The Tyrosine-Containing Cytoplasmic Tail of CD1b Is Essential for Its Efficient Presentation of Bacterial Lipid Antigens

Immunity, Vol. 8, 341–351, March, 1998, Copyright 1998 by Cell Press

The Tyrosine-Containing Cytoplasmic Tail of CD1b Is Essential for Its Efficient Presentation of Bacterial Lipid Antigens Robin M. Jackman,* Steffen Stenger,† Agnes Lee,* D. Branch Moody,* Rick A. Rogers,§ Kayvan R. Niazi,† Masahiko Sugita,* Robert L. Modlin, †‡ Peter J. Peters,k and Steven A. Porcelli*# * Lymphocyte Biology Section Division of Rheumatology, Immunology, and Allergy Brigham and Women’s Hospital Harvard Medical School Boston, Massachusetts 02115 † Department of Microbiology and Immunology ‡ Division of Dermatology University of California at Los Angeles School of Medicine Los Angeles, California 90095 § BioMedical Imaging Institute Harvard School of Public Health Boston, Massachusetts 02115 k Department of Cell Biology Utrecht University School of Medicine Heidelberglaan 100 3584 CX Utrecht The Netherlands

Summary CD1b is an antigen-presenting molecule that mediates recognition of bacterial lipid and glycolipid antigens by specific T cells. We demonstrate that the nine– amino acid cytoplasmic tail of CD1b contains all of the signals required for its normal endosomal targeting, and that the single cytoplasmic tyrosine is a critical component of the targeting motif. Mutant forms of CD1b lacking the endosomal targeting motif are expressed at high levels on the cell surface but are unable to efficiently present lipid antigens acquired either exogenously or from live intracellular organisms. These results define the functional role of the CD1b targeting motif in a physiologic setting and demonstrate its importance in delivery of this antigenpresenting molecule to appropriate intracellular compartments. Introduction CD1 molecules are b2-microglobulin–associated transmembrane glycoproteins that are related to major histocompatibility complex (MHC)-encoded antigen-presenting molecules in both structure and evolution (Porcelli, 1995). The human CD1 locus on chromosome 1 contains a family of five nonpolymorphic CD1 genes, four of which are known to encode expressed proteins. Three closely related human CD1 proteins, designated CD1a, CD1b, and CD1c, are prominently expressed on specialized antigen- presenting cells in a wide variety of lymphoid and nonlymphoid tissues. All three of these CD1 proteins # To whom correspondence should be addressed (e-mail: sporcelli@ rics.bwh.harvard.edu).

have been shown to be required for the specific responses of subpopulations of TCR (T cell receptor) ab1 CD42CD82 T cells or CD42CD81 T cells that recognize pathogenic mycobacteria, including Mycobacterium tuberculosis and Mycobacterium leprae (Beckman et al., 1994, 1996; Rosat and Brenner, submitted). A fourth human CD1 protein, CD1d, has also recently been shown to be involved in activation of a specific subset of T cells (Exley et al., 1997). Remarkably, the antigens recognized by human CD1b- or CD1c-restricted T cells are mycobacterial lipids and glycolipids (Beckman et al., 1994; Sieling et al., 1995; Moody et al., 1997). In mice, CD1d can also apparently present glycolipid antigens (Kawano et al., 1997). These findings, together with recent x-ray crystallographic data showing a deep hydrophobic ligand binding groove in CD1 (Zeng et al., 1997), have led to the hypothesis that CD1 proteins bind lipid antigens and display them in a manner that allows specific recognition by T cells (Moody et al., 1997). Recent studies on the intracellular localization of CD1b show that this protein is prominently found in the endocytic system of monocytes treated with cytokines to induce a dendritic cell phenotype (Sugita et al., 1996; Prigozy et al., 1997). CD1b is broadly distributed in endosomal and lysosomal compartments, including those in which antigen loading for MHC class II molecules (MHC class II compartments [MIICs]) is believed to occur. Studies of HeLa cells transfected with a truncated form of CD1b indicate that the endosomal localization of CD1b is dependent on signals contained in its cytoplasmic tail (Sugita et al., 1996). This nine–amino acid domain contains a tetrapeptide that conforms to a tyrosine-based signal denoted as YXXØ (where Y is tyrosine, X represents any amino acid, and Ø is a bulky hydrophobic amino acid; (Sandoval and Bakke, 1994; Boll et al., 1996; Marks et al., 1997). In a variety of other proteins, tyrosine-based signals of this general type have been shown to direct intracellular targeting by binding to adapter protein complexes that are believed to regulate the sorting of integral membrane proteins into transport vesicles (Ohno et al., 1995). These results suggest that targeting of CD1b to MIICs and other endosomal locations could be partially or completely due to the effect of this apparent tyrosine-based targeting motif. In the present study, mutagenesis and chimeric constructs have been employed to show that the tyrosinebased motif is in fact both necessary and sufficient to direct the normal trafficking of CD1b. Furthermore, we show that the tyrosine-based motif is critical for efficient presentation of M. tuberculosis antigens by either exogenous antigen-pulsed or M. tuberculosis– infected monocytes. These resultsdemonstrate the critical role of the tyrosine-based motif in the endosomal targeting of CD1b and link the intracellular trafficking properties of this protein with its function in the presentation of lipid and glycolipid antigens. Our findings also establish that an antigen-presenting molecule can be regulated in its delivery to various compartments within the cell by the system of adapter–protein complexes.

Immunity 342

Figure 1. Localization of CD1b.WT and CD1b. TD by Immuno-Electron Microscopy in Transfected THP-1 Monocytoid Cells (A) Labeling for CD1b.WT (10 nm gold particles) was seen in MIIC (*) and on the plasma membrane (p). Mitochondria (m) are not labeled with gold. THP-1 cells have large numbers of microvilli resulting in the finger-like projections and invaginations of the plasma membrane. Scale bars, 200 nm. (B and C) Colocalization of CD1b.WT (10 nm gold particles and arrows) with Lamp-1 (15 nm gold particles and arrowheads) is shown in MIIC (*). Strong labeling of CD1b.WT is seen in early endosomes (e) without Lamp-1 labeling. No labeling is seen on the endoplasmic reticulum (er) or on mitochondria (m). (D) CD1b.TD labeling (10 nm gold particles) is much stronger at the plasma membrane (p) than in MIIC compartments (*). (E) Quantitative EM was used to compare endosomal density (en) of CD1b.WT to CD1b.TD in these cells (gold particles/area, arbitrary units 6 1 standard deviation). Plasma membrane density (p) was also compared (gold particles/length, arbitrary units). The ratios of endosomal labeling to plasma membrane labeling (en/p) were then compared, showing a 15-fold decrease for CD1b.TD as compared to CD1b.WT.

