Published online before print September 22, 2008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
INSERM U662, Institut Universitaire dHématologie, Hôpital Saint-Louis, Paris, France
1 Correspondence: Inserm U662, Hôpital Saint-Louis, 1 av Claude Vellefaux 75010, Paris, France. E-mail: catherine.gelin{at}univ-paris-diderot.fr
|
|
|---|
Key Words: dendritic cells cell-surface molecules
|
|
|---|
An efficient T cell response to antigen presented by the APC requires a two-way interaction. Such interaction takes place in a defined context, and any alteration that could intervene before or during the interaction could alter the outcome. Thus, the precise organization and the compartmentalization of the antigen-presenting proteins expressed at the cell surface and the associated signaling molecules are crucial for an appropriate response.
For a long time, the T cell was considered as the only cell active in initiating the adaptive immune response. However, although it has been neglected, the contribution of the APC is increasingly clear. For example, the duration of the dendritic cell (DC)–T cell interaction, which is known to be critical for T cell priming, depends on the degree of DC maturation [1 2 3 ].
This review will address regulation of antigen presentation by the MHC II and CD1 antigen-presenting molecules, which can be regulated at, at least, three distinct levels: first, the expression and structure of the antigen-presenting molecules; second, the extracellular environment and the signals of danger detected; and finally, a third level of regulation determined by the molecules associated with MHC II or CD1, the localization of MHC II and CD1 in specialized microdomains, and the trafficking pathways used in these molecules.
This review will begin in the endoplasmic reticulum (ER), where MHC II and CD1 are synthesized, and shows how they can reach the cell surface of APC. We then describe how the association of MHC II and CD1 molecules with cell-surface proteins, cytoskeleton, and membrane microdomains is necessary to ensure specific T cell responses. Finally, we explore the extracellular environment to show how it can influence the type of antigens presented.
|
|
|---|
CD1 looks like MHC I
Like MHC I molecules, the CD1 proteins, usually identified as MHC class I nonclassic proteins, are formed of a transmembrane heavy chain with three
-domains and are noncovalently associated to β2-microglobulin. They are encoded on chromosome 1 by five genes—CD1A, CD1B, CD1C, CD1D, and CD1E—display an intra-exon structure similar to MHC I genes, and encode proteins with significant homologies to MHC I and MHC II molecules. Thus, similarities and divergences of amino acid and nucleic acid sequences of MHC I, MHC II, and CD1 proteins have led to the concept that these molecules have diverged from a common ancestral gene [4
]. Moreover, MHC I and CD1 molecules, which are synthesized in the ER, share the chaperones calnexin and calreticulin, which are necessary for their correct folding [5
6
7
]. Another similarity between MHC I and CD1 molecules has been revealed in a recent study that shows that one of the CD1 isoforms, CD1a, can undertake a similar trafficking pathway to that of MHC I molecules [8
].
The function of CD1 proteins has remained unknown for a long period, as their initial description and thus, the CD1 family can be considered as a relatively "young" family of antigen-presenting molecules. Although MHC I and MHC II molecules are conformed to present peptides, the binding groove of the CD1 molecules is hydrophobic and enfolds alkyl chains of lipid antigens, thereby allowing interactions of polar chains with the TcR.
The first member of the CD1 family was identified at the cell surface of human thymocytes in 1979 [9
], but it was the work of Porcelli et al. [10
] that revealed the role of CD1 in antigen presentation. In 1994, Beckman et al. [11
] first proposed the concept that CD1 molecules can present lipids to specific T cells. Since then, numerous specific, CD1-restricted T cell clones have been isolated [12
13
14
15
], and different kinds of lipids presented by CD1 have been identified. These can be divided in two major classes of antigen: lipids derived in a great majority from the wall of Mycobacteria tuberculosis, such as mycolic acid, lipoarabinomanan, didehydroxymycobactine, or mannosyl-β1-phosphoisoprenoide, and lipids derived from Sphingomonas bacteria such as glycosphingolipids [16
, 17
] or a diacylglycerol from Borrelia burgdorferi [18
]; and self-antigens such as gangliosides GM1, GD1a, or GD1b, mannosyl-β1-phosphodolichol, sulfatide, or glycerophospholipids. These lipids are presented by CD1 proteins, which can be divided in humans into two groups: Group I molecules including CD1a, CD1b, and CD1c isoforms that are recognized by conventional
β T cells and required for the T cell response to naturally occurring self as well as foreign antigens [19
] and Group II, of which the CD1d isoform is the sole member and is recognized by T cells expressing an invariant V
24 (in humans) or V
14 (in mice) TCR [20
]. The CD1e isoform has nucleotide and amino acid sequences leading to a classification intermediate to Groups I and II.
