A growing list of innate immune receptors is being defined that recognize polysaccharides of microbial cell walls. Fungal β-glucan recognition by the receptor Dectin-1 triggers inflammatory immune responses in macrophages and dendritic cells that are appropriate for defense against fungal pathogens. Among these responses is the specific recruitment of the autophagy-related protein light chain 3 (LC3) to phagosomes containing fungi. Studies documenting LC3's recruitment to phagosomes containing β-glucan and other nonsugar particles suggest that LC3 plays a role in regulating phagocytosis and its related immunological responses.
Innate immune receptors recognize conserved chemical motifs typically found on foreign cells, but not on self. A well-studied such “Pathogen-Associated Molecular Pattern” or “PAMP” is the β-glucan polymer that comprises a large portion of the cell walls of fungi. Fungal β-glucan is a glucose polymer that varies in length and typically consists of β-(1,3) linkages with occasional β-(1,6) branches (Figure 1A). Fungal β-glucan is recognized by innate immune cells such as macrophages, dendritic cells and neutrophils. The primary β-glucan receptor on these cells has been identified as the C-type lectin pattern recognition receptor Dectin-1 that triggers phagocytosis, respiratory burst and inflammatory responses upon engaging fungi (Goodridge et al. 2009). Dectin-1 is essential for effective mucosal immunity to fungi in mice and humans. Dectin-1 knockout mice are substantially more susceptible to infection with many different fungi including Candida albicans, Pneumocystis carinii and Aspergillus fumigatus (Saijo et al. 2007; Taylor et al. 2007; Ferwerda et al. 2009; Werner et al. 2009; Leal et al. 2010; Marodi and Erdos 2010). Macrophages and dendritic cells from mice deficient in Dectin-1 show a drastically reduced respiratory burst in response to fungi and generally produce lower levels of inflammatory cytokines. Importantly, humans with rare genetic deficiencies in Dectin-1 or CARD9, an adaptor protein involved in Dectin-1 signaling, are specifically susceptible to fungal infections (Ferwerda et al. 2009; Glocker et al. 2009; Marodi and Erdos 2010). Additional receptors have been implicated in detection of β-glucans including the complement receptor 3 and lactosylceramide, although their specific contributions to β-glucan-induced immune responses are less well-defined. This review will focus on how β-glucan recognition by Dectin-1 leads to a phagocytic process tailored to optimize immune responses to fungi.
Dectin-1 is a member of the C-type lectin family of pattern recognition receptors. This is a growing family of receptors that includes, among others, Dectin-2, the mannose receptor, Mincle, DC-SIGN and CLEC5a (Vautier et al. 2010). These related receptors all contain at least one extracellular C-type lectin domain and play roles in recognition of extracellular microbes. They differ in their ligand specificities and can differ in their mechanisms of intracellular signaling.
Dectin-1 is a type II transmembrane protein containing a single extracellular C-type lectin domain that recognizes β-glucan, and a short intracellular signaling tail (Brown and Gordon 2001). The tail contains a signaling motif similar to the immunoreceptor tyrosine-based activation motif (ITAM) found in T cell receptors, B cell receptors and Fc receptors, but while a true ITAM has two tyrosines that are essential for signaling, the Dectin-1 motif uses just one and is thus called a “hemITAM” (Rogers et al. 2005; Underhill et al. 2005). The functional consequences of this difference are not yet clear. Like other ITAM-containing receptors, Dectin-1's hemITAM becomes phosphorylated by Src family kinases and recruits and activates Syk. These kinases are responsible for triggering phagocytosis and the recruitment and activation of the nicotinamide adenine dinucleotide phosphate (NADPH) phagocyte oxidase leading to a release of antimicrobial reactive oxygen into the phagosome (Gantner et al. 2003). In addition, Dectin-1 activation triggers inflammatory transcriptional responses through activation of nuclear factor of activated T cells (NFAT) and nuclear factor kappa-light-chain-enhancer of activated B cells. A unique feature of Dectin-1 related to its role as a phagocytosis-inducing receptor is that while it binds both soluble and particulate forms of β-glucan, it is only activated by particulate glucans. This process is regulated by the presence of CD45 and CD148 membrane phosphatases on phagocytic cells (Goodridge et al. 2011).