Results Cytoplasmic Tail–Mediated Endosomal Localization of CD1b in Monocytic Cells Previous studies show that the cytoplasmic tail of CD1b is required for the prominent localization of this protein

within the endocytic system of transfected HeLa cells (Sugita et al., 1996). CD1b is expressed in extrathymic tissues mainly by dendritic cell populations and cytokine-differentiated monocytes. Because HeLa cells are representative of a cell lineage that does not normally express CD1b (i.e., epithelial origin), we extended the

Functional Effects of CD1b Targeting 343

Table 1. Cytoplasmic Domains of Cd1b Mutants, Endosomal Markers, and Other CD1 Family Members Cytoplasmic Domain Protein

307 311 315

CD1b.WT CD1b.TD CD1b.Y311A Lamp-1 HLA-DMb CD1a CD1c CD1d

RRRSYQNIP* RR* RRRSA AQNIP* RRKKSHAGYQTI* SWRRAGHSSYTPLPGSNYSEGWHIS* RKRCFC* KKHCSYQDIL* KRQTSYQGVL*

Amino acid sequences are indicated by standard single letter code, and the numbering above CD1b.WT corresponds to the residue number of the wild-type protein. The YXXØ targeting motif is underlined in each protein in which it is present, and the carboxy terminus is designated with an asterisk in each case. Alanine substitution for a wild-type residue is shown in bold.

analysis of the targeting effect of the cytoplasmic tail sequence to the non–CD1-expressing monocytic cell line THP-1 (Tsuchiya et al., 1980), which more closely approximates cell types that naturally express CD1b. Analysis of stably transfected THP-1 lines expressing the wild-type CD1b protein (CD1b.WT) or a tail-deleted form of this protein (CD1b.TD) was initially carried out by immunofluorescence microscopy. This showed punctate intracellular staining for CD1b.WT and strong surface localization for CD1b.TD (data not shown), indicating that the full-length form was required for endosomal localization in the monocytic cell line. This effect of the CD1b tail deletion on localization in THP-1 cells was confirmed and assessed at a greater level of resolution using immunogold labeling and electron microscopy (EM) of ultrathin cryosections. THP-1 cells that expressed CD1b.WT had the most prominent labeling in MIICs (which represent conventional late endocytic compartments including lysosomes [Kleijmeer et al., 1997]), with detectable but low surface staining and occasional labeling of electron lucent early endosomes (Figure 1A). Double-labeling with antibody to CD1b and lysosomeassociated membrane protein-1 (Lamp-1), a marker of late endosomes and lysosomes, showed extensive colocalization of these two proteins in electron-dense multivesicular or multilamellar endosomes (Figures 1B and 1C). As measured by quantitative EM analysis (Figure 1E), endosomal labeling was significantly higher than plasma membrane labeling. Thus the distribution of CD1b.WT in transfected THP-1 cells was similar to the distribution in monocytes that had been induced by cytokine treatment to express CD1b (Sugita et al., 1996; Prigozy et al., 1997). In contrast to the pattern observed in THP-1 cells expressing CD1b.WT, immunogold labeling of CD1b in THP-1 cells expressing CD1b.TD showed prominent expression on the plasma membrane with very low labeling in either early or late endosomes (Figure 1D). In marked contrast to the result with CD1b.WT, quantitation of the immunogold labeling of THP-1 cells transfected with CD1b.TD showed that plasma membrane CD1b-specific labeling was significantly higher than endosomal labeling (Figure 1E). Thus, the overall ratio of endosomal to

Figure 2. Immunofluorescence Microscopy of CD1b Mutants HeLa cells stably expressing the wild-type (A) or the TD (B) or Y311A (C) mutant forms of CD1b were grown on cover slips, fixed with paraformaldehyde, and permeabilized with digitonin. Double labeling for CD1b and Lamp-1 was carried out using 4A7.6 mouse antiCD1b monoclonal antibody (detected with CY3-conjugated goat anti-mouse Ig [left]) and 931B rabbit anti-human Lamp-1 antiserum (detected with FITC-conjugated goat anti-rabbit IgG [right]). Scale bar, 10 mm (all panels).

plasma membrane CD1b labeling for CD1b.WT was 15fold greater than for the cytoplasmic tail deletion mutant. These results confirmed the importance of the cytoplasmic tail sequence for efficient endosomal targeting of CD1b in a cell type belonging to a lineage that normally expresses this protein in vivo. Critical Role of the Cytoplasmic Tyrosine Residue for Endosomal Delivery of CD1b Because the cytoplasmic tail of CD1b contains a YXXØ motif and only five additional amino acids, it appeared likely that the lack of this motif was responsible for the redistribution of CD1b.TD to the plasma membrane. We further analyzed the molecular signal that directs CD1b localization by comparing the localization of CD1b.WT and CD1b.TD with that of a mutated form of CD1b in which the tyrosine residue of the YXXØ motif at position 311 (Y311) was replaced by an alanine (CD1b.Y311A) (Table 1). In both monocytoid cells and transfected HeLa cells, CD1b normally colocalizes with the late endosomal marker Lamp-1 (Figures 1B and 1C) (Sugita et al., 1996), which also depends upon a YXXØ motif for its correct intracellular localization (Table 1). This colocalization was confirmed for CD1b.WT by immunofluorescence microscopy of transfected HeLa cells (Figure 2A). In contrast, colocalization with Lamp-1 was completely abolished either by deletion of the entire YXXØ motif in CD1b.TD (Figure 2B) or by modification of the only

Immunity 344

Figure 3. Confocal Microscopy Comparing CD1b.WT to Targeting Motif Mutants HeLa cells stably expressing the indicated forms of CD1b were grown on cover slips and processed for immunofluorescence microscopy after 48 hr in culture. Cells were double-labeled with 4A7.6 (CD1b, green vesicles) (detected with FITC-conjugated goat anti-mouse Ig) and with rabbit antiserum to human Lamp-1 (red vesicles) (detected with Texas Red-conjugated goat anti-rabbit IgG). Fluorescent confocal images were obtained for CD1b expression (green; A, D, and G) and Lamp-1 expression (red; B, E, and H). The two images were then superimposed (C, F, and I) to show vesicles expressing both CD1b and Lamp-1 (yellow) as well as vesicles expressing only CD1b (green) or Lamp-1 (red). Enlargements of the boxed areas in C, F, and I are shown in C9, F9, and I9. Scale bars, 10 mm and 5 mm in enlargements.

tyrosine residue within the motif in CD1b.Y311A (Figure 2C). This loss of endosomal localization was associated with the acquisition of strong CD1b-specific surface staining for both CD1b.TD and CD1b.Y311A, presumably as a result of the mis-sorting of CD1b from intracellular compartments to the plasma membrane. To better observe any residual endocytic staining obscured by the extremely bright surface fluorescence of the CD1b cytoplasmic tail mutants, fluorescent confocal microscopy was used to examine optical sections of individual cells. HeLa cells expressing either the WT, TD, or Y311A forms of CD1b were double-labeled with antibodies to CD1b (Figures 3A, 3D, and 3G; green) and Lamp-1 (Figures 3B, 3E, and 3H; red). CD1b.WT was found in vesicles distributed both centrally and peripherally throughout the cells, showing a nearly complete intracellular colocalization with Lamp-1 (Figures 3A–3C). In contrast, endosomal localization of CD1b.TD and CD1b.Y311A was markedly reduced and instead showed a marked increase in staining on the plasma membrane (Figures 3D and 3G). Colocalization of CD1b.TD and CD1b.Y311A with Lamp-1 was virtually abolished in the peripherally distributed vesicles as shown by the large

number of Lamp-1–positive vesicles without CD1b coexpression (Figures 3F and 3I; red vesicles). Coexpression of these forms of CD1b with Lamp-1 was limited mainly to a relatively few perinuclear vesicles that were most likely terminal lysosomes (Figures 3F and 3I; central yellow vesicles). Similar results were obtained by both immunofluorescence microscopy and confocal microscopy using cotransfected human leukocyte antigen (HLA)–DM instead of endogenous Lamp-1 as a marker for late endosomes (data not shown). Thus, the point mutation of residue Y311 to alanine appeared to have the same effect on overall localization as deletion of the entire cytoplasmic domain. The Cytoplasmic Tail of CD1b Was Sufficient for Endosomal Delivery of a Chimeric Protein Although the effects of deletion or mutation of the CD1b cytoplasmic tail demonstrated its necessary role for endosomal localization, it remained possible that sequences within other portions of the protein also contributed to this localization. Thus, to show that the cytoplasmic domain of CD1b was sufficient in itself to control