CD1 acts like MHC II
MHC II and CD1 families are functionally close. Both molecules traffic via endosomal compartments to load antigen, and some CD1 and MHC II molecules are detected within the same compartments. This is the case in DC, which are uniquely capable of naïve T lymphocyte activation via MHC II or CD1 molecules. Notably, the trafficking pathways used by MHC II and CD1 molecules in DC have certain similarities [21
] in terms of endosomal compartment distribution and protein interactions with the invariant chain (Ii) of MHC II molecules. The following section will show how MHC II, CD1, and the Ii chain work together.
|
|
|---|
![]() View larger version (59K): [in a new window] |
Figure 1. MHC II and CD1 molecules coexist in endosomal compartments. (A) MHC II β-peptide complexes reach the surface after trafficking through MHC Ii-loading antigen compartments (MIICs). Three potential pathways are used by MHC II molecules to reach MIICs. The first and predominant pathway (Path 1) drives MHC II complexes directly through MIIC in an adaptor protein 1 (AP-1)-dependent mechanism. Second, part of the βIi complexes can be directed to the cell surface and internalized to endosomes before entering MIICs (Path 2). Finally, the third pathway, intermediate between Paths 1 and 2, is used by MHC II βIi complexes, which can pass through endosomes before entering MIICs (Road 3). Whichever path is taken, the βIi complexes reach MIICs, in which Ii is degraded to allow peptide loading on β complexes. The β-peptide complexes then reach the cell surface, where they can be recycled through endosomes. (B) CD1 molecules reach the surface before trafficking through MIICs. The major route taken by CD1 molecules after egress from the trans-Golgi network (TGN) is directly to the cell surface (Path 1). Once at the cell surface, CD1 molecules are recycled to different endosomal compartments to meet lipid antigens. The pathway taken is conditioned by the topography of the protein groove, the alkyl chain length of the antigen they will present, and their intracytoplasmic domain. Thus, CD1a molecules, which do not display any signaling motifs in their short intracytoplasmic domain, are predominantly detected at the surface and in recycling endosomes. The CD1c and CD1d molecules have a YXXZ tyrosine motif that allows their interaction with AP-2 and their access to late endosome (LE). The CD1b isoform is the only one to possess a YXXZ motif, allowing interaction with AP-3 and then their access to MIICs. After passage in their respective compartments, the CD1a, CD1b, CD1c, and CD1d lipid-loaded molecules can be recycled to the cell surface. The CD1e and CD1d molecules can use an alternative pathway (Path 2). The traffic used by the CD1e isoform is clearly different from the one described above: CD1e is mainly detected in TGN and can be mobilized to endosomes under DC maturation. A part of the CD1d molecules, which are associated to βIi complexes, can also be derived from TGN to MIIC (black arrow).
|
βIi complexes have also been detected at the B cell surface, and it remains unclear whether the complex can reach the cytoplasmic membrane directly after egress from the TGN or passes through endosomal compartments and recycles to the cell surface [27
]. The MHCII/Ii complex is directed to endosomes [28
, 29
] and then to immature lysosomes, where the Ii chain is proteolyzed [30
], and the Ii chain-associated protein (CLIP) fragment is replaced by an antigenic peptide [31
], or the complex reaches multivesicular MIIC antigen-loading compartments directly. This targeting is guided by the dileucine motif of the Ii chain. Whether the MHC II/Ii complexes pass through early endosomes, the main feature of MHC II trafficking in B cells is to leave the cell surface. Thus, in B cells, MHC class II molecules are functionally dedicated to the presentation of exogenous antigens internalized by the BCR in late endosomal compartments.