Details of the signaling pathway activated by Dectin-1 upon binding to β-glucan have been discussed in other recent reviews (Kerrigan and Brown 2011; Goodridge et al. 2012; Sancho and Reis e Sousa 2012). We will focus our discussion here on a relatively new addition to our understanding of the phagocytic response to β-glucans mediated by Dectin-1 called “LC3-associated phagocytosis”. Light chain 3 (LC3) is a microtubule-associated protein highly conserved across species. While well-known for its role in autophagy, recent findings have suggested that it may play a related but distinct role in the process of traditional phagocytosis (Sanjuan et al. 2007). Autophagy is a process whereby cells sequester cytosolic debris, organelles and pathogens in double-membrane autophagosomes (Levine et al. 2011). These newly formed intracellular compartments fuse with lysosomes which degrade and recycle their contents. In a state of metabolic stress or cytosolic infection, autophagy is activated and contributes to cell survival. Activation of autophagy and formation of autophagosomes is often measured by observing the accumulation of LC3 puntae. The formation of these puntae is directed by an autophagy signaling pathway that involves ATG5, ATG7, ATG12, ATG4 and Beclin. Newly synthesized LC3 exists as LC3I in the cytosol, and signaling by this pathway causes LC3I to become conjugated to phosphatidylethanolamine. This lipidated LC3 called LC3II becomes membrane-associated, localizing specifically to autophagosomes (Mizushima et al. 2008). Despite its wide use as a marker for autophagy, the exact role that LC3 plays in the process is largely unknown.
The appearance of LC3 as a highly specific marker of autophagy has been recently challenged. In 2007, Green and coworkers found that phagosomes in which Toll-like receptor 2 (TLR2) signaling was activated recruited GFP-tagged LC3 expressed in macrophages (Sanjuan et al. 2007). Plastic beads coated with a TLR2 ligand led to recruitment of LC3, while beads without a TLR ligand did not. By electron microscopy, the investigators confirmed that the organelle LC3 was recruited to be a traditional phagosome enclosed in a single membrane, not the double-membrane compartment formed during autophagy. Even though LC3 is recruited to a compartment that is not an autophagosome, the recruitment nevertheless requires additional signaling components utilized in autophagy. The investigators noted that LC3 recruitment to these phagosomes required Atg5 and Atg7 using siRNA knockdown and genetically deficient macrophages. They suggested that the process was important for phagosome maturation, since LAMP1 recruitment was delayed in cells lacking Atg5, and killing of live yeast was impaired in these cells. These findings suggest that we should begin to think about LC3 (and perhaps some other components of the traditional autophagy pathway) as potential regulators of immunologically important processes beyond autophagy.
We, therefore, investigated whether recognition of β-glucan by Dectin-1 could trigger formation of phagosomes that specifically recruit LC3. We found that phagosomes containing yeast or β-glucan particles specifically and rapidly trigger conversion of LC3I into LC3II and accumulation of the protein on phagosomes (Ma et al. 2012) (Figure 1B). We further demonstrated by directly cross-linking the receptor that Dectin-1 signaling is entirely sufficient to trigger LC3-associated phagocytosis. In cells lacking Dectin-1, recruitment of LC3 to β-glucan-containing phagosomes was completely abrogated, suggesting that Dectin-1 is the only receptor directing this response. In contrast, we found that conversion of LC3I into LC3II was partially deficient in Dectin-1-deficient cells exposed to C. albicans yeast, suggesting that while Dectin-1 is a key receptor for this response, others may also participate. As discussed above, TLRs are likely candidates.
In 2009, Brummel and coworkers observed that Fcγ receptors can trigger LC3 recruitment to phagosomes containing IgG-opsonized beads (Huang et al. 2009). Using inhibitors and NADPH oxidase-deficient cells, they demonstrated that reactive oxygen production by the NADPH oxidase is required for LC3's recruitment to Fcγ receptor-triggered phagosomes. Since Dectin-1 activates the NADPH oxidase in a manner very similar to Fcγ receptors, we examined whether reactive oxygen production is similarly required for β-glucan-induced LC3 recruitment. β-glucan-induced LC3 recruitment was completely blocked by inhibitors of the NADPH oxidase as well as in cells genetically deficient in the enzyme. Furthermore, we and others have previously shown that Syk is required for Dectin-1 to activate reactive oxygen production, so we examined whether the kinase is also required for Dectin-1-triggered LC3 recruitment. Inhibition or knockout of Syk blocked LC3 recruitment. While these data demonstrate that signaling through Syk and activation of reactive oxygen production are necessary for LC3-recruitment to Fcγ receptor and Dectin-1 phagosomes, the precise mechanism by which reactive oxygen promotes LC3II formation and recruitment has not yet been understood. Further, TLR signaling does not involve Syk and is generally not associated with NADPH oxidase activation in most situations, suggesting that the mechanism of TLR-induced LC3 association might be different.