Functional Effects of CD1b Targeting 345

Figure 4. Localization of a Chimeric Protein with a CD1b Cytoplasmic Tail HeLa cells stably expressing HLA-B27 (top) or B27/CYT.b (bottom) were transiently transfected with CD1b.WT and grown on cover slips for 48 hr, then fixed with paraformaldehyde and permeabilized with digitonin. Double-labeling was carried out with antiHLA–B27 mAb ME-1 (mouse IgG1, detected with tetramethylrhodamine isothiocyanateconjugated goat anti-mouse IgG1–specific antibody [left]) and anti-CD1b mAb 4A7.6 (mouse IgG2a, detected with FITC-conjugated goat F(ab9)2 anti-mouse IgG2a-specific antibody [right]). Both detecting antibodies had minimal cross reactivity, as indicated by control samples stained with ME-1 or with 4A7.6 alone (data not shown).

the observed CD1b steady-state localization, a chimeric protein (B27/CYT.b) composed of the extracellular and transmembrane domains of the MHC class I molecule HLA-B27 fused to the nine amino acids of the cytoplasmic domain of CD1b was analyzed. As for all MHC class I molecules, HLA-B27 that has passed through the biosynthetic pathway localizes predominantly at the cell surface and is not normally found in endosomes to any significant degree (Peters et al., 1991b). Effects of the CD1b cytoplasmic tail could therefore be monitored independently of the CD1b transmembrane and extracellular domains by examining the localization of the HLAB27 extracellular domain in cells expressing the B27/ CYT.b chimera. HeLa cells, which express several class I proteins normally but do not posses the HLA-B27 allele, were stably transfected with either the wild-type HLA-B27 construct or the chimeric B27/CYT.b construct. Subsequently, the CD1b.WT construct was transiently introduced into these cells to serve as a marker for wildtype CD1b localization. Both wild-type HLA-B27 and the B27/CYT.b chimera were detected by immunofluorescence with monoclonal antibody (mAb) ME-1, which specifically recognizes an epitope present on the extracellular domains of HLA-B27. ME-1 staining of the wildtype HLA-B27 transfectants was seen primarily on the cell surface, with little or no intracellular punctate staining (Figure 4, top left), and there was virtually no intracellular colocalization with the wild-type CD1b protein (Figure 4, top right). In contrast, the localization of the chimeric B27/CYT.b molecule (Figure 4, bottom left, again detected with ME-1) was indistinguishable from that of wild-type CD1b (Figure 4, bottom right). Addition of protease inhibitors such as leupeptin significantly increased intracellular and plasma membrane

staining of B27/CYT.b but did not affect levels of CD1b.WT (data not shown). Such differential protease susceptibility of these molecules is to be expected, since the luminal portion of HLA-B27 does not normally traffic to the late endosome and presumably has not evolved protease resistance. In contrast to the high surface expression of wild-type HLA-B27, the surface levels of B27/CYT.b were low when measured by flow cytometry despite the strong intracellular staining seen by immunofluorescence microscopy. Even in the presence of leupeptin, the level of ME-1 surface staining of B27/ CYT.b transfectants remained only about 15% of that of wild-type HLA-B27, consistent with the profound retention of the chimeric protein in intracellular compartments. Thus, the patterns of localization of wild-type HLA-B27 and the B27/CYT.b chimera could be directly correlated with the cytoplasmic tail sequences of the expressed proteins (i.e. the amino acid sequences below the images in Figure 4). These results prove that the short CD1b cytoplasmic tail was sufficient to redirect a normally nonendosomal protein to a pattern of localization indistinguishable from that of wild-type CD1b. Enhancement of Exogenous Antigen Presentation by Endosomal Targeting of CD1b B lymphoblastoid cell lines (BLCL) are known to be efficient at presenting exogenously acquired antigens via the MHC class II pathway, and previous studies have shown that CD1b-transfected BLCLs can present exogenous lipid antigens to CD1b-restricted T cells (Porcelli et al., 1992; Behar et al., 1995). Therefore, BLCLs transfected with wild-type or mutant forms of CD1b were used as a model system to assess the effects of cytoplasmic tail-mediated endosomal targeting on antigen presentation.

Immunity 346

Figure 5. Reduced Efficiency of Exogenous Antigen Presentation by CD1b Endosomal Targeting Mutants (A) LDN5 T cells specific for GMM were coincubated with C1R cells expressing CD1b wild-type, mutants, or mock-transfected controls at various antigen doses for 48 hr, at which time the supernatants were harvested and levels of IFNg and GM-CSF determined by ELISA. Background cytokine release in the absence of antigen was below detection limits. Experiments were done in triplicate with error bars showing 61 standard deviation. (B) Same as (A) except responses of mycolic acid specific DN1 T cells are shown. Because of low-level autoreactivity in DN1, cytokine secreted in the absence of antigen in response to each APC was subtracted from the values shown (less than 1.3 ng/ml for IFNg and 1.0 ng/ml for GM-CSF). (C) C1R clones used above were stained for flow cytometry with anti-CD1b mAb BCD1b3.1 followed by FITC-conjugated goat F(ab9)2 anti-mouse Ig and analyzed by flow cytometry. Mean fluorescence intensity (MFI) corresponding to surface expression, is shown. (D) The same C1R clones were analyzed for total immunoprecipitable CD1b by Western blotting. The cells were lysed and immunoprecipitations for CD1b were done using the 4A7.6 antibody. The protein was then deglycosylated and run on an SDS-PAGE gel, transferred to a polyvinylidene difluoride membrane, and immunoblotted with rabbit anti-human CD1b antiserum followed by horseradish peroxi-

