• Antigen loading;
  • Antigen presentation;
  • CD1;
  • Lipid recognition


  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. Supporting Information

Self-glycosphingolipids bind to surface CD1 molecules and are readily displaced by other CD1 ligands. This capacity to exchange antigens at the cell surface is not common to other antigen-presenting molecules and its physiological importance is unclear. Here we show that a large pool of cell-surface CD1a, but not CD1b molecules, is stabilized by exogenous lipids present in serum. Under serum deprivation CD1a molecules are altered and functionally inactive, as they are unable to present lipid antigens to T cells. Glycosphingolipids and phospholipids bind to, and restore functionality to CD1a without the contribution of newly synthesized and recycling CD1a molecules. The dependence of CD1a stability on exogenous lipids is not related to its intracellular traffic and rather to its antigen-binding pockets. These results indicate a functional dichotomy between CD1a and CD1b molecules and provide new information on how the lipid antigenic repertoire is immunologically sampled.




brefeldin A


cytochalasin D


glycosphingolipids: HS: high serum


low serum


Langerhans cells


phosphatidic acid












  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. Supporting Information

The immune system has evolved different pathways to present antigens to T cells, including presentation of peptides by MHC molecules and presentation of lipids by CD1 molecules 1, 2. The two pathways complement each other and contribute to elimination of infectious agents and immunoregulation.

CD1a, b, c and d molecules present a variety of microbial and self-lipids to T cells 3. CD1 ligands are made of a polar head group, interacting with the TCR and of a hydrophobic tail, responsible for CD1 binding as documented by the analysis of the crystal structure of CD1a, CD1b and CD1d 4.

Several studies have addressed the function of lipid-reactive cells, the structure of lipid antigens and the intracellular trafficking of CD1 molecules. However, little is known about the maturation of CD1 molecules, their mode of interaction with antigenic lipids, and the mechanisms of antigen loading 3.

In a previous study, we demonstrated that CD1a and CD1b molecules reach the cell surface and are immunologically competent in the absence of de novo synthesis of self-glycosphingolipids (GSL) 5. However, CD1 molecules may associate during their assembly with endogenous lipids different from GSL such as phospholipid phosphatidylinositol 6. A second possibility is that newly synthesized CD1 molecules leave the ER in association with stabilizing chaperones that guide CD1 traffic to other cellular compartments in which the glycolipid antigens are loaded. One example is the association of CD1d with the invariant chain, which is not absolutely required for refolding and maturation of this CD1 molecule, but nevertheless contributes to trafficking in a late endosomal compartment, likely a main station of CD1d loading 7, 8.

A third possible scenario is that mature CD1 molecules bind to, and are stabilized by, exogenous lipids on the cell membrane. An important distinction from MHC class I molecules is that recombinant soluble CD1a and CD1b molecules successfully refold in vitro in the absence of added ligands and remain available for ligand binding 9, 10, suggesting that exogenous ligands might bind to CD1 molecules on the cell surface. In this case, serum lipids capable of binding to CD1 molecules would be the most abundant source of CD1 ligands.

Here, we show that CD1a, but not CD1b, requires binding to exogenous lipids to remain properly folded and immunologically functional. Thus, the repertoire of CD1a-bound ligands is continuously shaped by the repertoire of ligands present in the extracellular environment.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. Supporting Information

CD1a expression on the cell surface is affected by serum deprivation

To investigate whether exogenous ligands present in the serum participate in CD1 maturation and stabilization on the cell surface we cultivated MOLT-4 cells, which express both CD1a and CD1b molecules, in medium containing very low amounts (0.2%) of FCS (LS) for 24 h and then performed immunofluorescence analysis. Under these LS conditions, CD1a expression was detectable at about 50% of that found under control (10% serum, HS) culture conditions. CD1b, MHC class I molecules (Fig. 1A), CD71 (transferrin receptor) and MHC class II molecules (data not shown) remained unchanged, thus ruling out the possibility that culture in LS induces a general reduction of surface protein levels. Comparable results were obtained with human DC and B cells transfected with CD1A or CD1B genes (data not shown), thus also excluding an effect specific to the MOLT-4 cells. CD1a surface expression in LS was analyzed with nine CD1a-specific mAb (Fig. 1C). Reduction of expression levels ranged from 3 to 56%. The differences observed with various CD1a-specific mAb infers that CD1a loss in LS is not due to its internalization and suggest that different CD1a epitopes are no longer available for mAb recognition, likely due to conformational changes. These findings show that CD1a, but not CD1b, requires serum components to maintain an unaltered expression on the cell surface.

thumbnail image

Figure 1. CD1a surface expression is altered in LS. (A) MOLT-4 or (B) freshly isolated human CD4+CD8+ thymocytes were cultured in LS (A and B, dotted lines), in HS (A and B, thin lines) for 24 h or analyzed immediately after isolation (B, bold lines) for surface expression of CD1a, CD1b, MHC class I (OKT-6, WM-25, W6/32 mAb, respectively). Background staining is indicated by broken lines. (C) MOLT-4 cultured for 24 h in HS (white bars) or LS (black bars) were stained with various anti-CD1a mAb. Numbers indicate percentage of CD1a modulation as compared to HS conditions. BBM.1 is specific for human β2m, C15 is a negative control mAb.