In DC, 50–80% of MHC II molecules are directed to the cell surface with Ii [32
, 33
] and are internalized rapidly to endocytic compartments to be degraded or to load peptide followed by cell-surface recycling. In epithelial cells, the endocytosis of
β/Ii complexes in epithelial cells is highly dependent on the Ii chain dileucine motif, which allows clathrin recruitment through AP-2 interaction [34
, 35
], whereas endocytosis of mature
β/peptide is regulated by polyubiquitinylation of the β-chain intracytoplasmic tail. In this context, it has been shown recently that the down-regulation of the E3 ubiquitin ligase membrane-associated RING-CH I (MARCH I) is a major biochemical event leading to MHC II surface stabilization during DC activation [36
].
CD1 uses a MHC II-like pathway
The trafficking of CD1 proteins in APCs is similar to that of MHC class II proteins, as both leave the ER by the secretory pathway to reach the cell surface. This notion is supported by kinetic studies of newly synthesized CD1b [37
] and CD1d [38
] at the cell surface but remains to be confirmed for other CD1 isoforms. Once CD1 molecules reach the plasma membrane,they are internalized and transported through endosomal compartments, within which they load lipid antigen, before returning to the cell surface, where they can encounter antigen-specific T lymphocytes. However, loading of self-antigens such as sulfatide and GM1 ganglioside can take place directly at the cell surface [39
], and it has been shown that this surface loading of antigen is necessary for correct folding and function of CD1a [40
].
The endocytosis of CD1 molecules from the cell surface appears to be largely dependent on the intracytoplasmic tail YXXZ tyrosine motif, allowing interaction with AP-2 [41 ]. CD1 proteins express this tyrosine motif with the exception of the CD1a isoform. Whereas CD1c and human CD1d cytoplasmic tails interact with AP-2 and thereby limit their expression to early and late endosomes, the CD1b and the murine CD1d tyrosine motif also interact with AP-3. This interaction gives CD1b and murine CD1d access to late endocytic compartments and to acidic lysosomal vesicles [37 , 42 43 44 45 ]. Thus, CD1 isoforms specifically reach all endocytic compartments and can sample a broad spectrum of lipid antigens. Therefore, the CD1 system seems to be dedicated to the surveillance of the endocytic compartments.
However, the CD1e isoform does not conform to this schema. First, this molecule is absent from the cell surface and is constitutively localized in the TGN of immature DC (iDC) [46 ]. Ubiquitinylation of CD1e allows its egress from Golgi and its entrance into the endosomal pathway [47 ]. During DC maturation, CD1e is found in the endosomal network [48 ], wherein cleavage leads to generation of the functional, soluble lysosomal form that is active in CD1b lipid antigen processing [49 ]. In this respect, CD1e plays a similar role to that of the HLA-DM molecule in HLA class II antigen presentation [50 ].
Ii regulates MHC II and CD1 trafficking
CD1 trafficking can also be regulated by the Ii chain association with MHC-class II. Such associations have been shown for the CD1d and CD1a isoforms. CD1d associates with Ii and MHC class II in the ER in mouse and human APC [38
, 51
]. Moreover, the Ii chain rescues an intracytoplasmic tail-truncated CD1d isoform from an exclusive surface localization and induces endosomal targeting [52
]. However, in contrast to MHC II, CD1d alone can also reach endosomal compartments, even in the absence of Ii. The functional consequences of Ii association with CD1d have been shown in the study of Jayawardena-Wolf and colleagues [38
], which demonstrates that CD1d trafficking in mice is regulated by two different pathways. Thus, in the absence of Ii (intrinsic pathway), CD1d molecules reach the cell surface directly and present antigens independently of endosomal localization. In the presence of Ii (extrinsic pathway), part of the CD1d molecules traffic first to the endosomal compartments before reaching the cell surface associated with antigens. The CD1d trafficking can also be influenced by cathepsins S [53
] and L [54
], proteases critical in Ii processing and cleavage.
