature 416: 603–607. Mace, G., Miaczynska, M., Zerial, M., and Nebreda, A.R. (2005) Phosphorylation of EEA1 by p38 MAP kinase regulates mu opioid receptor endocytosis. EMBO J 24: 3235– 3246. McHeyzer-Williams, L.J., and McHeyzer-Williams, M.G. (2005) Antigen-specific memory B cell development. Annu Rev Immunol 23: 487–513. MacMicking, J., Xie, Q.W., and Nathan, C. (1997) Nitric oxide and macrophage function. Annu Rev Immunol 15: 323– 350. McPherson, P.S., Kay, B.K., and Hussain, N.K. (2001) Signaling on the endocytic pathway. Traffic 2: 375–384. Marsh, M., and Helenius, A. (2006) Virus entry: open sesame. Cell 124: 729–740. Mellman, I. (2005) Antigen processing and presentation by dendritic cells: cell biological mechanisms. Adv Exp Med Biol 560: 63–67.


Signalling and phagocytosis 299 Miaczynska, M., Pelkmans, L., and Zerial, M. (2004) Not just a sink: endosomes in control of signal transduction. Curr Opin Cell Biol 16: 400–406. Niedergang, F., and Chavrier, P. (2004) Signaling and membrane dynamics during phagocytosis: many roads lead to the phagos(R)ome. Curr Opin Cell Biol 16: 422–428. Nishi, T., and Forgac, M. (2002) The vacuolar (H+)-ATPases – nature’s most versatile proton pumps. Nat Rev Mol Cell Biol 3: 94–103. O’Riordan, M., and Portnoy, D.A. (2002) The host cytosol: front-line or home front? Trends Microbiol 10: 361–364. Ozinsky, A., Underhill, D.M., Fontenot, J.D., Hajjar, A.M., Smith, K.D., Wilson, C.B., et al. (2000) The repertoire for pattern recognition of pathogens by the innate immune system is defined by cooperation between toll-like receptors. Proc Natl Acad Sci USA 97: 13766–13771. Pelkmans, L., Fava, E., Grabner, H., Hannus, M., Habermann, B., Krausz, E., and Zerial, M. (2005) Genome-wide analysis of human kinases in clathrin- and caveolae/ raft-mediated endocytosis. Nature 436: 78–86. Pierre, P., Scheel, J., Rickard, J.E., and Kreis, T.E. (1992) CLIP-170 links endocytic vesicles to microtubules. Cell 70: 887–900. Rickard, J.E., and Kreis, T.E. (1996) CLIPs for organelle– microtubule interactions. Trends Cell Biol 6: 178–183. Rock, K.L., and Shen, L. (2005) Cross-presentation: underlying mechanisms and role in immune surveillance. Immunol Rev 207: 166–183. Roy, C.R., Salcedo, S.P., and Gorvel, J.P. (2006) Pathogen– endoplasmic-reticulum interactions: in through the out door. Nat Rev Immunol 6: 136–147. Savina, A., Jancic, C., Hugues, S., Guermonprez, P., Vargas, P., Moura, I.C., et al. (2006) NOX2 controls phagosomal pH to regulate antigen processing during crosspresentation by dendritic cells. Cell 126: 205–218. Sturgill-Koszycki, S., Schlesinger, P.H., Chakraborty, P., Haddix, P.L., Collins, H.L., Fok, A.K., et al. (1994) Lack of acidification in Mycobacterium phagosomes produced by exclusion of the vesicular proton-ATPase. Science 263: 678–681. Suomalainen, M., Nakano, M.Y., Boucke, K., Keller, S., and

Greber, U.F. (2001) Adenovirus-activated PKA and p38/ MAPK pathways boost microtubule-mediated nuclear targeting of virus. EMBO J 20: 1310–1319. Swanson, J.A., and Hoppe, A.D. (2004) The coordination of signaling during Fc receptor-mediated phagocytosis. J Leukoc Biol 76: 1093–1103. Trombetta, E.S., and Mellman, I. (2005) Cell biology of antigen processing in vitro and in vivo. Annu Rev Immunol 23: 975–1028. Trombetta, E.S., Ebersold, M., Garrett, W., Pypaert, M., and Mellman, I. (2003) Activation of lysosomal function during dendritic cell maturation. Science 299: 1400–1403. Turley, S.J., Inaba, K., Garrett, W.S., Ebersold, M., Unternaehrer, J., Steinman, R.M., and Mellman, I. (2000) Transport of peptide–MHC class II complexes in developing dendritic cells. Science 288: 522–527. Underhill, D.M., Ozinsky, A., Hajjar, A.M., Stevens, A., Wilson, C.B., Bassetti, M., and Aderem, A. (1999) The Toll-like receptor 2 is recruited to macrophage phagosomes and discriminates between pathogens. Nature 401: 811–815. Veiga, E., and Cossart, P. (2005) Listeria hijacks the clathrindependent endocytic machinery to invade mammalian cells. Nat Cell Biol 7: 894–900. Vergne, I., Chua, J., Singh, S.B., and Deretic, V. (2004) Cell biology of Mycobacterium tuberculosis phagosome. Annu Rev Cell Dev Biol 20: 367–394. Vieira, O.V., Botelho, R.J., and Grinstein, S. (2002) Phagosome maturation: aging gracefully. Biochem J 366: 689– 704. Viglianti, G.A., Lau, C.M., Hanley, T.M., Miko, B.A., Shlomchik, M.J., and Marshak-Rothstein, A. (2003) Activation of autoreactive B cells by CpG dsDNA. Immunity 19: 837– 847. Vulcano, M., Dusi, S., Lissandrini, D., Badolato, R., Mazzi, P., Riboldi, E., et al. (2004) Toll receptor-mediated regulation of NADPH oxidase in human dendritic cells. J Immunol 173: 5749–5756. Watts, C. (2004) The exogenous pathway for antigen presentation on major histocompatibility complex class II and CD1 molecules. Nat Immunol 5: 685–692.
