Cell Signaling Pathways That Regulate Antigen Presentation

What allows cells to respond to stimuli in specific ways? A priori, one can envisage a stimulus and a cellular response. Unfortunately, in between that stimulus and response is a black box. An external or internal event triggers within a cell a cascade (linear or branched), resulting in the phosphorylation and/or dephosphorylation of specific proteins. As a consequence, these cell signaling pathways have effects upon or within the cell that result in a specific response. For example, the activation of these pathways can stimulate cell migration or arrest. This can be due to the polymerization or depolymerization of cytoskeletal proteins or the rearrangement of proteins or intracellular compartments. Another potential consequence of the activation of these pathways could be activation-induced cell death.

Studies to dissect the specific paths (i.e., proteins that are phosphorylated or dephosphorylated and consequent effects on downstream intracellular proteins) followed by cell signaling pathways have helped us to understand many ways in which cells react to the environment, including infections by pathogens. Because this is a Brief Review, my comments are focused on cell signaling pathways that control the presentation of Ag to conventional and unconventional T cells.

Why would cell signaling pathways even be important for the immune system? Certainly, APCs need to be able to capture and internalize Ag in various forms and by various routes (1–5). This could be by phagocytosis, pinocytosis, or infection by a pathogen or even taken from within a cell as an endogenous protein or lipid. Ag-loaded molecules need to be able to be transported intracellularly along the cytoskeletal network to their ultimate expression on the cell surface for recognition by T cells. That movement needs direction; this is provided by cell signaling pathways.

APCs primarily use three pathways of Ag presentation; Ags are presented by MHC or MHC-like molecules (1–5). The first pathway involves MHC class I molecules (2, 6–18). In the cytosol, endogenously synthesized polypeptide chains are threaded into a barrel-like structure called the proteasome, which contains a variety of proteolytic enzyme activities. As the polypeptide is cleaved, a diverse array of peptides consisting of ∼9–15 aa is generated. These peptides are delivered into the endoplasmic reticulum by the TAP. Upon being loaded onto peptide-receptive MHC class I molecules, endoplasmic reticulum–resident peptidases cleave amino terminal amino acids, resulting in the 8–9-aa peptides that are usually found associated with MHC class I molecules. The peptide-loaded MHC class I molecules are transported through the Golgi to the cell surface, where they are recognized by CD8⁺ T cells. The second Ag-presentation pathway involves MHC class II molecules (17, 19–31). Rather than being synthesized intracellularly within an APC, Ags are taken up by phagocytosis (or pinocytosis) and delivered to late endocytic compartments where they are processed into longer (e.g., 15–20) peptide-sized fragments than those loaded onto MHC class I molecules. Initially, MHC class II molecules are complexed with the invariant chain (Ii). The CLIP portion of the Ii prevents peptide loading until the MHC class II molecules traffic to late endocytic compartments (e.g., MHC class II compartment). It is here that an antigenic peptide replaces the Ii’s CLIP. The newly loaded MHC class II molecule is transported to the cell surface where it is recognized by CD4⁺ T cells. The third Ag-presentation pathway does not involve peptide Ag presentation. CD1 molecules are MHC class I–like molecules that generally present lipid Ags to invariant, relatively oligoclonal, or even diverse T cells. These lipids can be of microbial origin or from mammalian cells themselves (5, 32). The CD1 family of Ag-presenting molecules consists of two members (based on the human CD1 molecules): group 1 includes CD1a, CD1b, and CD1c molecules, whereas group 2 contains CD1d as its sole member (33). Each of these molecules differs in how they traffic intracellularly and, thereby, how they acquire Ag.

For the sake of completeness, I will note that there are the very interesting MHC class I–like MR1 molecules, which present microbial vitamin B–derived metabolites to a novel T cell subpopulation called mucosal-associated invariant T cells (34, 35). Because so little is known about the cell signaling pathways regulating their function, they are not discussed in this article.

The presentation of specific Ags to T cells by cell–cell interactions results in the production of various cytokines that, in turn, stimulate APCs. For T cells to be stimulated by APCs, cell signaling pathways within the APCs themselves need to be activated; this results in the proper Ags being processed and loaded onto the appropriate Ag-presenting molecule to be expressed on the surface. This is absolutely required for Ag-specific T cells to be activated and perform their effector cell functions.