Stable transfectant clones of the BLCL C1R expressing either CD1b.WT, CD1b.TD, or CD1b.Y311A were compared for their ability to present exogenous bacterial lipid and glycolipid antigens to the CD1b-restricted T cell lines DN1 and LDN5. This was assessed by stimulation of interferon g (IFNg) and granulocyte–macrophage colony-stimulating factor (GM-CSF) release. LDN5, which is specific for the mycobacterial glycolipid antigen glucose monomycolate (Moody et al., 1997), responded in the presence of C1R cells expressing CD1b.WT at an antigen concentration 100-fold lower than the amount required to stimulate an equivalent response using C1R cells expressing either CD1b.TD or CD1b.Y311A (Figure 5A). A similar result was obtained with T cell line DN1, specific for the mycobacterial lipid antigen mycolic acid (Porcelli et al., 1992; Beckman et al., 1994), for which the TD and Y311A forms of CD1b also had a significantly reduced ability to present antigen as compared to CD1b.WT (Figure 5B). The reduced ability of CD1b.TD and CD1b.Y311A to present antigen to LDN5 and DN1 was demonstrated consistently despite the significantly higher surface expression of both mutant forms of CD1b as compared with CD1b.WT in the C1R transfectants (Figure 5C). To demonstrate that the greater efficiency of CD1b.WT in exogenous antigen presentation was not due to a higher total cellular content of CD1b, quantitation was done by Western blot analysis of immunoprecipitable CD1b (Figure 5D). This revealed that the C1R CD1b.WT transfectant had up to 6-fold less total immunoprecipitable CD1b compared with CD1b.TD and CD1b. Y311A transfectants, despite presenting exogenous antigen up to 100-fold more efficiently. In addition, a panel of ten different conformation-sensitive anti-CD1b monoclonal antibodies showed identical patterns of reactivity with CD1b.WT and CD1b.TD transfectants, indicating that gross conformational differences were unlikely to be present (data not shown). This possibility was assessed more stringently by examining the responses of a CD1breactive, foreign antigen–independent T cell line. Previous studies have described such directly reactive or antigen-independent T cell lines that recognize target cells expressing CD1a, CD1c, or CD1d without a requirement for a cognate foreign antigen (Porcelli et al., 1992; Beckman et al., 1996; Exley et al., 1997). Recently, we isolated a CD81 TCRab1 T cell line designated CD8.3 that shows such antigen-independent direct reactivity to CD1b (D.B. Moody, R.M. Jackman and S.A. Porcelli, unpublished data), thus providing a useful tool with which to assess the presence of native CD1b protein on the surface of antigen-presenting cells (APCs) while also verifying the integrity of general APC functions.

dase–conjugated protein A and enhanced chemiluminescence for detection. Differences in band intensity are quantified below each lane on a relative scale by densitometry. As a control, class II a chain concentration was determined in parallel using L243 for immunoprecipitation and rabbit anti-human class II a antiserum for detection. (E) CD1b directly reactive T cell line CD8.3 was coincubated with an equal number of the same C1R clones expressing the CD1b forms shown. IFNg release was determined after 48 hr of culture.

Functional Effects of CD1b Targeting 347

Indeed, CD8.3 responded more strongly to C1R cells expressing each of the mutant forms of CD1b than to C1R cells expressing CD1b.WT (Figure 5E). Taken together, these results strongly support the hypothesis that the reduction in exogenous antigen presentation by APCs expressing the CD1b mutants was a direct consequence of the alteration in their intracellular trafficking and likely reflected reduced antigen loading of CD1b in intracellular compartments. Effects of Endosomal Targeting of CD1b on Presentation of Antigens from Live Intracellular Mycobacteria In infected tissues in vivo, mycobacterial pathogens are found predominantly intracellularly, in the vacuoles of monocyte-derived cells. We therefore sought to elucidate the importance of intracellular localization of CD1b on presentation of antigens derived from live bacteria growing within APCs. Four independent CD1b-restricted M. tuberculosis antigen-specific T cell lines were tested for recognition of THP-1 promonocytic cells expressing the wild-type or tail-deleted forms of CD1b and infected with live virulent M. tuberculosis (strain H37Rv). Recognition of the infected cells by each T cell line was measured by specific induction of cytokine secretion. Although each T cell line released different absolute levels of IFNg, in every case infected THP-1 cells expressing the CD1b.WT molecule presented endogenously derived mycobacterial antigen significantly better (Figure 6A) than those expressing CD1b.TD (Figure 6A). The presentation of antigen by CD1b.TD was reduced despite its surface expression at a level approximately five times higher than that of CD1b.WT. To show again that the reduced presentation by CD1b.TD was most likely due to defective antigen loading of CD1b, cytokine responses of the antigen-independent CD1breactive T cell line CD8.3 to uninfected THP-1 cells expressing CD1b.WT or CD1b.TD were measured. In contrast to the results obtained with the mycobacterial antigen-specific T cell lines, CD8.3 responded better to THP-1 cells expressing CD1b.TD (Figure 6B), consistent with its antigen-independent recognition. Thus, for the presentation of exogenous bacterial antigens, the efficient presentation by CD1b of antigens originating from live bacteria growing within an APC was also critically dependent on the correct intracellular trafficking of this protein. Discussion A variety of recent studies support the view that CD1 is a third distinct family of antigen-presenting molecules for specific T cell responses (reviewed by Porcelli, 1995; Melian et al., 1996; Sugita et al., 1998). CD1b has unique intracellular trafficking properties that are distinct from those of other known antigen-presenting molecules. Thus, while CD1b shows strong localization to a variety of endocytic compartments in a pattern that substantially overlaps that of MIICs, its intracellular trafficking and steady-state distribution is controlled by signals present in its own short cytoplasmic tail. This distinguishes it from MHC class I molecules that traffic directly

Figure 6. Reduced Efficiency of Endogenously Derived Antigen Presentation by Tail-Deleted CD1b (A) THP-1 cells transfected with CD1b.WT (crosshatched bars), CD1b.TD (filled bars), or untransfected THP-1 cells were infected with M. tuberculosis (strain H37Rv), washed, and coincubated with an equal number of the indicated T cell lines. The supernatant was harvested after 22–30 hr, followed by cytokine ELISA. Nonspecific IFNg secretion by T cells in response to uninfected APCs was subtracted from values shown. (B) CD8.3 cells were coincubated with uninfected THP-1 cells that were either untransfected, positive for CD1b.WT (crosshatched bar), or CD1b.TD (filled bar). Supernatants were harvested after 48 hr, and the IFNg response to the untransfected THP-1 (1.76 ng/ml) was subtracted as background.

to the cell surface through the default secretory pathway, and from MIICs that are targeted to the endocytic system primarily by the modified dileucine motif present in associated invariant chain molecules (Bakke and Dobberstein, 1990; Neefjes and Ploegh, 1992). In this study we demonstrate that despite the apparent structural similarity of CD1b to MHC class I molecules (which are loaded with peptide antigens predominantly in the endoplasmic reticulum), the ability of CD1b to present antigen efficiently is critically dependent on its endosomal class II–like distribution. In fact, CD1b largely lost its ability to present antigen when it assumed a more class I–like surface distribution as a result of deletion or mutation of its targeting signal. Furthermore, our results showed that efficient recognition of M. tuberculosis– infected monocytoid cells by CD1b-restricted T cells was dependent on correct targeting of CD1b to the late endosome/MIIC. All of these findings are likely to be relevant to the normal biology of this antigen-presenting pathway. Previous studies in mice have shown that both MHC class II–restricted T cells and b 2m–dependent T cells (i.e., presumably MHC class I-restricted) participate in