The association of CD1 with Ii may be a general feature, as we have recently identified a CD1a/Ii complex at the surface of human iDC [55 ]. Ii can therefore associate with all of the antigen-presenting molecules, as MHC I/Ii complexes have been identified in the ER, and this is favored by the absence of antigenic peptide. Mutations of the CLIP peptide alter this interaction, suggesting that Ii interacts with MHC class I in a similar manner as for MHC class II. However, for CD1 and MHC class I protein, the physiological relevance of such interactions remains to be determined. Chen et al. [56 ], who demonstrated that CD1d associated with Ii could be dedicated to foreign antigen presentation, characterized CD1/Ii associations. The precise role of Ii in CD1 endocytosis requires further investigations, notably, to understand how the CD1/Ii association may influence the presentation of intra- versus extracellular antigens.
|
|
|---|
MHC II, CD1, and the Ii chain
The Ii chain (CD74) was described initially as a nonpolymorphic type II integral membrane protein [25
], which is a chaperone for MHC II molecules and a major actor for the correct folding and the functional stability of MHC II molecules [57
]. The association with Ii prevents MHC II molecules from premature loading of self-peptide in the ER and also participates in the sorting of MHC II toward the endocytic pathway. In addition to these activities, Ii transmits signals [58
] and acts as a receptor for the macrophage migration inhibitor [59
], as well as regulating the proteases that usually act in antigen degradation.
Ii can also be a chaperone for CD1 molecules. Ii alone or MHC II–Ii complexes are involved in CD1d trafficking, and several studies have demonstrated CD1d/Ii and CD1d/MHC II associations [38 , 51 , 53 ]. These associations have been documented on the cell surface and in late endosomal compartments in B cell lines and DC from mice and humans. The overall results obtained in human cells have led to the conclusion that CD1d can be sorted to MIICs by association with class II–Ii chain complexes and that human CD1d trafficking is regulated by the CD1d association with the Ii chain, independently of the tyrosine-based motif present in its intracellular domain. The recent and elegant study of Chen and colleagues [56 ] has demonstrated that in human, the Ii association with CD1d could participate in the selection of extracellular antigens, and CD1d-nonassociated Ii molecules would be dedicated to the presentation of autoantigens. This study reveals the difference with mice in which autoreactive NKT cell responses are dependent on CD1d lysosomal localization [52 , 60 ].
We have recently determined whether the CD1a isoform was associated with Ii in iDC. This hypothesis was based on the observation that although CD1a was observed at the cell surface and in early endosomes, this protein does not possess intracellular motifs that would allow access to endosomes. The association between CD1d and Ii has been described, and we have examined the potential association of CD1a with Ii at the cell surface of iDC. Biochemical and imaging studies revealed the cell-surface association of CD1a with several molecules, including MHC II, Iip33, Iip35, and CD9 molecules, and that Ii and lipid rafts are key regulators of the molecular organization and the immunological function of CD1a [55 ].
This Ii association with MHC II and CD1, observed at the cell surface, has potentially major consequences for the pathway used by these molecules for loading antigens.
MHC II, CD1, and other surface-associated proteins
Although signal-transducing functions mediated by MHC II or CD1a molecules have been documented [61
, 62
], the absence of known signaling motifs in the intracytoplasmic domain of these molecules has raised the question of how they can transduce signals and has led to the notion that MHC II- and CD1-associated proteins may be involved in this process.
MHC II molecules associate with the F-actin cytoskeleton, and the actin cytoskeleton regulates signaling via MHC II molecules as well as antigen processing and presentation [63
]. This association can be enhanced by the oligomerization of HLA-DR, and although neither the cytoplasmic domain nor the transmembrane domains were necessary, both improved the association of class II molecules with actin filaments [64
], which is regulated by amino acids present in the
- and β-chain [65
].