Which are the most widely studied cell signaling pathways that control classical (i.e., MHC class I and II) and nonclassical (e.g., CD1d) Ag presentation and what are the mechanisms (if known) by which they do this? One of the best understood cell signaling pathway families is that mediated by MAPK (Fig. 1, left panel). Following a cell stimulus, a MAPKKK is phosphorylated which, in turn, phosphorylates a MAPKK. Finally, the MAPKK activates the MAPK, which is transported into the nucleus to mediate its function. The MAPK family consists of three main members: p38, JNK, and ERK (reviewed in Ref. 36). Upon receiving a cell stimulus (e.g., infection with a pathogen or exposure to inflammatory cytokines), the subsequent activation of the MAPK pathways results in inflammation, cell growth, differentiation, or apoptosis. Consequently, the activation of MAPK pathways has the potential to impact the ability of a host’s immune cells to appropriately respond to an infection. Of these three pathways, p38 has been studied the most with regard to Ag processing and presentation.

MAPK p38 can be phosphorylated by two upstream kinases: MKK3 and MKK6 (37). For macrophage (MΦ) maturation (20) and production of IL-12 (38), p38 is required via its stimulation by MKK3 (38). This occurs because MKK3 indirectly activates the IL-12 p40 promoter through its phosphorylation of p38 (39). In MΦs, anisomycin (a p38 activator) enhances LPS-induced IL-12 production; this is blocked by the p38-specific inhibitor SB203580 (39). Blocking p38 also impairs dendritic cell (DC)/T cell clustering, which can, of course, reduce effector T cell activation (39). Interestingly, p38 inhibits (whereas ERK promotes) the differentiation of monocytes into DCs (40). As discussed below, p38 and ERK have similar reciprocal control over CD1d-mediated Ag presentation (41–43). Unstimulated DCs have a basal level of p38 that is enhanced following stimulation with anisomycin, a drug that activates p38 and JNK (44). In human (44) and mouse (45) DCs, activation of TLR4 results in the phosphorylation of p38. TLRs have a variety of known effects on APCs; these are discussed later. The other MAPKK that can activate p38, MKK6, increases the APC activity of Langerhans cells by stimulating the NF-κB pathway (46).

MAPK p38 was shown to be important for Ag presentation by classical MHC class I and II molecules. Ag uptake is increased in DCs treated with cyclophin A, which results in p38 activation (47). p38 (and ERK) activation by osteoprotegerin ligand increases costimulatory molecule expression (48), readying APCs for interaction with Ag-specific T cells. In the context of Ag presentation by MHC class I, p38 can help or hinder. For example, p38 phosphorylation following CD40L engagement results in the activation of DCs and expansion of HIV-specific memory CD8⁺ T cells (49). It was presumed that the expansion of these T cells is MHC class I dependent; however, it was shown recently that SIV peptides can be presented to CD8⁺ T cells by MHC class II molecules (50). I indicated above that TLR engagement activates p38 in DCs (44, 45); this also results in enhanced Ag presentation by DCs to antitumor T cells via MHC class I molecules (51). In contrast, there is increased Ag uptake for cross-presentation by MHC class I molecules in a Japanese encephalitis virus model when p38 is inhibited (7); thus, p38 appears to be a negative regulator in this model.

p38 activation in DCs (52) and its induction by fragments of the food product gliadin (53) increase costimulatory and MHC class II molecule expression, which, in turn, enhance Ag presentation to CD4⁺ T cells. This is in contrast to one report, in which the role of p38 in MHC class II–mediated Ag presentation by multiple myeloma patient DCs was studied. In that system, p38 activation inhibited Ag presentation by MHC class II molecules, concomitant with reduced costimulatory molecules and MHC class II expression on the cell surface (54).

We have done extensive analyses on signal-transduction pathways that can regulate lipid Ag presentation by CD1d. It is well known that viruses (along with other external stimuli) activate p38 in cells (43). We analyzed the role of p38 in regulating CD1d-mediated Ag presentation following infection with vaccinia virus or vesicular stomatitis virus and found that it was inhibitory; blocking p38 actually increased Ag presentation to NKT cells (42, 43). However, this does not only occur in a virus infection. Inducing apoptosis in CD1d⁺ APCs stimulates p38 activation; not surprisingly, these cells are poorer stimulators of NKT cells (41). Interestingly, another group looked at Ag presentation by group 1 CD1 molecules. As we (55) observed with CD1d, p38 is inhibitory in that system as well. Notably, p38 has an NKT cell–intrinsic negative effect, because treatment of invariant NKT cells with a p38-specific inhibitor impairs their stimulation by the CD1d-binding (and NKT cell–stimulating) glycolipid α-galactosylceramide (56) or by an anti-CD3–specific mAb (57). Therefore, p38 can affect the APC and T cell sides of a host’s immune response.