Download figure to PowerPoint

To asses whether the serum dependence of CD1a high expression is also observed ex vivo, freshly isolated human CD4+CD8+ thymocytes were analyzed. CD1a and CD1b expression on fresh thymocytes was investigated either immediately or after 24 h of culture in HS or LS. Incubation in LS and not in HS induced a 50% reduction of surface CD1a, without decreasing expression levels of CD1b or MHC class I (Fig. 1B). Thus, CD1a molecules complexed with ligands captured in vivo in the thymus also undergo alterations in LS.

Surface CD1a molecules are the ones affected by serum deprivation

To investigate whether surface, recycling, or de novo synthesized CD1a molecules are altered in LS conditions, MOLT-4 cells were cultivated in LS in the presence of brefeldin A (BFA) or cytochalasin D (CHD). Amongst its effects, BFA blocks protein egression from the trans-Golgi network, thus affecting surface expression of de novo synthesized proteins. CHD inhibits fluid-phase, receptor- or caveolae-mediated uptake of extracellular and transmembrane molecules 1113. In the presence of BFA or CHD, CD1a continued to be altered after 24 h of culture in LS (Table 1). These results show that CD1a alteration occurs in the absence of de novo synthesized proteins and protein internalization, and suggest that CD1a molecules already present on the cell surface are affected.

Table 1. Effect of BFA and CHD on the expression of CD1a, CD1b and MHC class Ia)

Surface molecule


  1. a) MOLT-4 were cultured in HS or LS, without or with inclusion of BFA (1 µg/mL) or CHD (10 µg/mL). After 24 h cells were labeled with the indicated mAb and analyzed by FCM. Results are expressed as percentage of MFI of the positive control (cells cultivated in HS without drugs).

CD1a (OKT-6)10048582910950
CD1b (WM-25)1009867719793
MHC class I (W6/32)1001037367162119

Surface CD1a molecules unfold during serum deprivation and become immunologically inactive

To examine whether unfolded CD1a molecules remain on the cell surface in LS, we generated transfectants expressing CD1a fused with GFP at the cytoplasmic tail. The addition of GFP tail did not modify CD1a intracellular trafficking as detected by confocal analysis (supplementary online information). These hybrid molecules were studied with specific mAb to follow the extracellular CD1a domains and with GFP to localize the intracellular tail. A disappearance of surface staining with anti-CD1a mAb and concomitant detection of membrane-associated GFP would provide evidence for presence of altered CD1a on the cell surface. CD1a-transfected cells were incubated 24 h under HS or LS conditions, and then capping of surface CD1a was induced using anti-CD1a mAb. Confocal microscopy of cells cultivated in HS showed capping of all surface CD1a and colocalization of the anti-CD1a mAb and GFP signals (Fig. 2). In contrast, cells cultured in LS showed diffused distribution of membrane-associated GFP staining, capping of the CD1a molecules still detectable with mAb, and only partial colocalization of the mAb and GFP signals (Fig. 2).

thumbnail image

Figure 2. Cell surface CD1a molecules are unfolded in LS (A–F). J558 transfected with CD1a-GFP were cultured for 24 h with HS (A–C) or LS (D–F), and then capping of surface CD1a molecules was induced by OKT-6. Confocal microscopy detected GFP as a green signal and CD1a in red. Scale bar: 5 µm. Restoration of CD1a expression by sulfatide (G–J). MOLT-4 were cultured in HS (G) or LS (H, I, and J) for 24 h. Then cells were further cultured in LS alone (H), or with inclusion of sulfatide (I), or sulfatide and BFA+CHD (J). CD1a was detected with OKT-6 and Cy3-conjugated anti-mouse IgG and analyzed by confocal microscopy. Scale bar: 5 µm.

Download figure to PowerPoint

These findings confirm that part of the CD1a molecules unfolds in the absence of serum and remains anchored to the cell surface. The fact that specific mAb do not detect the altered molecules, raises the question whether CD1a also becomes functionally inactive. Therefore, we studied whether antigen presentation to CD1a-restricted T cells is also affected. MOLT-4 cells expressing 50% of CD1a after 24 h cultivation in LS were incubated with sulfatide (sulfo-galactosylceramide) or GM1, and used to stimulate CD1a- or CD1b-restricted T cell clones, respectively. MOLT-4 cells cultivated in HS were used as control APC. The response of a CD1a-restricted T cell clone was significantly reduced after antigen presentation with MOLT-4 kept in LS (Fig. 3A), whereas antigen presentation to a CD1b-restricted T cell clone was not affected, thus excluding a generic reduced capacity of antigen presentation after LS culture (Fig. 3B). These findings show that the altered CD1a molecules that remain on the cell surface are functionally inactive, and that serum components are required to maintain their antigen-presenting function.