In trying to understand the functional significance of LC3 recruitment to phagosomes, investigators have typically relied on the use of inhibitors and genetic mutants affecting upstream components of the autophagy signaling pathway including mammalian target of rapamycin, ATG5, ATG7 and ATG12 (Sanjuan et al. 2007; Lee et al. 2010). Interpretation of such studies with respect to the specific role of LC3 recruitment to phagosomes can be complicated by the fact that these proteins have been implicated in regulating diverse processes in addition to LC3II formation including roles in regulating apoptosis, metabolic regulation and nonfungal immunity (Levine et al. 2011). We, therefore, made use of cells specifically deficient in LC3 in order to understand the significance of LC3 recruitment to β-glucan-containing phagosomes.
Two isoforms of LC3 have been identified in mice, LC3α and LC3β. In macrophages and dendritic cells, LC3β appears to be the only isoform expressed (Ma et al. 2012). Mice lacking the gene for LC3β (Map1lc3b) have been made, and unlike many other autophagy gene knockouts that are lethal, the animals are healthy (Cann et al. 2008). Using fungi expressing the model antigen ovalbumin, we showed that dendritic cells lacking LC3β are significantly less efficient at processing and presenting fungal-derived antigens on major histocompatibility complex class II (MHCII) compared with wild-type cells. Phagosomes in cells lacking LC3β demonstrated reduced ability to accumulate MHCII molecules, suggesting that LC3 might be important in directing phagosome association with MHCII-containing vacuoles (Figure 1B). LC3β did not appear to be important in regulating cross presentation on MHCI.
These data are consistent with the above-discussed reports suggesting that LC3 recruitment to phagosomes might regulate phagosome maturation. Other studies have also associated LC3 recruitment with killing of ingested microbes and with clearance of apoptotic and necrotic cells, although these analyses have been done using inhibition or genetic deletion of upstream autophagy regulators which may participate in processes independent of LC3 (Sanjuan et al. 2007; Martinez et al. 2011). Further studies will be required to elucidate the specific role of LC3 in these processes.
In addition to triggering phagocytosis, respiratory burst and directing inflammatory responses that are essential for effective host defense against pathogenic fungi, recognition of β-glucan-containing particles by Dectin-1 directs LC3 recruitment to phagosomes. As discussed above, this process influences how fungi are handled by immune cells. We hypothesize that other innate immune receptors mediating recognition of other microbial sugars (e.g., mannans or chitin) are similarly poised to activate LC3 recruitment to phagosomes. Dectin-2, a pattern recognition receptor implicated in sensing fungal mannans, is expressed at the plasma membrane associated with the Fc receptor common γ chain and signals in a manner that is highly similar to Dectin-1 (Sato et al. 2006). Thus, it would seem logical to anticipate that this receptor might also be able to regulate LC3 recruitment to phagosomes, even though unlike Dectin-1, Dectin-2 has not been demonstrated to be phagocytic. As illustrated by data suggesting that TLR signaling can trigger LC3 recruitment to phagosomes, it is not clear that a receptor must be “phagocytic” in order to promote LC3 association. Further studies will be required to determine which microbial sugars and which receptors are able to modulate LC3 recruitment to phagosomes.
Recent studies have started to elucidate the signaling pathway which activates LC3 recruitment to phagosomes. As noted above, many upstream components of the autophagy signaling pathway including ATG5, ATG7 and ATG12 are also required for LC3 recruitment during LC3-associated phagocytosis. However, a couple of recent studies have demonstrated that components further upstream in the autophagy signaling pathway including ULK1, FIP200 and ATG13 are not required (Martinez et al. 2011; Henault et al. 2012). The data suggest that there are specific signals upstream of ATG5, ATG7 and ATG12 leading to LC3-associated phagocytosis that are distinct from the more widely studied pathway for autophagy. Future investigations are necessary to fully elucidate how microbial sugar recognition by receptors such as Dectin-1 drives LC3-associated phagocytosis and to define the role of this process in orchestrating effective immunity.
This effort was supported in part by a grant from the National Institutes of Health (R01 AI071116).
Conflict of interest
ITAM, immunoreceptor tyrosine-based activation motif; LC3, light chain 3; NFAT, nuclear factor of activated T cells.