As indicated above, JNK has effects on many of the same cellular functions as p38, although there is not total overlap (37). Certainly, analyses in DCs found that, like for p38, anisomycin induces JNK in human monocyte-derived DCs, although at higher concentrations (39). Unlike p38, there is no detectable JNK in human DCs (39) at baseline, and TLR4 activation does not stimulate JNK in murine DCs (45). Does JNK have any role in Ag processing and presentation? It was shown that JNK2-deficient mice generate more CD8⁺ T cells in response to IL-2 (58), and more antiviral CD8⁺ cells are found in these mice upon infection with lymphocytic choriomeningitis virus (59). This suggests that JNK can have a negative effect on MHC class I–mediated Ag presentation; alternatively (or additionally), this could be T cell intrinsic. In line with those two articles, we found that JNK2 is a negative regulator of Ag presentation by CD1d (J. Liu, R.M. Gallo, M.A. Khan, A.K. Iyer, I.A. Kratzke, and R.R. Brutkiewicz, manuscript in preparation). Studies on the JNK signaling pathway in the control of Ag presentation by classical and/or nonclassical MHC molecules are still limited.

A few studies asked whether the ERK MAPK pathway is important for MΦ or DC maturation or whether it can be activated in either cell population. For example, one study suggested that LPS or GM-CSF induces ERK in MΦs, albeit with different kinetics: the latter induces ERK activation much more quickly (5 min versus 15 min) (60). LPS activates ERK in MΦs (44, 61); however, it apparently does not do so in DCs (62). In terms of MΦ maturation, it is ERK and p38 dependent (20). In contrast, ERK promotes, but p38 impairs, the differentiation of human monocytes into DCs (40). We (43) reported that, in the context of lipid Ag presentation by CD1d, ERK is a positive regulator.

As indicated above, a number of cytokines can activate MAPK signaling, with an impact on APC-maturation or Ag-presentation abilities. In some cases, such as with TGF-β, although the cytokine can activate MAPKs, which negatively regulate CD1d-mediated Ag presentation (41–43) (J. Liu et al., manuscript in preparation), the ultimate effect on Ag presentation by CD1d in the TGF-β model appears to be (at least) p38 independent (21).

The protein kinase C (PKC) family members are Ser/Thr kinases that consist of three main subgroups: conventional/classical (PKCα, β [I and II], and γ); atypical (PKC ζ and λ/ι); and novel (δ, ε, and θ) (reviewed in Ref. 63). Within these subgroups are 10 kinases that play distinct roles in the regulation of gene expression and cell proliferation. The activation of PKCs modifies them from a quiescent cytosolic form to an active membrane-associated form (64). Beyond their role as cytoplasmic signal-transduction molecules, there is evidence to suggest that they can also serve as nuclear kinases. In terms of the control of Ag presentation, PKC isoforms were shown to promote Ag presentation by MHC class II molecules. In particular, the use of PKC activators, such as bryostatin, which activates PKC-α, δ, and ι, or PKC activation by phorbol esters and ionomycin increases the surface expression of MHC class II molecules and, thereby, increases Ag presentation (27, 65, 66). The use of PKC-specific (i.e., PKC-α and/or PKC-δ) inhibitors by our laboratory (67) and other investigators (23, 27) further supports the positive correlation between PKC-δ activation and Ag presentation by MHC class II molecules. Interestingly, we (67) found that Ag presentation by CD1d, an MHC class I–like molecule that traffics intracellularly more like MHC class II molecules, is promoted by PKC-δ. In contrast to MHC class II and CD1d, Ag presentation by MHC class I molecules is not impacted by PKC-δ (27, 67).