thumbnail image

Figure 3. Altered CD1a molecules are functionally inactive and HS or sulfatide restore their functional expression. (A, B) MOLT-4 were cultured for 24 h in LS (circles), or in HS (squares) and then used as APC to stimulate CD1a-restricted and sulfatide-specific (A) or CD1b-restricted and GM1-specific (B) T cell clones. IL-4 released in the supernatants was measured by ELISA and expressed in pg/mL ± SD. (C) MOLT-4 cells were cultured in LS for 24 h before inclusion of sulfatide and further cultured for 24 h. Surface expression of CD1a (OKT-6) and CD1b (WM-25) was measured by FCM. MFI of cells cultured throughout in HS (positive controls) was considered as 100%. Each line represents a separate experiment. (D) MOLT-4 cultured for 24 h in LS and then further 24 h in HS (LS/HS) were used to present sulfatide (10 µg/mL) to the CD1a-restricted sulfatide-specific T cell clone. MOLT-4 always kept in HS or in LS served as controls. (E) Sulfatide efficiently stabilizes CD1a molecules present on the cell surface. MOLT-4 were kept in LS for 24 h, and then additional 24 h in the presence of BFA or CHD alone, or both together and without or with inclusion of sulfatide as indicated. Cells were stained with OKT-6 and analyzed by FCM. Results are expressed as percentage of MFI of the positive control (untreated cells in HS). MFI values are indicated next to the respective bar. The results represent one of three independent experiments.

Download figure to PowerPoint

The extracellular CD1a domain and not CD1a recycling pathway confers susceptibility to serum deprivation

CD1a is internalized in early endosomes, whereas the other CD1 molecules reach mature late endosomes 1417. This difference could expose CD1a to a set of ligands distinct from those for the other CD1 molecules, thus providing an explanation for the dependence of CD1a on exogenous ligands.

We studied transfectants expressing either hybrid CD1a molecules fused with the cytoplasmic tail of CD1b (CD1ab) that recycle in late endosomes, or hybrid CD1b molecules expressing the cytoplasmic tail of CD1a (CD1ba) and recycling to early endosomes (manuscript in preparation). When transfectants were cultured in LS, hybrid molecules with extracellular CD1a domain continued to be altered, similarly to control transfectants expressing wild type CD1a (Table 2). In contrast, hybrid molecules with extracellular CD1b domain were not altered.

Table 2. The extracellular CD1a domain is responsible for dependence on exogenous lipidsa)
mAb anti-
  1. a) Numbers show MFI of staining of T2 cells expressing CD1a or CD1b or CD1 hybrid molecules. The percentage of CD1a alteration detected by staining with OKT-6 is indicated in parentheses as negative value.

CD1aCD1bMHC class I
CD1a729272 (-63)911291266
CD1ab232119 (-49)1115307296

These findings show that the extracellular domain of CD1a and not recycling to early endosomes confers susceptibility to deprivation of exogenous ligands.

Sulfatide stabilizes CD1a on the cell surface

With respect to the nature of active serum components, we speculated that ligands binding to CD1a could be important for stabilization. Initially we studied sulfatide, which is very abundant in serum. MOLT-4 cells were cultivated first for 24 h in LS and then for a further 24 h in LS supplemented with sulfatide. Sulfatide as well as HS restored up to 90% of CD1a, whereas CD1b (Fig. 3C) and MHC class I (data not shown) molecules were not affected. Moreover, restored CD1a molecules also re-acquired their immunological activity. MOLT-4 cells incubated in LS and then in HS for additional 24 h presented sulfatide to CD1a-restricted T cells very efficiently (Fig. 3D), showing the functional flexibility of surface CD1a molecules.

Further experiments were performed to investigate whether sulfatide stabilizes newly synthesized CD1a molecules when they reach the cell surface and/or restores the conformation detectable with mAb of the molecules already present on the cell surface. MOLT-4 cells maintained for 24 h in LS were cultured for a further 24 h in the presence of BFA, CHD or both together, and with or without inclusion of sulfatide (Fig. 3E). Cells maintained first in LS but then further cultured in HS served as controls. In the presence of BFA, sulfatide restored CD1a to 80% of that on control cells (MFI: 307 vs. 388), indicating that cell surface and recycling CD1a molecules, not blocked by BFA 18, are stabilized. When CHD was used, sulfatide restored CD1a expression to level higher than on control cells (MFI: 339 vs. 291), indicating that the recycling and the de novo expressed CD1a molecules, blocked at the cell surface by CHD, are stabilized very efficiently. In the combined presence of BFA and CHD there was a more profound reduction of CD1a, which, however, was efficiently restored by sulfatide (91%, MFI: 264 vs. 289). This latter result suggests that surface CD1a molecules are rescued without a requirement for internalization, even when already altered.

Restoration of CD1a expression was also investigated by confocal microscopy (Fig. 2, G–J). MOLT-4 cells cultured for 24 h in LS conditions exhibited a marked reduction of both surface and intracellular CD1a (Fig. 2G and H). Inclusion of sulfatide during the additional 24 h period of culture in LS resulted in a restoration of CD1a to levels almost comparable to those of cells cultured in HS (Fig. 2G and I). Both surface and intracellular molecules were restored by culture in the presence of sulfatide. Furthermore, when sulfatide was added simultaneously with CHD, a strong CD1a staining was detected on the cell surface but not within intracellular compartments (Fig. 2J). Since CHD has a variety of effects including blocking of internalization of surface proteins, this finding indicates that CD1a internalization and recycling are not required for its stabilization.