Immunity 348

host defense to M. tuberculosis (Kaufmann, 1993). However, after entering macrophages, mycobacteria can sequester themselves in phagolysosomes and alter the normal trafficking and maturation of these compartments (Clemens and Horwitz, 1995). Phagolysosomes containing pathogenic mycobacteria have been shown to become depleted of the vesicular proton-ATPase, resulting in a failure of these compartments to acidify normally (Xu et al., 1994). These effects of mycobacterial infection may have profound negative consequences on the presentation of mycobacterial protein antigens to MHC-restricted T cells. Indeed, monocyte-derived macrophages that are chronically infected with M. tuberculosis have been shown to become ineffective at stimulating CD41 MHC class II–restricted T cells in vitro (Pancholi et al., 1993). Presentation of lipid antigens by the CD1 system in infected cells may help to overcome this putative defect, since mycobacterial lipids such as LAM have been shown to traffic actively out of the M. tuberculosis phagosome and become widely dispersed throughout the endocytic system and on the plasma membrane of infected cells (Xu et al., 1994). Thus, CD1b could allow T cell recognition of infected macrophages in situations in which MHC class II function is severely impaired or effectively eliminated. A mechanism for the cotargeting of relevant glycolipid antigens into the MIIC was recently demonstrated (Prigozy et al., 1997). Previous studies have also shown that treatment of CD1b1 APCs with inhibitors of endosomal acidification or fixation of the APCs prior to antigen pulsing completely blocks presentation of lipid antigens to CD1b-specific T cells (Porcelli et al., 1992; Sieling et al., 1995). Together these findings suggest that lipid antigen loading of CD1b occurs in an acidic intracellular compartment. This is consistent with our findings that a major intracellular site of CD1b accumulation is the MIIC, and that failure to target to this internal low-pH compartment (due to lack of a proper targeting sequence) abolishes antigen-dependent recognition by T cells. Our EM findings showed that the cytoplasmic tail of CD1b was critical for its localization to the late endosomal/MIIC compartments in a physiologically relevant cell type. In the absence of the cytoplasmic tail, CD1b was concentrated on the monocyte cell surface with relatively little localized in late endosomal compartments. These findings were extended using both immunofluorescence and confocal microscopy to show that this targeting was dependent on the single tyrosine residue found within the YXXØ motif in the cytoplasmic tail of CD1b. Similarly, other studies have shown that mutation of the tyrosine in the YXXØ targeting signals in molecules such as Lamp-1 (Williams and Fukuda, 1990) can cause the redistribution of these molecules to the cell surface. Thus, the current studies directly confirm the role of residue Y311 and the YXXØ motif in controlling the steady-state localization of CD1b. Our results also show that the nine–amino acid cytoplasmic tail of CD1b is both necessary and sufficient to confer its wild-type localization on a nonendocytic reporter protein, such as an MHC class I heavy chain. It is remarkable that the motif, when placed on HLAB27, replicates CD1b localization even to the level of

substantially lower surface expression. Although the altered localization of this reporter molecule was apparent without protease inhibitors, addition of leupeptin significantly increased intracellular staining of B27/CYT.b while having little effect on wild-type CD1b. This suggests that CD1b has evolved protease resistance to allow its traffic through the endocytic and lysosomal vesicles, whereas the class I luminal domain does not normally traffic through these vesicles and has presumably no need for protease resistance. CD1b is the first antigen-presenting molecule that has been shown to be dependent on a YXXØ motif for its localization. HLA-DM, a molecule that facilitates peptide loading of MHC class II molecules, also uses such a motif for its localization (Table 1 and Marks et al., 1995). This is significant considering the demonstrated colocalization of MHC class II, HLA-DM, and CD1b in MIIC. Compared to other late endosomal/lysosomal targeted proteins that contain such a targeting motif and are found almost exclusively intracellularly, the endosomal targeting motif of CD1b appears to be relatively weak and thus allows a higher proportion of CD1b to remain on the cell surface. A YXXØ motif is predicted to induce a “tight turn” structure (Bansal and Gierasch, 1991; Eberle et al., 1991) that associates with adapter complexes and allows clathrin-mediated internalization. The identity of the Y11 and Y12 amino acids in a YXXØ motif can influence the strength of its interaction with the AP-2 adapter complex (which mediates internalization from the plasma membrane). This interaction has been reported to be strongest when both of these are polar residues (Boll et al., 1996). CD1b contains asparagine and isoleucine at these positions, predicting a relatively lower affinity for AP-2 and therefore a relatively slow internalization from the plasma membrane. In Lamp-1, the spacing of the YXXØ motif relative to the membrane has also been shown to be critical for effective endocytic targeting. Placing the motif a single amino acid closer to the membrane significantly reduced the rate of internalization from the plasma membrane, whereas extending the distance from the membrane increased internalization (Rohrer et al., 1996), presumably because of changes in the relative accessibility to the AP-1 and AP-2 adapter complexes. In CD1b the tyrosine-based motif is very close to the membrane, two amino acids closer than in Lamp-1. A CD1b mutant that increases the spacing of the motif relative to the membrane by five amino acids shows a marked increase in endocytic localization relative to surface localization (R. M. J., unpublished data). These data, together with the prominent visualization of CD1b in clathrin-coated vesicles (Sugita et al., 1996), suggest a recycling pathway whereby CD1b is serially internalized and delivered to endosomes. As is now believed to be the case for MHC class II (Pinet et al., 1995), recycling may provide an important alternative pathway for presentation of certain lipid antigens. However, the relative importance of CD1b targeting to relevant endosomal compartments via a recycling pathway versus direct targeting from the trans–Golgi network remains to be established. Although these studies focus exclusively on CD1b, it appears likely that endosomal targeting will also prove important for several other CD1 proteins. Both CD1c

Functional Effects of CD1b Targeting 349

and CD1d have cytoplasmic tail motifs similar to that of CD1b (Table 1), and preliminary studies show that both of these molecules also localize within the endocytic system (R. M. J. and S. A. P., unpublished data). However, the localization of CD1c and CD1d to endosomes in transfected HeLa cells appears to be less prominent than observed for CD1b, and a larger cohort of the former proteins seem to reside on the plasma membrane. The precise identity of the endocytic compartments to which CD1c and CD1d are transported and the functional significance of this localization remain to be established. Thus, while human CD1c and murine CD1d are known to present lipid antigens to T cells (Beckman et al., 1996; Kawano et al., 1997), the dependence of this presentation on endosomal targeting has not yet been shown. Interestingly, CD1a has no discernible targeting motif in its cytoplasmic tail (Table 1), and has been found to localize to the plasma membrane with little late endosomal localization in most cell types examined (M. S. et al., unpublished data). Nevertheless, this molecule has been found to present mycobacterial lipid antigens to T cells (Rosat and Brenner, submitted). Although further studies are required, these preliminary findings suggest the possibility that the individual CD1 family members may have evolved to efficiently survey different cellular compartments and to present distinct species of lipid antigens that associate with these compartments. In conclusion, we have shown that the cytoplasmic tail of CD1b is both necessary and sufficient for normal targeting of CD1b, and that the mutation of a single tyrosine within the tyrosine-based endocytic motif is sufficient to abrogate this targeting. Furthermore, this targeting was shown to be critical for presentation of both exogenous lipid and glycolipid antigens as well as antigens derived from live mycobacteria growing within antigen-presenting cells. Together, this information implicates the late endosome/MIIC as the primary location for antigen loading of CD1b. Our results provide a link between the intracellular trafficking properties of CD1b and its probable function in the host response to persistent intracellular pathogens such as M. tuberculosis. Experimental Procedures Cell Lines and Antibodies The antigen-specific T cell lines DN1 (Porcelli et al., 1992), DN.OR and CD8.TX (Stenger et al., 1997), and LDN5 (Moody et al., 1997) have been described. CD8.3 was derived from human peripheral blood lymphocytes selected for the CD81 phenotype. This line was established from cultures expanded by two rounds of stimulation with 2.3 mg/ml trehalose dimycolate (Sigma, St. Louis, MO) and autologous, irradiated monocytes previously cultured with 300 U/ml GM-CSF and 200 U/ml IL-4 to induce CD1 expression. Subsequently, CD8.3 was stimulated with trehalose dimycolate and random donor irradiated monocytes previously stimulated with GMCSF and IL-4. HeLa and THP-1 cells were obtained from the American Type Culture Collection (ATCC, Rockville, MD). C1R cells were obtained from Peter Cresswell (Yale University, New Haven, CT). The mouse monoclonal antibodies 4A7.6 (anti-CD1b, IgG2a [Olive et al., 1984]) and BCD1b3.1 (anti-CD1b, IgG1 [Behar et al., 1995]) have been described. ME-1 (anti-HLA–B27, –B7, and –Bw22; IgG1 [Ellis et al., 1982]) was purified from a hybridoma obtained from the ATCC. Rabbit anti-human Lamp-1 antiserum (Fukuda et al., 1988), rabbit anti-human HLA-DMb antiserum (Denzin et al., 1994), rabbit antihuman class II a chain antiserum (Neefjes et al., 1990), and rabbit