In general, when one thinks about what constitutes the innate immune response, the first thing that often comes to mind is a response via TLRs (reviewed in Ref. 68). The TLR family of molecules consists of up to 13 structurally similar molecules in mammals that are homologs of the Drosophila toll gene product (69). TLRs are molecules on the cell surface and/or in intracellular (e.g., endocytic) compartments (i.e., TLR3, TLR7/8, TLR9) to which microbial or viral products bind; this results in the activation of various cell signaling pathways, including MAPK and other kinases (70). The majority of TLRs (TLR3 is the exception) use the MyD88 adaptor molecule for signaling down the TLR-activation cascade (68, 70). These responses can have important effects on Ag presentation by classical MHC molecules (17), as well as by CD1d (71–76). Globally, the binding of a ligand to its specific TLR can regulate immunodominance; this was shown in the context of the Toxoplasma gondii–encoded protein, profilin, via MHC class II molecules (77). In contrast, TLR7/8 agonists can impair the differentiation of monocytes into DCs (78). Thus, activation via TLRs can promote (or inhibit) APC function.

In terms of MHC class I–mediated Ag presentation, TLR engagement can enhance (or have a minimal effect on) overall surface MHC class I expression (79, 80). However, ligands for TLR2 and other TLRs can impair conventional Ag presentation by MHC class I (80, 81). TLR signals can also have very important regulatory (albeit not direct) roles in triggering the delivery of MHC class I molecules to phagosomes for cross-presentation. These signals result in the IκB-kinase 2–mediated phosphorylation of phagosome-associated SNAP23, which stabilizes SNARE complexes, ultimately facilitating cross-presentation (11). Similar to endogenous Ag presentation by MHC class I, cross-presentation can also be reduced upon TLR engagement (7, 15, 18). Alternatively, cross-presentation can be enhanced via the binding of TLR2-specific (13, 82), TLR4-specific (18), TLR7-specific (8), or TLR9-specific (83) agonists. Differences in terms of the activation versus inhibition of cross-presentation are likely due to the timing of APC exposure to the TLR ligands compared with the cross-presented Ags.

Ag presentation by MHC class II molecules is also differentially affected by the engagement of distinct TLRs. Interestingly, in some cases, it seems to be a means of immune evasion by a pathogen. For example, 19- and 24-kDa lipoproteins from Mycobacterium tuberculosis can inhibit MHC class II expression and Ag processing via TLR2 signaling (25, 84, 85). In contrast, a variety of TLR ligands (including those for TLR2) can upregulate MHC class II molecules on microglia (29). Ligands for TLR1/2, TLR4, TLR7, and TLR9 were shown to enhance the ability of APCs to present Ag-85B of bacillus Calmette-Guérin to CD4⁺ T cells specific for that Ag via the upregulation of MHC class II molecules (86). Moreover, the stimulation of TLR signaling and delivery of a TLR-specific ligand (in this case, the TLR4 ligand LPS) into phagosomes can contribute to the generation of peptide/MHC class II complexes; this is as a means to segregate self-peptides versus nonself-peptides in a phagosome-autonomous manner (22). This delivery of TLR4 ligands is due to adaptor protein-3–dependent transport of TLR4 from endosomes to phagosomes (87). Sometimes, MHC class II–mediated responses, in this case, flagellin-specific CD4⁺ T cell responses, can be enhanced in a TLR5-dependent (but nonconventional TLR signaling pathway) manner (26). In that study, DCs from MyD88-deficient mice increased flagellin-specific CD4⁺ T cells in a TLR5-dependent manner that was comparable to DCs from wild-type mice (26). This response is regulated by CD103⁻CD11b⁺ DCs (88). Type I IFNs can work with TLR ligand–mediated signals to generate type B peptide/MHC class II complexes (pMHCs); type B pMHCs are formed in early endosomes from exogenous peptides (89). In fact, type I IFNs are very important for such peptide generation because DCs deficient in the receptor for type I IFNs are impaired in their ability to generate type B pMHCs (89). However, in the context of DC maturation, although TLR ligands can inhibit MHC class II synthesis and presence in intracellular compartments, type I IFNs can prevent this from occurring (30).