Members of the tetraspanin family are typically involved in the molecular organization of cells and facilitate functional, multimolecular complexes [66 ]. Tetraspanins are composed of four transmembrane domains, are broadly expressed, and regulate cell migration, fusion, and signaling events. However, and according to our current knowledge, these molecules do not have specific receptors or ligands. Two major features characterize these proteins: their ability to self-associate and to associate with other molecules (such as integrins, BCR, TCR, and diverse coreceptors) and their role in organizing proteins in signaling complexes at the cell surface [67 ]. This explains why many studies of MHC II- or CD1-associated molecules have detected tetraspanin molecules.
Among these tetraspanin associations on B cells, the CD23/CD81/MHC II complex regulates signaling through soluble immunoglobulin or MHC class II [68 ]. CD9, CD63, CD81, and CD82 associate in a surface tetraspanin network connected to HLA-DR and VLA integrins [69 ] at the surface of B cells.
Although a study in the THP-1 myeloid cell line revealed HLA-DR-associated β2 integrin CD18 [70 ], the molecular organization and associations of HLA-DR at the surface of primary monocytes remain poorly understood. Immunoprecipitation experiments revealed the association of MHC II with the ectoenzyme CD38 and the tetraspanin CD9 and that a portion of these molecules are associated in the same trimolecular complex within lipid rafts [71 ]. This association regulates the T cell response to bacterial superantigen, and engagement of the CD9 molecules led to increased monocyte–T cell interactions.
Under inflammatory conditions, recruitment and differentiation of monocytes replenish tissue-localized DC [72 , 73 ]. In addition to the known phenotypic changes, dynamic changes in cell-surface molecular associations may occur during monocyte differentiation to DC or macrophages. We have studied monocyte-to-DC differentiation to determine how these complexes evolve during activation/differentiation of monocytes to professional APC. The MHC II/CD9 association has been described in murine DC [74 ], and we have revealed the CD1a/MHC II/CD9 complexes at the surface of human DC and the specific CD1a/Ii p33 association in surface membrane lipid rafts [55 ].
MHC II, Ii, and tetraspanins
MHC class II molecules are present at the surface of the APC in different types of membrane domains including microdomains, where they associate with tetraspanin molecules. The key role of tetraspanin molecules in MHC class II function was demonstrated recently by the identification of a novel tetraspanin, MYPS, associated with MHC class II molecules and essential for the transduction of the MHC class II-mediated cell death signal [75
]. Localization of MHC class II molecules (irrespective of their antigen loading) in lipid-rich microdomains, known as membrane rafts [76
, 77
], optimizes T cell activation and polarization [78
, 79
]. It has been proposed that tetraspanins localize in specialized membrane microdomains enriched in a population of MHC class II molecules characterized by the CDw78 epitope of MHC II, which are loaded with a relatively homogeneous population of antigenic peptides [67
, 80
]. MHC II–tetraspanin complexes are detected in lipid raft microdomains of primary monocytes [71
]. Although initially, tetraspanin molecules were thought to localize to distinct microdomains, their localization within lipid-rich microdomains has also been documented [81
].
A more recent study has shown that Ii contributes to the HLA-DR tetraspanin organization. Although CDw78 was propsed initially as an epitope characterizing HLA-DR tetraspanin domains, it now appears that CDw78 binding identifies HLA-DR molecules, which have trafficked via lysosomes. As Ii expression is required for lysosomal trafficking of MHC II-peptide complexes, Ii is therefore also required for CDw78 expression [82 ].
|
|
|---|
![]() View larger version (44K): [in a new window] |
Figure 2. Three levels of regulation for antigen presentation in DC. MHC class II and CD1 molecules constitutively [1
] colocalize with membrane raft markers and [2
] associate with tetraspanin molecules within and outside DRMs [3
]. Ii, which is essential for appropriate MHC II trafficking to encounter antigen, is mainly intracellular, and a portion of these molecules localized with MHC II and CD1 within DRMs. Engagement of MHC II will reorganize the actin cytoskeleton and lead to membrane raft recruitment and detection of an Ii/MHC II/tetraspanin molecular complex. This organization could facilitate interaction with the CD4–T cells during antigen presentation by enriching MHC II complexes, which would interact with a range of affinities, and tetraspanin interactions, which would augment lateral tetraspanin and Ii interactions. These could therefore represent common mechanisms for enhancing antigen presentation.