anti-human CD1b antiserum (Sugita et al., 1997) also have been previously described. Commercially available antibodies used were indocarbocyanine (CY3)–conjugated goat anti-mouse IgG and Texas Red (TR)–conjugated goat anti-rabbit IgG (Jackson Immunoresearch Laboratories, West Grove, PA); fluorescein isothiocyanate (FITC)–conjugated goat F(ab9)2 anti-mouse Ig and FITC–conjugated goat F(ab9)2 anti-rabbit IgG (Biosource International, Camarillo, CA); and tetramethylrhodamine isothiocyanate–conjugated goat antimouse IgG1–specific antibody and FITC-conjugated goat F(ab9)2 anti-mouse IgG2a–specific antibody (Southern Biotechnology Associates, Birmingham, AL). Construction of CD1b Mutants and Transfections Construction of CD1b.WT and CD1b.TD with subcloning into pSRaneo has been described previously (Porcelli et al., 1989; Sugita et al., 1996). cDNA for CD1b.Y311A was generated by polymerase chain reaction using CD1b.WT as template DNA. Primers used were 59-AGTGAACATGCCTTCCAGGGGCCGACC-39 and 59-TCGGGATC CTATGGGATATTCTGAGCTGACCGGCGCCTC-39. The amplified fragment was cloned and sequenced, and the EcoRI-to-BamHI region (39 terminus) was subcloned into similarly digested pSRaneoCD1b.WT to give the mutant construct. Construction of the HLA-B27 construct has been described (Sugita and Brenner, 1995). Polymerase chain reaction was used to generate B27/CYT.b using HLA-B2705 as template. Primers used were 59-ATGGCTGCGACGTGGGGCC-39 (plus strand within B27) and 59-TCTAGATCATGGGATATTCTGATATGACCGCCTCCTACAC ATCACAGCAGCG-39 (minus strand, cytoplasmic domain of CD1b fused to HLA-B27 transmembrane region). The amplified fragment was cloned and sequenced, and the SalI-to-XbaI region was subcloned into the HLA-B27 construct for expression. Electroporation was used (Gene Pulser Apparatus, Bio-Rad Laboratories, Hercules, CA) to introduce the plasmid expression constructs into suspension cells (C1R: 300 V, 960 mF; THP-1: 200 V, 960 mF), and CaPO4-mediated DNA precipitation was used for HeLa cells, followed by selection for G418 resistance and cloning. Immunofluorescence and Confocal Microscopy HeLa cells were grown for immunofluorescence on glass coverslips, fixed with 3.7% paraformaldehyde, permeabilized with 0.1% digitonin, and incubated with primary and secondary antibodies in 5% goat serum. Cover slips were mounted for observation on glass slides in Vectashield mounting medium (Vector Laboratories, Burlingame, CA) as an antifading agent. For confocal microscopy, cells were fixed and labeled as described above and then examined using a Leica TCS-NT confocal laser scanning microscope fitted with krypton and argon lasers. Cells were illuminated with 488 and 568 nm light after filtering through an acoustic optical device. Images of cells decorated with FITC and Texas Red were recorded simultaneously through separate optical detectors using a 530 6 15 nm bandpass filter and a 590 nm long pass filter, respectively. Pairs of images were superimposed for colocalization analysis (Rogers et al., 1993). Immunogold EM Procedures used for EM have been described previously (Peters et al., 1991a). Standard quantitative analysis (Liou et al., 1997) was performed on photographic prints of electron micrographs of 25 cells chosen at random for each cell type. Each print (12,0003, corresponding to approximately 38 mm 2) contained plasma membrane and an area of cytoplasm. A transparent sheet of single 1 cm2 lattice grid area was superimposed on the prints, and the plasma membrane and endosomal surface areas (arbitrary units) were estimated by scoring intersections with the grid. Endosomal gold particles counted were 762 for CD1b.WT (n 5 21 photographs) and 292 for CD1b.TD (n 5 20) on surface areas of 836 and 657 units, respectively. Similarly, the plasma membrane gold particle count was 150 for CD1b.WT and 1301 for CD1b.TD on surface lengths of 805 and 1005 units, respectively. Cytokine Assays T cell activation for cytokine release was done using 1 3 10 5 T cells/ well in 96-well U-bottomed plates stimulated with an equal number