The activation of TLR signaling pathways by TLR-specific ligands also was investigated in terms of its effect on the CD1d/NKT cell axis. Mimicking APC stimulation of NKT cells (i.e., a panel of murine NKT cell hybridomas) using anti-CD3 and IFN-α upregulates TLRs on the cell surface (75). Moreover, the exposure of NKT cells to a variety of TLR ligands enhances NKT cell production of IFN-γ, IL-4, and TNF-α (75). However, another study indicated that, although human NKT cells express all TLRs (except TLR8), they are not activated when directly exposed to TLR ligands; yet, they are stimulated when TLR ligands are added to total PBMCs (73). The differences could simply be due to the fact that these studies analyzed mouse NKT cell hybridomas (75) as opposed to normal human NKT cells from PBMCs (73). Culture of murine mononuclear cells with the CD1d-binding glycolipid α-galactosylceramide and polyinosinic-polycytidylic acid (a TLR3 ligand) resulted in NKT cell activation in a model of airway inflammation (76). Certainly, the effects observed in that study could be NKT cell and/or APC specific. Murine bone marrow–derived DCs exposed to LPS or infected with Salmonella typhimurium were able to stimulate NKT cells at a high level. This suggests that APC TLR4 signaling enhanced Ag presentation by CD1d molecules (71). Additionally, TLR4 was shown to work with Nod1 and Nod2, two members of the Nod-like receptor family that are cytosolic pattern-recognition receptors. In that situation, LPS treatment of bone marrow–derived DCs resulted in IFN-γ production by NKT cells (74). An in vivo infection with S. typhimurium (which contains LPS) can activate invariant NKT cells (71, 72) but does so without the need for a CD1d-presented lipid Ag (72).

The JAK/STAT signaling pathway is activated by most cytokines involved in the development and regulation of the host’s immune response (90). There are four receptor-associated JAKs that, upon phosphorylation of the receptor they are bound to, recruit one of seven STATs to the receptor to be phosphorylated by that JAK. This process results in the dimerization of the p-STATs, which are translocated into the nucleus where they regulate the expression of a variety of genes (91).

JAK/STAT signaling upon activation by IFN-γ is critical for MHC gene expression and, ultimately, expression of MHC class I and II molecules on the cell surface (90, 91). Of importance, it was shown that p-STAT1 facilitates human MHC locus chromatin remodeling as a first step for the subsequent expression of HLA genes (92). Impairing the ability of IFN-γ to stimulate JAK/STAT signaling and, thereby, preventing MHC gene expression, is one mechanism by which viruses can evade recognition by the host’s immune response (16). For example, MHC class I and/or class II molecules can be targeted by herpesviruses (19, 28, 93) or influenza virus (31). In each of these cases, it is likely that a virus-encoded protein(s) prevented IFN-γ–induced activation of the JAK/STAT pathway. Similarly, immune evasion by solid tumors can be due to an impairment of JAK/STAT signaling that prevents the upregulation of MHC class I molecules (12, 14) or TAP1 expression (9).

In contrast to the immune-evasion examples presented above, activation of JAK/STAT signaling can have beneficial effects. For example, IL-10, which activates a JAK1/STAT3 complex, inhibits DC maturation by mesenchymal stem cells (94). An example for which this could be important is in the prevention of NKT cell–mediated colitis. Here, cross-linking CD1d on colonic epithelial cells results in STAT3 activation; the transcription of the il-10 gene (by STAT3) and consequent IL-10 production control inflammation in those cells (95). Recently, we (96) showed that STAT3 is essential for CD1d-mediated Ag presentation because of its ability to regulate the transcription of UDP glucose ceramide glycosyltransferase, an enzyme involved in the first step of glycosphingolipid biosynthesis, from which the natural ligands of CD1d are derived. The critical nature of STAT3 in the CD1d/NKT cell axis is evident in patients with loss-of-function mutations in STAT3. Those patients have a significantly reduced number of NKT cells (97).

Rho GTPases are well known for mediating intracellular protein traffic (reviewed in Ref. 98). This is due to their control over the generation of F-actin. Thus, Rho GTPases activate the Rho kinase (ROCK) which then phosphorylates the LIM kinase, which then activates cofilin. Normally, cofilin, in its nonphosphorylated form, prevents the polymerization of actin. The phosphorylation of cofilin by LIM kinase permits F-actin formation (99, 100). We found that F-actin actually impairs Ag presentation by CD1d (101). When we disrupted the actin cytoskeleton with cytochalasin D, Ag presentation was enhanced. Thus, ROCK is a negative regulator of CD1d-mediated Ag presentation. Interestingly, we found that ROCK actually promoted Ag presentation by MHC class II molecules (101). This was another example in which we have found specific cell signaling pathways that differentially regulate Ag presentation by MHC class II versus CD1d (21, 96, 101).