|
Studies in B lymphocytes have implicated a membrane raft localization of MHC II in optimizing antigen presentation and particularly, under conditions of limited antigen availability [78 , 87 ]. This latter property could be a result of a preferential localization of peptide-bound MHC class II molecules in DRMs, as indicated by the higher fraction of SDS-stable MHC II complexes [83 ]. Lipid-rich microdomains were required to transport MHC II molecules to the site of interaction or immunological synapse formed with the CD4–T cells, and cholesterol depletion inhibits the translational diffusion of MHC II in the plasma membrane [88 ]. A further role for MHC II-enriched, cholesterol-rich microdomains in B lymphocytes was in maintenance of the stringency of the peptide-dependent interaction with the CD4 T lymphocyte [89 ].
CD1 and membrane rafts
DC are the most efficient APC, and recent studies confirmed the localization of MHC II molecules to membrane rafts and their role in antigen presentation [90
, 91
].
Members of the CD1 family also localize to detergent-resistant microdomains. Disruption of cholesterol-dependent microdomains inhibits lipid antigen presentation (
-galactosylceramide) by CD1d to NKT cells [92
, 93
]. A later study suggested that the lipid raft localization only influences antigen presentation involving the endocytic pathway [94
]. The role of cholesterol-dependent membrane rafts in CD1a-mediated sulfatide antigen presentation to
βT lymphocytes has been demonstrated [55
]. Similar to MHC II antigen presentation, the integrity of cholesterol-dependent microdomains was particularly important under conditions of limited availability of antigen. Moreover, the CD1a association with Ii is predominantly (but not exclusively) detected in lipid-rich fractions. As the Ii chain function is essential for antigen presentation by MHC class II molecules traversing the endocytic pathway, lipid raft localization may also be important for CD1a-mediated presentation of antigens requiring the endocytic pathway. The lipid raft localization of CD1a and of MHC class II may therefore be required for sorting of molecules associated with the Ii chain to the required compartment(s), although a surface-to-endocytic pathway would be favored for CD1a, whereas a Golgi-to-endocytic pathway would be favored for MHC class II. Moreover, this raises the question of whether CD1a and MHC II compete for Ii binding within lipid-rich microdomains. Similarities between the CD1 and MHC II pathways of antigen presentation include the requirement for the PKC
isoform in CD1d-mediated antigen presentation [95
]. Interestingly, the PKC
isoform has been implicated in MHC II signaling [96
97
98
] and colocalizes with MHC II in lipid-rich microdomains of mature DC and in iDC after MHC II engagement [86
]. PKC
is also enriched in B lymphocyte lipid rafts after antigen-specific T cell interaction [89
]. Acylated signaling proteins are commonly localized in lipid-rich microdomains, and such segregation controls their signaling activity [99
].
Membrane rafts and actin cytoskeleton
MHC class II-rich microdomains are enriched consistently in F-actin [87
, 100
, 101
], and the actin cytoskeleton is essential for the targeting of lipid-rich microdomains to the site of the immunological synapse [100
, 101
]. Membrane rafts and the actin cytoskeleton are therefore intimately associated. In T lymphocytes, the recruitment of membrane rafts to the immunological synapse was abrogated when actin cross-linking proteins known as filamins were knocked down using a small interfering RNA approach [102
]. B cell antigen receptor signaling induces dissociation of ezrin from lipid rafts and thereby, allows coalescence of lipid rafts at the B lymphocyte plasma membrane [103
].