Glycosphingolipids and phospholipids restore surface CD1a

We tested the ability of several lipids to restore CD1a conformation under LS conditions (Table 3). Sulfatide, GM1, sphingomyelin, phosphatidylserine (PS), phosphatidylcholine (PC) and phosphatidylinositol (PI) restored CD1a expression albeit with different efficiencies. In contrast, cholesterol, distearin, phosphatidylethanolamine (PEA) and phosphatidic acid (PA) were not able to restore surface CD1a expression, thus ruling out that the alterations observed in LS can be rescued by a generalized increase in cellular lipid content. These findings support the hypothesis that different ligands stabilize CD1a.

Table 3. Restoration capacity of different types of lipidsa)
Exogenous lipidCD1a (MFI)% Restoration
  1. a) MOLT-4 were first cultured in LS for 24 h and then for additional 24 h in the presence of 10 µg/mL of indicated lipid. CD1a expression is reported as MFI of OKT-6 staining and % restoration calculated according to the Materials and methods section.

HS 749100

Stabilization of CD1a on the cell surface by sulfatide and β2-microglobulin

In addition to antigen binding to CD1a, we tested whether exogenous β2-microglobulin (β2m) stabilizes CD1a in a similar manner. Purified human β2m at a concentration of 0.2 or 1 μg/mL, equivalent to the concentration range present in HS medium, restored up to 80% of CD1a expression in LS-pretreated MOLT-4 cells, as detected by surface staining with OKT-6 mAb, similar to what was observed with sulfatide (Fig. 4A). The addition of both sulfatide and β2m restored CD1a expression to an even higher extent (96% of control), suggesting that the two molecules act independently.

thumbnail image

Figure 4. Distinct role of sulfatide and β2m in CD1a cell surface stabilization. (A) Restoration of CD1a expression after 24 h cultivation of MOLT-4 in LS was tested by adding sulfatide (10 µg/mL) or β2m (1 µg/mL) or both or HS as control and culturing cells for additional 24 h. (B) Prevention of CD1a alteration in MOLT-4 cultured 24 h in LS supplemented with sulfatide (10 µg/mL) or β2m (1 µg/mL) or both or HS as control. Results in (A) and (B) are expressed as percentage of control MFI of staining for the indicated molecules.

Download figure to PowerPoint

Stabilization of the surface CD1a in its immunologically active conformation was also observed in experiments aimed at preventing the alteration observed in LS. In fact, by culturing cells with LS in the presence of sulfatide alone or sulfatide and β2m, CD1a was not altered (Fig. 4B). On the contrary, addition of β2m alone did not completely prevent alteration. These experiments further confirm that exogenous GSL stabilize CD1a and that they exert this effect with a mechanism different from that of β2m.

CD1a expression in human epidermal Langerhans cells

We next quantified the levels of CD1a expression on the surface of freshly isolated Langerhans cells (LC) as short as 1 h after incubation with FCS or sulfatide (Table 4). LC incubated in LS supplemented with sulfatide up-regulated CD1a expression (MFI average 37%) similarly to what observed with HS (MFI average 43%). The levels of MHC class I surface expression did not significantly change in any of these conditions. These findings show that also with freshly isolated LC, surface CD1a can be up-regulated in the presence of serum or of an appropriate lipid, thus further supporting the conclusion that CD1a is stabilized by exogenous ligands.

Table 4. Basal levels of surface CD1a in freshly isolated LC are up-regulated by serum and by sulfatidea)
CD1aMHC class I
  1. a) LC were incubated 1 h at 37°C with LS, HS, or LS supplemented with 10 µg/mL of sulfatide. Numbers show MFI. The percentage of modulation with respect to LS is indicated in parentheses as positive or negative values. ND: not done.

Exp. 19781777 (+82)1130 (+56)NDNDND
Exp. 29541275 (+34)1369 (+44)15671484 (-5)1152 (-26)
Exp. 322182515 (+13)2455 (+11)1244876 (-30)1376 (+11)


  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. Supporting Information

In this study we show that plasma membrane CD1a, but not CD1b, is kept in a functional conformation by exogenous ligands. CD1a efficiently presents lipid antigens to T cells only when present in this conformation. Surface levels of CD1b, MHC class I, MHC class II and CD71 are not affected by cell culture in LS, indicating that this phenomenon is specific to CD1a and not associated with a general reduced expression of surface molecules. CD1a functional alteration occurs independently of cellular internalization and mostly affects molecules residing on the cell surface. This is shown by differential staining with different CD1a-specific mAb using cytofluorimetry and confocal evidence of sustained surface GFP-tagged CD1a expression after LS-induced modulation.

CD1a functional alteration also occurs when freshly isolated thymocytes are incubated in LS, also showing that CD1a molecules that are likely to be associated with in vivo-bound ligands are stabilized by exogenous lipids. These findings demonstrate that surface CD1a becomes unstable and unfolds in the absence of exogenous ligands.

Previous studies have shown the unique capacity of CD1a to exchange associated β2m with that present in the serum 19, 20. Our results confirm that β2m participates in CD1a stability, however, they also add the novel finding that lipids are necessary for CD1a stabilization. The observation that sulfatide, but not β2m, efficiently prevents CD1a alteration occurring in LS further outlines the importance of lipids in stabilizing CD1a.