Immunity 350

of C1R APCs with the stated concentration of antigen. For C1R assays, LDN5 T cells were incubated in 0.1 nM IL-2 and DN1 T cells in 0.75 nM IL-2 with 10 ng/ml phorbol myristate acetate (PMA), representing the optimal stimulation conditions for each T cell line and giving the highest signal-to-noise ratio. Qualitatively similar results were obtained for LDN5 and DN1 in the absence of IL-2 and PMA. For stimulation of CD8.3 to C1R cells, the cells were incubated with 0.5 nM IL-2 and 4 ng/ml PMA; for THP-1 cells, 0.1 nM IL-2 with no PMA was used. For recognition of antigen presented by cells infected with live M. tuberculosis., the respective THP-1 lines were activated overnight with 16 ng/ml PMA to enhance adhesion, detached with EDTA treatment, plated in a 96-well flat-bottomed plate at 2.5 3 10 4 cells/ well, and infected with M. tuberculosis H37Rv (five bacilli per cell). After 5 hr the cells were washed to remove noninternalized bacteria, and then duplicate samples were stained with auramine/rhodamine stain to verify intracellular infection, showing a multiplicity of infection of 3.2 to 4.8 with 61%–68% of cells infected depending on the experiment. For the assay, 1 3 10 4 T cells were added per well, and released cytokine levels were determined after 30–48 hr by enzymelinked immunosorbent assay in relation to recombinant cytokine standards (IFNg antibodies were from Endogen [Woburn, MA] and GM-CSF antibodies were produced from hybridomas BVD2–23B6.4 and BVD2–21C11.3 obtained from the ATCC). IFNg secretion by T cells in response to uninfected APCs was measured to distinguish nonspecific background reactivity from antigen-specific responses to infected cells. Uninfected THP-1 transfected with CD1b.WT gave background IFNg secretions of 0.28, 0.38, 0.41, and 0.46 ng/ml with CD8.TX, DN.OR, DN1, and LDN5 respectively, while the background responses for THP-1 transfected with CD1b.TD were 0.27, 0.42, 1.64, and 0.85 ng/ml. Western Blots C1R cells (3 3 106/ml) expressing the wild-type or various mutants of CD1b were lysed in 1% polyoxyethylene 9 laurylether (C12E9, Sigma) in lysis buffer (150 nM NaCl, 10 mM bicine [pH 8.2], 7.5 nM iodoacetamide, and 1 mM PMSF). Immunoprecipitation was carried out with 0.5 ml lysate samples using the indicated antibodies, followed by incubation with protein A–sepharose CL-4B beads (Pharmacia, Piscataway, NJ). Beads were washed four times and then treated with 200 U of PNGase F as directed by the manufacturer (New England Biolabs, Beverly, MA). The samples were suspended in 13 sample buffer containing 1.5% SDS and 5% b-mercaptoethanol, boiled, and resolved by 9% SDS-PAGE (polyacrylamide gel electrophoresis). The proteins were transferred to a polyvinylidene difluoride membrane (Millipore, Bedford, MA), and divided in half. The membrane containing CD1b was incubated with a 1:2000 dilution of rabbit antiserum against denatured human CD1b, and the other half containing MHC class II was incubated with a 1:500 dilution of rabbit antiserum against the MHC class II a chain. Immunoblotting was performed as described (Reedquist et al., 1996) using horseradish peroxidase–conjugated protein A (Zymed Laboratories, San Francisco, CA) as a second-step reagent and enhanced chemiluminescence (Du Pont, Boston, MA). A Molecular Dynamics Personal Densitometer was used to quantitate the intensity of the bands using an autoradiograph with linear signal. Acknowledgments We thank Jean Lai and Juliete Mervis for expert help with confocal microscopy. We also gratefully acknowledge the following colleagues for providing antisera: Minoru Fukuda (Lamp-1), Peter Cresswell (HLA-DMb), and Hidde Ploegh (class II a chain). R. M. J. was supported by a National Science Foundation predoctoral fellowship, D. B. M. by National Institute of Arthritis and Musculoskeletal Disease-National Institutes of Health grant AR01988, and R. A. R. by National Heart, Lung, and Blood Institute–National Institutes of Health grants HL33009 and HL43510. This research was supported by the Arthritis Foundation (S. A. P, Investigator Award), a grant from the American Cancer Society, and National Institutes of Health grant AI40135 to S. A. P. Received December 17, 1997; revised February 5, 1998.

References Bakke, O., and Dobberstein, B. (1990). MHC class II-associated invariant chain contains a sorting signal for endosomal compartments. Cell 63, 707–716. Bansal, A., and Gierasch, L.M. (1991). The NPXY internalization signal of the LDL receptor adopts a reverse-turn conformation. Cell 67, 1195–1201. Beckman, E.M., Porcelli, S.A., Morita, C.T., Behar, S.M., Furlong, S.T., and Brenner, M.B. (1994). Recognition of a lipid antigen by CD1-restricted ab1 T cells. Nature 372, 691–694. Beckman, E.M., Melian, A., Behar, S.M., Sieling, P.A., Chatterjee, D., Furlong, S.T., Matsumoto, R., Rosat, J.P., Modlin, R.L., and Porcelli, S.A. (1996). CD1c restricts responses of mycobacteria-specific T cells: evidence for antigen presentation by a second member of the human CD1 family. J. Immunol. 157, 2795–2803. Behar, S.M., Porcelli, S.A., Beckman, E.M., and Brenner, M.B. (1995). A pathway of costimulation that prevents anergy in CD28-T cells: B7-independent costimulation of CD1-restricted T cells. J. Exp. Med. 182, 2007–2018. Boll, W., Ohno, H., Songyang, Z., Rapoport, I., Cantley, L.C., Bonifacino, J.S., and Kirchhausen, T. (1996). Sequence requirements for the recognition of tyrosine-based endocytic signals by clathrin AP-2 complexes. EMBO J. 15, 5789–5795. Clemens, D.L., and Horwitz, M.A. (1995). Characterization of the Mycobacterium tuberculosis phagosome and evidence that phagosomal maturation is inhibited. J. Exp. Med. 181, 257–270. Denzin, L.K., Robbins, N.F., Carboy-Newcomb, C., and Cresswell, P. (1994). Assembly and intracellular transport of HLA-DM and correction of the class II antigen-processing defect in T2 cells. Immunity 1, 595–606. Eberle, W., Sander, C., Klaus, W., Schmidt, B., von Figura, K., and Peters, C. (1991). The essential tyrosine of the internalization signal in lysosomal acid phosphatase is part of a b turn. Cell 67, 1203–1209. Ellis, S.A., Taylor, C., and McMichael, A. (1982). Recognition of HLAB27 and related antigen by a monoclonal antibody. Hum. Immunol. 5, 49–59. Exley, M., Garcia, J., Balk, S.P., and Porcelli, S.A. (1997). Requirements for CD1d recognition by human invariant Va24 1 CD42CD82 T cells. J. Exp. Med. 186, 109–120. Fukuda, M., Viitala, J., Matteson, J., and Carlsson, S.R. (1988). Cloning of cDNAs encoding human lysosomal membrane glycoproteins, h-lamp-1 and h-lamp-2: comparison of their deduced amino acid sequences. J. Biol. Chem. 263, 18920–18928. Kaufmann, S.H.E. (1993). Immunity to intracellular bacteria. Annu. Rev. Immunol. 11, 129–163. Kawano, T., Cui, J., Koezuka, Y., Toura, I., Kaneko, Y., Motoki, K., Ueno, H., Nakagawa, R., Sato, H., Kondo, E., et al. (1997). CD1drestricted and TCR-mediated activation of Va14 NKT cells by glycosylceramides. Science 278, 1626–1629. Kleijmeer, M.J., Morkowski, S., Griffith, J.M., Rudensky, A.Y., and Geuze, H.J. (1997). Major histocompatibility complex class II compartments in human and mouse B lymphoblasts represent conventional endocytic compartments. J. Cell Biol. 139, 639–649. Liou, W., Geuze, H.J., Geelen, M.J., and Slot, J.W. (1997). The autophagic and endocytic pathways converge at the nascent autophagic vacuoles. J. Cell Biol. 136, 61–70. Marks, M.S., Roche, P.A., van Donselaar, E., Woodruff, L., Peters, P.J., and Bonifacino, J.S. (1995). A lysosomal targeting signal in the cytoplasmic tail of the beta chain directs HLA-DM to MHC class II compartments. J. Cell Biol. 131, 351–369. Marks, M.S., Ohno, H., Kirchhausen, T., and Bonifacino, S.J. (1997). Protein sorting by tyrosine-based signals: adapting to the Ys and wherefores. Trends Cell Biol. 7, 124–128. Melian, A., Beckman, E.M., Porcelli, S.A., and Brenner, M.B. (1996). Antigen presentation by CD1 and MHC-encoded class I–like molecules. Curr. Opinion Immunol. 8, 82–88. Moody, D.B., Reinhold, B.B., Guy, M.R., Beckman, E.M., Frederique, D.E., Furlong, S.T., Ye, S., Reinhold, V.N., Sieling, P.A., Modlin, R., et al. (1997). Structural requirements for glycolipid antigen recognition by CD1b-restricted T cells. Science 278, 283–288.