Knowing that ROCK impairs Ag presentation by CD1d via the polymerization of actin is only part of the mechanism by which this occurs. How does actin actually do it? Recently, it has been reported that actin segregates CD1d nanoclusters on the cell surface via direct interaction between actin and the CD1d cytoplasmic tail (102). This interaction keeps them at a certain density on the surface. Allowing the formation of larger nanoclusters (by blocking the ability of CD1d’s cytoplasmic tail to bind to actin) results in enhanced NKT cell activation—exactly the same observation we made when the actin cytoskeleton was disrupted in our system by cytochalasin D (101).

For the most part, cell signaling pathways appear to be important in inflammatory diseases (including cancer), which have a clear–or likely–immune component, where Ag processing and presentation would come into play. Of the MAPK, upstream ERK pathway components (i.e., Ras-Raf-ERK) are more involved. Either Ras or Raf contribute to malignant transformation and are thereby targeted with Ras- or Raf-specific inhibitors (103–106). Moreover, blocking ERK in esophageal and gastric cancers, or in melanoma (in the context of immunotherapy), results in an increase in MHC class I molecules (107–109).

For the p38 MAPK, inflammation was shown to increase tumor-induced (and immunosuppressive) myeloid-derived suppressor cells; the activation of p38 is believed to be important for their survival (110). For the treatment of rheumatoid arthritis, p38 inhibitors have been considered (111). Moreover, p38 is also believed to be a culprit in the development of acquired immune deficiency syndrome and HIV-associated neurocognitive disorders in HIV patients, which could have an Ag presentation component (112).

An important aspect of p38 signaling needs to be taken into account when considering using inhibitors against that MAPK: there is reciprocal cross-talk between p38 and STAT3 [reviewed in (113)]. In the context of the CD1d/NKT cell axis, this could be critical in the disease that one is attempting to mitigate, as p38 is a negative (whereas STAT3 is a positive) regulator of lipid Ag presentation by CD1d (41–43, 96, 97, 101).

The JAK/STAT pathway (in particular, STAT3) is also important in cancer and inflammation. In a recent study of patients with nonrefractory Hodgkin’s lymphoma being treated using a programmed cell death-1–blockade immunotherapy paradigm, p-STAT3 was detected in the nucleus of Reed-Sternberg cells (the presence of which is necessary for the diagnosis of Hodgkin’s lymphoma) (114). STAT3 is activated in Behçet’s disease, an inflammatory disorder resulting in vasculitis (115). In contrast, there appears to be attenuated IFN-α–induced STAT3 signaling in DCs from patients with Crohn’s disease, and IL-10 induces enhanced STAT3 activation, which could impact the DC’s ability to present Ag (116). Relating Crohn’s disease effects to the various cell signaling pathways in this Brief Review, there appears to be an association with an ICOSL loss-of-function mutation resulting in reduced cell signaling in these patients (117); this likely impacts immune homeostasis.

Lastly, signaling via TLRs has been predominantly associated with infectious diseases (118–124); however, a variety of single-nucleotide polymorphisms in human TLRs have not only been associated with infectious and inflammatory (including autoimmune) diseases, but also with cancer (121, 123–126). As such, TLRs have been looked at as potential targets for immune-based therapy against infectious diseases (118, 119), as well as sepsis-associated pathology (118). Moreover, another approach that is being considered is the specific targeting of TLR-associated adaptors that are negative regulators of TLR signaling (127). This is reminiscent of the CTLA-4 and inhibitory receptor programmed cell death-1 targeting of antitumor T cells that has recently shown promise in clinical trials with cancer patients (128, 129). This approach effectively removes inhibitory signals coming from the APCs to the effector T cells.

Although cell signaling pathways have been studied for a number of years, very little has been focused on Ag processing and presentation and how they affect APCs and/or the T cells that recognize them. Nonetheless, Fig. 2 summarizes what I described in this Brief Review in terms of how these cell signaling pathways affect MHC class I, MHC class II, and/or CD1d. To help the readers, I also included a table (Table I), indicating the reports cited in this study, that have studied these specific cell signaling pathways and Ag presentation. As is clear from the work done so far, the cell signaling pathways described above have at least some impact on various components of a host’s immune response; these could potentially be exploited for adding new weapons to the arsenal against infectious and autoimmune diseases, as well as cancer. We have just begun to scratch the surface.

I thank the members of my laboratory, past and present, who have made many important contributions to our understanding of how cell signaling pathways regulate Ag presentation by CD1d. I thank Jean Liu and Abhirami Iyer for careful reading of the manuscript and I also thank Abhirami Iyer for the idea regarding a summary figure for this review.

The author has no financial conflicts of interest.