Glycosphingolipids such as sulfatide, ganglioside GM1 and sphingomyelin as well as phospholipids such as PS, PC and PI restore the expression of previously altered CD1a molecules with different efficiencies. The three GSL ligands share a ceramide tail, which anchors them to CD1a. The crystal analysis of the CD1a-sulfatide complex has outlined the importance of sphingosine which is inserted into the A1′ pocket and is likely of major importance for binding and stabilizing the entire complex 21. In contrast, PA, PEA, distearin and cholesterol do not restore the CD1a conformation. This may reflect a low binding to CD1a, although we cannot exclude other factors such as low availability of free molecules in aqueous media that could also be relevant.

CD1a functional alteration shows important and unique features. Altered CD1a molecules become immunologically incompetent as they lose their antigen presentation capacity. Thus, there is a direct relationship between surface CD1a levels and exogenous ligands availability, a relationship not observed for other antigen-presenting molecules.

Surface CD1a molecules are restored on the cell surface without absolute requirement for internalization, or de novo synthesis, as demonstrated by flow cytometry and confocal microscopy in the presence of CHD and BFA.

Stabilization by exogenous ligands is a feature of CD1a and does not apply to CD1b. This difference is not imposed by the intracellular distribution and traffic capacities, which differ between the two molecules, as shown by the behavior of CD1ab and CD1ba hybrid molecules. Therefore, it is likely that the distinct manner of lipid binding accounts for the observed different stability requirements. CD1a is characterized by a F’ pocket, which is relatively large, is exposed to the surface 21 and allows binding of exogenous lipids in the absence of acidification 10. The structure of this pocket might facilitate ready exchange of bound lipids, and when empty might induce collapse of the molecule. CD1b is instead characterized by a more intricate net of pockets 22, which allow binding of glycolipids with long acyl tails and may be stabilized by a variety of self ligands 10, 23. CD1a structural characteristics render therefore this molecule unique for its property of becoming more available for antigen-presentation when increased ligand concentration is available.

Surface binding and stabilization by exogenous ligands also differentiates CD1a from MHC molecules. Important features of MHC class I molecules are that they do not readily exchange peptide ligands on the plasma membrane 24, in the absence of peptides they become unstable, dissociate from β2m 25 and after this denaturation step they lose the capacity to bind exogenous peptides 26. This is a key mechanism to retain immunological identity and prevent killing of cells inappropriately presenting exogenous peptides. In the case of CD1a, bound ligands may be readily exchanged on the cell surface and perhaps the role of CD1a molecules is not to confer immunological identity in surveillance for virus-infected cells.

An important difference is also evident with respect to MHC class II molecules. Physiological antigen presentation by MHC class II molecules occurs after antigen internalization, processing and loading in specialized endosomal compartments in which the class II-associated invariant chain is exchanged with antigenic peptides. On the contrary, most immunogenic GSL may bind to CD1a on the cell surface and are immunogenic in the absence of processing.

Our results also show that CD1a expression on the surface of LC freshly isolated from the epidermis is up-regulated by serum or sulfatide. This up-regulation is very efficient and fast, as up to 43% of CD1a increase is observed already after 1 h of incubation. Thus, normal LC in the skin present a reservoir of CD1a molecules that upon ligand binding are recognized by specific mAb and likely become immunologically competent.

According to these results, we propose the following physiological role for CD1a. Newly synthesized CD1a molecules assemble in the ER and reach the cell surface in association with lipid ligands which are not GSL 5, with unknown chaperones or, alternatively, with an empty antigen-binding groove. CD1a molecules reaching the plasma membrane become available for lipid exchange or, if empty, are susceptible to unfolding and after cellular internalization undergo subsequent degradation. Loading of exogenous lipids stabilizes CD1a and might prolong its surface permanence in an immunologically competent form. According to this model, the repertoire of CD1a-bound antigens is represented by exogenous ligands, and thus CD1a may continuously sample the extracellular space for the presence of abundant lipid antigens. These ligands may be self-lipids, which are integral components of lipoproteins present in serum 27 and may be exchanged with other ligands bound to CD1a 10, according to their affinity and concentration. A second source of CD1a ligands is represented by foreign lipids of microbial origin. One microbial antigen that binds to CD1a is the lipopeptide didehydromycobactin, which is secreted by Mycobacterium tuberculosis28. This ligand, as well as other still unknown microbial ones 29 released by intracellular pathogens might traffic from phagosomes to early endosomes 30, or be released in the serum. If CD1a loading of soluble bacterial antigens is as efficient as that of self-GSL, it might also ensure immediate and adequate presentation when microbial antigens are generated in CD1a-negative cells. Thus, CD1a may sense lipid variations occurring within the extracellular milieu and, accordingly, readily present the repertoire of available antigens to T cells.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. Supporting Information


MOLT-4 and T2 cells are from American Type Culture Collection (ATCC, Manassas, VA). C1R-CD1a and C1R-CD1b were obtained from S. Porcelli, J558 from K. Karjalainen. Cells were grown in RPMI 1640 supplemented with 10% FCS, 10 mM HEPES, 2 mM UltraGlutamine II, MEM nonessential amino acids, 1 mM Na-pyruvate (all from Cambrex, Brussels, Belgium) and 100 µg/mL kanamycin (Invitrogen, Basel, Switzerland). DC and T cell clones were generated as described 23. Fresh thymocytes, isolated from thymi of children subjected to heart surgery, were kindly provided by M. Heberer. Fresh LC were obtained from surgical specimens of human skin 31.