Functional Effects of CD1b Targeting 351

Neefjes, J.J., and Ploegh, H.L. (1992). Intracellular transport of MHC class II molecules. Immunol. Today 13, 179-189. Neefjes, J.J., Stollorz, V., Peters, P.J., Geuze, H.J., and Ploegh, H.L. (1990). The biosynthetic pathway of MHC class II but not class I molecules intersects the endocytic route. Cell 61, 171–183. Ohno, H., Stewart, J., Fournier, M.C., Bosshart, H., Rhee, I., Miyatake, S., Saito, T., Gallusser, A., Kirchhausen, T., and Bonifacino, J.S. (1995). Interaction of tyrosine-based sorting signals with clathrinassociated proteins. Science 269, 1872–1875. Olive, D., Dubreuil, P., and Mawas, C. (1984). Two distinct TL-like molecular subsets defined by monoclonal antibodies on the surface of human thymocytes with different expression on leukemia lines. Immunogenetics 20, 253–264. Pancholi, P., Mirza, A., Bhardwaj, N., and Steinman, R.M. (1993). Sequestration from immune CD41 T cells of mycobacteria growing in human macrophages. Science 260, 984–986. Peters, P.J., Borst, J., Oorschot, V., Fukuda, M., Krahenbuhl, O., Tschopp, J., Slot, J.W., and Geuze, H.J. (1991a). Cytotoxic T lymphocyte granules are secretory lysosomes, containing both perforin and granzymes. J. Exp. Med. 173, 1099–1109. Peters, P.J., Neefjes, J.J., Oorschot, V., Ploegh, H.L., and Geuze, H.J. (1991b). Segregation of MHC class II molecules from MHC class I molecules in the Golgi complex for transport to lysosomal compartments. Nature 349, 669–676. Pinet, V., Vergelli, M., Martin, R., Bakke, O., and Long, E.O. (1995). Antigen presentation mediated by recycling of surface HLA-DR molecules. Nature 375, 603–606. Porcelli, S.A. (1995). The CD1 family: a third lineage of antigenpresenting molecules. Adv. Immunol. 1995; 59, 1–98. Porcelli, S., Brenner, M.B., Greenstein, J.L., Balk, S.P., Terhorst, C., and Bleicher, P.A. (1989). Recognition of cluster of differentiation 1 antigens by human CD42CD82 cytolytic T lymphocytes. Nature 341, 447–450. Porcelli, S.A., Morita, C.T., and Brenner, M.B. (1992). CD1b restricts the response of human CD4282 T lymphocytes to a microbial antigen. Nature 360, 593–597. Prigozy, T.I., Sieling, P.A., Clemens, D., Stewart, P.L., Behar, S.M., Porcelli, S.A., Brenner, M.B., Modlin, R.L., and Kronenberg, M. (1997). The mannose receptor delivers lipoglycan antigens to endosomes for presentation to T cells by CD1b molecules. Immunity 6, 187–197. Reedquist, K.A., Fukazawa, T., Panchamoorthy, G., Langdon, W.Y., Shoelson, S.E., Druker, B.J., and Band, H. (1996). Stimulation through the T cell receptor induces Cbl association with Crk proteins and the guanine nucleotide exchange protein C3G. J. Biol. Chem. 271, 8435–8442. Rogers, R.A., Jack, R.M., and Furlong, S.T. (1993). Lipid and membrane protein transfer from human neutrophils to schistosomes is mediated by ligand binding. J. Cell Sci. 106, 485–491. Rohrer, J., Schweizer, A., Russell, D., and Kornfeld, S. (1996). The targeting of Lamp-1 to lysosomes is dependent on the spacing of its cytoplasmic tail tyrosine sorting motif relative to the membrane. J. Cell Biol. 132, 565–576. Sandoval, I.V., and Bakke, O. (1994). Targeting of membrane proteins to endosomes and lysosomes. Trends Cell Biol. 4, 292–297. Sieling, P.A., Chatterjee, D., Porcelli, S.A., Prigozy, T.I., Mazzaccaro, R.J., Soriano, T., Bloom, B.R., Brenner, M.B., Kronenberg, M., Brennan, P.J., et al. (1995). CD1-restricted T cell recognition of microbial lipoglycan antigens. Science 269, 227–230. Stenger, S., Mazzaccaro, R.J., Uyemura, K., Cho, S., Barnes, P.F., Rosat, J.P., Sette, A., Brenner, M.B., Porcelli, S.A., Bloom, B.R., and Modlin, R.L. (1997). Differential effects of cytolytic T cell subsets on intracellular infection. Science 276, 1684–1687. Sugita, M., and Brenner, M.B. (1995). Association of the invariant chain with major histocompatibility complex class I molecules directs trafficking to endocytic compartments. J. Biol. Chem. 270, 1443–1448. Sugita, M., Jackman, R.M., van Donselaar, E., Behar, S.M., Rogers, R.A., Peters, P.J., Brenner, M.B., and Porcelli, S.A. (1996). Cytoplasmic tail-dependent localization of CD1b antigen-presenting molecules to MIICs. Science 273, 349–352.

Sugita, M., Porcelli, S.A., and Brenner, M.B. (1997). Assembly and retention of CD1b heavy chains in the endoplasmic reticulum. J. Immunol. 159, 2358–2365. Sugita, M., Moody, D.B., Jackman, R.M., Grant, E.P., Rosat, J.P., Behar, S.M., Peters, P.J., Porcelli, S.A., and Brenner, M.B. (1998). CD1: a new paradigm for antigen presentation. Clin. Immunol. Immunopathol. 86, in press. Tsuchiya, S., Yamabe, M., Yamaguchi, Y., Kobayashi, Y., Konno, T., and Tada, K. (1980). Establishment and characterization of a human acute monocytic leukemia cell line (THP-1). Int. J. Cancer 26, 171–176. Williams, M.A., and Fukuda, M. (1990). Accumulation of membrane glycoproteins in lysosomes requires a tyrosine residue at a particular position in the cytoplasmic tail. J. Cell Biol. 111, 955–966. Xu, S., Cooper, A., Sturgill-Koszycki, S., van Heyningen, T., Chatterjee, D., Orme, I., Allen, P., and Russell, D.G. (1994). Intracellular trafficking in Mycobacterium tuberculosis and Mycobacterium avium-infected macrophages. J. Immunol. 153, 2568–2578. Zeng, Z.H., Castan˜o, A.R., Segelke, B.W., Stura, E.A., Peterson, P.A., and Wilson, I.A. (1997). Crystal structure of mouse CD1: an MHClike fold with a large hydrophobic binding groove. Science 277, 339–345.