Although the localization/recruitment of F-actin to lipid-rich microdomains is common to DC and B lymphocytes, the role of the actin cytoskeleton in antigen presentation by HLA-DR varies. The actin cytoskeleton has been documented as contributing to peptide presentation in B lymphocytes, whereas disruption of the DC actin cytoskeleton actually enhanced presentation of the same peptide [104 ]. Generally, the integrity of the actin cytoskeleton has been considered as having a positive influence in signaling, but this is not the case exclusively, and several examples of actin cytoskeleton restriction of cellular activation via diverse cell-surface receptors exist [105 106 107 ]. The mechanisms by which the actin cytoskeleton regulates MHC II-mediated antigen presentation remain to be determined.
|
|
|---|
This general scheme is obviously dependent on the extracellular environment. Indeed, iDC are sentinels detecting a variety of external stimuli such as pathogens, cytokines, apoptotic cells, and immune complexes, which trigger the entire presentation process. In this context, the activation of DC by microbial antigens, detected by their pathogen-associated molecular patterns (PAMPs), plays a decisive role in the primary immune response and induces increased MHC II and costimulatory molecule expression. These PAMPs are recognized by three families of proteins expressed on DC: the TLRs, C-type lectin receptors, and NOD-like receptors. Recognition of microbial components detected by the DC and their activation through TLRs are central to the induction of the adaptative immune response [108 ] and the orientation of Th cells into different effector subsets.
Apart from innate recognition of microbial stimuli, the lipid microenvironment can also influence the antigen presentation by CD1 molecules, and a range of molecules present in serum can potentially modulate DC function. Previous studies have shown that the peroxisome proliferator-activated receptor-
(PPAR
) regulates expression of genes linked to lipid metabolism [109
], and the Igs and serum lipids [110
] modify DC differentiation and antigen presentation. The activation of PPAR
in DC changes the expression pattern of the cell-surface receptor and regulates CD1 expression. This activation leads to reducing CD1a expression and enhancing CD1d expression, thus modulating the phenotype and function of DC and favoring activation of invariant NKT cells [111
, 112
]. Igs can also influence DC activation and differentiation and regulate CD1 expression. Monocyte-derived DC cultured in low concentrations of Ig display a phenotype characteristic of CD1 Group 1 molecules, and DC derived in high Ig concentrations, such as adult human serum, are CD1d+, and this modification depends on the activation of Fc
RIIa. These observations show that the concentration in lipids present in the extracellular environment "shape" the DC to express CD1 Group I or Group II molecules and thereby, induce activation of T or NKT cells [110
, 113
].
It appears that the CD1a-restricted T cell response is shaped by the repertoire of lipids present in the extracellular environment. It has been demonstrated that a portion of the CD1a molecules can be stabilized at the cell surface by exogenous lipids that maintain CD1a in a functional conformation [40 ]. One of the lipid ligands binding to CD1a is sulfatide, a self-antigen abundant in serum. When DC are cultured in low-serum conditions, surface expression and functional abilities of CD1a are reduced significantly without modifying their trafficking pathway or the expression of the CD1b, CD1c, or MHC II molecules. However, addition of glycosphingolipids, such as sulfatide or ganglioside GM1, restores the expression and functional ability of CD1a. This study therefore reveals a direct relationship between exogenous lipids present in the extracellular environment and CD1a surface expression and the capacity of CD1a to survey exogenous signals.
|
|
|---|
Dissection of the precise mechanisms regulating both pathways will allow development of more effective vaccination strategies and provide targets for modification of immunogenicity.
Received February 28, 2008; revised September 2, 2008; accepted September 2, 2008.
|
|
|---|
β+ T cells Nature 372,691-694[CrossRef][Medline]
14i NKT cells J. Exp. Med. 198,1133-1146
-galactosylceramide enhances humoral immunity to T-dependent and T-independent antigen in a CD1d-dependent manner Immunology 119,116-125[CrossRef][Medline]
is a critical regulator of CD1d-mediated antigen presentation Eur. J. Immunol. 37,2390-2395[CrossRef][Medline]
isoenzyme Int. Immunol. 14,935-942
stimulates antigen presentation by class II MHC in murine dendritic cells Int. Immunol. 19,719-732
)RI and lipid raft components are regulated by the actin cytoskeleton J. Cell Sci. 113,1009-1019[Abstract]
regulates the function of human dendritic cells primarily by altering lipid metabolism Blood 110,3271-3280
Blood 109,643-652
controls CD1d expression by turning on retinoic acid synthesis in developing human dendritic cells J. Exp. Med. 203,2351-2362
RIIa Blood 111,5037-5046
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||