Cell culture in the presence of exogenous compounds and drugs

Cells were washed twice with PBS and 5 × 105 cells/well were plated in HS or LS. The optimal serum concentration (0.2%) used in LS has been found by titrating FCS from 0.05 to 0.5% and evaluating cell viability and change in surface levels of CD1a, b and MHC class I expression over a 48-h period by FCM. The capacity of different compounds to restore CD1a expression altered after 24 h in LS was tested by incubating cells further 24 h with 10 µg/mL of the following compounds: sulfatide, PEA, sphingomyelin (all from Fluka Chemie AG, Buchs, Switzerland or from L. Panza), GM1, PA (Matreya, Pleasant Gap, PA), PC, PS, PI (Avanti Polar Lipids, Alabaster, AL), cholesterol or distearin (Sigma). Purified human β2m (Sigma) was used at 0.2 or 1 µg/mL. The restoration of CD1a expression was calculated as follows: % Restoration = (MFIcompound – MFILS)/(MFIHS – MFILS) × 100, where MFI is the CD1a median fluorescence intensity in the presence of the compound, in LS, or in HS. This latter value was taken as control for 100% restoration. In some experiments, BFA (1 µg/mL, Sigma) and CHD (10 µg/mL, Sigma) were added alone or together on MOLT-4 for 24 h. Cell viability after these treatments was always above 98%. To control the activity of BFA, monocytes were treated with 1 µg/mL BFA for 24 h in the presence of GM-CSF and IL-4 in order to induce differentiation of DC and de novo expression of CD1 molecules. The absence of CD1a and CD1b on the surface of BFA-treated cells confirmed the activity of the drug (not shown). The control for CHD activity was done by confocal microscopy, demonstrating CD1a expression increased at cell surface and decreased intracellularly (not shown and Fig. 2J).

Antigen-presentation assay

MOLT-4 were washed twice with PBS, incubated in HS or LS, as described above and after 24 h, washed twice and incubated (5 × 104 cells/well) with HS or sulfatide (0.3–10 µg/mL) for 2 h before adding T cells (6 × 104/well) in medium containing 0.5% FCS. CD1a alteration after incubation in 0.5% FCS is the same as that observed in LS (data not shown). Supernatants were harvested after 24 h and cytokine measured by ELISA 23.

Antibodies, FCM and confocal imaging

Cells were stained with 10 µg/mL of the following CD1a-specific mAb: Vit6b, Vit6c, Vit6d, B17, 10D12.2, 10H3.9.3, NA1/34 (W. Knapp and A. Woolfson), OKT-6 (Instrumentation Laboratory, Schlieren, Switzerland) and AD16 (obtained in our laboratory). Evidence that OKT-6, AD16 and B17 recognize different epitopes was obtained by competition assays (data not shown). Other mAb used were WM-25 (anti-CD1b, Immunokontakt, Switzerland), BCD1b3.1 (anti-CD1b, S. Porcelli), W6/32 (anti-MHC class I, ATCC), L243 (anti-MHC class II, ATCC), anti-CD71 (anti-transferrin receptor, PharMingen, Basel, Switzerland), BBM.1 (anti-β2m, ATCC-HB28). The mAb were revealed with FITC-conjugated goat anti-mouse Ig (Southern Biotechnology Associates, Birmingham, AL) and cells analyzed on a FACScan flow cytometer (Becton Dickinson, Allschwil, Switzerland). Background fluorescence was evaluated using isotype-matched irrelevant mAb. For confocal microscopy, cells were fixed permeabilized and stained as previously described 5. For CD1a detection OKT-6 was used followed by Cy3-conjugated sheep anti-mouse IgG (Sigma). Capping of surface CD1a molecules was induced in J558 cells transfected with CD1a-GFP by incubating 30 min at 37°C with OKT-6, followed by cell fixation and staining with sheep anti-mouse IgG. Cells were examined by confocal laser scanning microscope (Zeiss, Jena, Germany). The 0.5-µm Z-sections were analyzed by LSM software (Zeiss).

Generation of stable CD1a- and CD1b-expressing cell lines

Cell lines expressing WT CD1A and CD1B cDNA were obtained as described 5. CD1A-GFP fusion construct was prepared in the pEGFP-N3 expression vector (Clontech, Basel, Switzerland) after amplification of pCD1A-Bluescript II with the following primers: 5′XhoCD1A 5′TTCTCGAGATGCTGTTTTTGCTA and 3′CD1ABam 5′GGATCCACAGAAACAGCGTTTCC, lacking the stop codon. Hybrid CD1A/B and CD1B/A cDNA constructs were obtained using as templates pCD1A- and pCD1B-Bluescript II and the Advantage-HF 2 PCR kit (Clontech). CD1A/B contains extracellular (XC) and transmembrane (TM) gene segments of CD1A and cytoplasmic (CY) CD1B gene segment. CD1B/A contains XC and TM of CD1B and CY of CD1A. The following primers were used: 5′XhoXC1A_for 5′ATGCCTCGAGATGCTGTTTTTGCTACTTC, 3′CY1BTM1A_rev 5′ACCGGCGCCTGAACCAAAGCGCAAGACC, 5′TM1ACY1B_for 5′GCGCTTTGGTTCAGGCGCCGGTCA-TATC, 3′HincCY1B_rev 5′ATCTGTCGACTCATGGGATATTCTGATATG, 5′SalXC1B_for 5′TATAGTCGACATGCTGCTGCTGCCATTTCAAC, 3′CY1ATM1B_rev 5′AGCGTTTCCTCATATACCATAATGC, 5′TM1BCY1A_for 5′TTATGGTATATGAGGAAACGCTGTTTC, 3′NotCY1A_rev 5′ATAAGAATGCGGCCGCTTAACAGAAACAGCGTTTCC.

The amplified CD1A/B and CD1B/A hybrid constructs were sequenced after cloning into pBluescript II KS+ (Stratagene, La Jolla, CA), subcloned into the vector BCMGSNeo and transfected by electroporation into T2 cells. Stable transfectants expressing CD1ab or CD1ba hybrid molecules were selected with G418 sulfate (0.8 mg/mL, Calbiochem, La Jolla, CA) and analyzed at FACS for surface expression of CD1a or CD1b or at confocal microscope for intracellular colocalization studies. For CD1a modulation studies cells were incubated with BFA (1 µg/mL) for 24 h before staining.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. Supporting Information

We acknowledge A. Donda and A. Shamshiev for helpful discussions, Q. Schefer for preliminary experiments, L. Angman for excellent technical assistance, M. Heberer, W. Knapp, M. Kronenberg, L. Panza, S. Porcelli, and A. Woolfson for providing us with precious reagents, and B. Erne for assistance in confocal imaging. We also thank P. Dellabona, T. Resink, H. de la Salle and C. Watts for reading the manuscript and G. van Meer for stimulating discussions. This project was supported by grants to GDL from Swiss National Found (3100–66769.01 and 3100A0–109918) and Human Frontier Science Program (RG0168/2000-M). The work on LC was funded by ARMESA and by the Scientific Council of EFS (CS/2002/018). VM received an EMBO fellowship.

  • 1


  • 2


  • 3


  • 4


  • 1
    Germain, R. N. and Margulies, D. H., The biochemistry and cell biology of antigen processing and presentation. Annu. Rev. Immunol. 1993. 11: 403450.
  • 2
    Porcelli, S. A., Morita, C. T. and Brenner, M. B., Cd1b restricts the response of human CD48 T lymphocytes to a microbial antigen. Nature 1992. 360: 593597.
  • 3
    De Libero, G. and Mori, L., Recognition of lipid antigens by T cells. Nat. Rev. Immunol. 2005. 5: 485496.
  • 4
    Moody, D. B., Zajonc, D. M. and Wilson, I. A., Anatomy of CD1-lipid antigen complexes. Nat. Rev. Immunol. 2005. 5: 387399.
  • 5
    Manolova, V., Hirabayashi, Y., Mori, L. and De Libero, G., CD1a and CD1b surface expression is independent from de novo synthesized glycosphingolipids. Eur. J. Immunol. 2003. 33: 2937.
  • 6
    Joyce, S., Woods, A. S., Yewdell, J. W., Bennink, J. R., De Silva, A. D., Boesteanu, A., Balk, S. P. et al., Natural ligand of mouse CD1d1: cellular glycosylphosphatidylinositol. Science 1998. 279: 15411544.
  • 7
    Jayawardena-Wolf, J., Benlagha, K., Chiu, Y. H., Mehr, R. and Bendelac, A., CD1d endosomal trafficking is independently regulated by an intrinsic CD1d-encoded tyrosine motif and by the invariant chain. Immunity 2001. 15: 897908.
  • 8
    Kang, S. J. and Cresswell, P., Regulation of intracellular trafficking of human CD1d by association with MHC class II molecules. EMBO J. 2002. 21: 16501660.
  • 9
    Altamirano, M. M., Woolfson, A., Donda, A., Shamshiev, A., Briseno-Roa, L., Foster, N. W., Veprintsev, D. B. et al., Ligand-independent assembly of recombinant human CD1 by using oxidative refolding chromatography. Proc. Natl. Acad. Sci. USA 2001. 98: 32883293.
  • 10
    Shamshiev, A., Gober, H. J., Donda, A., Mazorra, Z., Mori, L. and De Libero, G., Presentation of the same glycolipid by different CD1 molecules. J. Exp. Med. 2002. 195: 10131021.
  • 11
    Helms, J. B. and Rothman, J. E., Inhibition by brefeldin A of a Golgi membrane enzyme that catalyses exchange of guanine nucleotide bound to ARF. Nature 1992. 360: 352354.
  • 12
    Poussin, C., Foti, M., Carpentier, J. L. and Pugin, J., CD14-dependent endotoxin internalization via a macropinocytic pathway. J. Biol. Chem. 1998. 273: 2028520291.
  • 13
    Parton, R. G., Joggerst, B. and Simons, K., Regulated internalization of caveolae. J. Cell Biol. 1994. 127: 11991215.
  • 14
    Sugita, M., Grant, E. P., van Donselaar, E., Hsu, V. W., Rogers, R. A., Peters, P. J. and Brenner, M. B., Separate pathways for antigen presentation by CD1 molecules. Immunity 1999. 11: 743752.
  • 15
    Salamero, J., Bausinger, H., Mommaas, A. M., Lipsker, D., Proamer, F., Cazenave, J.-P., Goud, B. et al., CD1a molecules traffic through the early recycling endosomal pathway in human Langerhans cells. J. Invest. Dermatol. 2001. 116: 401408.
  • 16
    Moody, D. B. and Porcelli, S. A., Intracellular pathways of CD1 antigen presentation. Nat. Rev. Immunol. 2003. 3: 1122.
  • 17
    Jackman, R. M., Stenger, S., Lee, A., Moody, D. B., Rogers, R. A., Niazi, K. R., Sugita, M. et al., The tyrosine-containing cytoplasmic tail of CD1b is essential for its efficient presentation of bacterial lipid antigens. Immunity 1998. 8: 341351.
  • 18
    Mc Dermott, R., Ziylan, U., Spehner, D., Bausinger, H., Lipsker, D., Mommaas, M., Cazenave, J. P. et al., Birbeck granules are subdomains of endosomal recycling compartment in human epidermal Langerhans cells, which form where Langerin accumulates. Mol. Biol. Cell 2002. 13: 317335.
  • 19
    Kefford, R. F., Calabi, F., Fearnley, I. M., Burrone, O. R. and Milstein, C., Serum beta 2-microglobulin binds to a T-cell differentiation antigen and increases its expression. Nature 1984. 308: 641642.
  • 20
    Bernabeu, C., van de Rijn, M., Lerch, P. G. and Terhorst, C. P., Beta 2-microglobulin from serum associates with MHC class I antigens on the surface of cultured cells. Nature 1984. 308: 642645.
  • 21
    Zajonc, D. M., Elsliger, M. A., Teyton, L. and Wilson, I. A., Crystal structure of CD1a in complex with a sulfatide self antigen at a resolution of 2.15 A. Nat. Immunol. 2003. 29: 29.
  • 22
    Gadola, S. D., Zaccai, N. R., Harlos, K., Shepherd, D., Castro-Palomino, J. C., Ritter, G., Schmidt, R. R. et al., Structure of human CD1b with bound ligands at 2.3 A, a maze for alkyl chains. Nat. Immunol. 2002. 3: 721726.
  • 23
    Shamshiev, A., Donda, A., Carena, I., Mori, L., Kappos, L. and De Libero, G., Self glycolipids as T-cell autoantigens. Eur. J. Immunol. 1999. 29: 16671675.
  • 24
    Schumacher, T. N., Heemels, M. T., Neefjes, J. J., Kast, W. M., Melief, C. J. and Ploegh, H. L., Direct binding of peptide to empty MHC class I molecules on intact cells and invitro. Cell 1990. 62: 563567.
  • 25
    York, I. A. and Rock, K. L., Antigen processing and presentation by the class I major histocompatibility complex. Annu. Rev. Immunol. 1996. 14: 369396.
  • 26
    Rock, K. L., Gamble, S., Rothstein, L., Gramm, C. and Benacerraf, B., Dissociation of beta 2-microglobulin leads to the accumulation of a substantial pool of inactive class I MHC heavy chains on the cell surface. Cell 1991. 65: 611620.
  • 27
    Merrill, A. H., Jr., Lingrell, S., Wang, E., Nikolova-Karakashian, M., Vales, T. R. and Vance, D. E., Sphingolipid biosynthesis de novo by rat hepatocytes in culture. Ceramide and sphingomyelin are associated with, but not required for, very low density lipoprotein secretion. J. Biol. Chem. 1995. 270: 1383413841.
  • 28
    Moody, D. B., Young, D. C., Cheng, T.-Y., Rosat, J.-P., Roura-mir, C., O'Connor, P. B., Zajonc, D. M. et al., T cell activation by lipopeptide antigens. Science 2004. 303: 527531.
  • 29
    Rosat, J. P., Grant, E. P., Beckman, E. M., Dascher, C. C., Sieling, P. A., Frederique, D., Modlin, R. L. et al., CD1-restricted microbial lipid antigen-specific recognition found in the CD8+ alpha beta T cell pool. J. Immunol. 1999. 162: 366371.
  • 30
    Schaible, U. E., Hagens, K., Fischer, K., Collins, H. L. and Kaufmann, S. H., Intersection of group I CD1 molecules and mycobacteria in different intracellular compartments of dendritic cells. J. Immunol. 2000. 164: 48434852.
  • 31
    Lipsker, D., Ziylan, U., Spehner, D., Proamer, F., Bausinger, H., Jeannin, P., Bohbot, A. et al., Heat shock proteins 70 and 60 share common receptors which are expressed on human monocyte-derived but not epidermal dendritic cells. Eur. J. Immunol. 2002. 32: 322332.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. Supporting Information

Supporting information for this article is available on the WWW under or from the author.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.