Peptide-Independent Folding and CD8αα Binding by the Nonclassical Class I Molecule, Thymic Leukemia Antigen

Dominique A. Weber, Antoine Attinger, Christopher C. Kemball, Jerrod L. Wigal, Jan Pohl, Yi Xiong, Ellis L. Reinherz, Hilde Cheroutre, Mitchell Kronenberg and Peter E. Jensen
J Immunol November 15, 2002, 169 (10) 5708-5714; DOI: https://doi.org/10.4049/jimmunol.169.10.5708

Abstract

The nonclassical class I molecule, thymic leukemia (TL), has been shown to be expressed on intestinal epithelial cells and to interact with CD8+ intraepithelial T lymphocytes. We generated recombinant soluble TL (T18d) H chains in bacteria as inclusion bodies and refolded them with β2-microglobulin in the presence or absence of a random peptide library. Using a mAb, HD168, that recognizes a conformational epitope on native TL molecules, we observed that protein folds efficiently in the absence of peptide. Circular dichroism analysis demonstrated that TL molecules have structural features similar to classical class I molecules. Moreover, thermal denaturation experiments indicated that the melting temperature for peptide-free TL is similar to values reported previously for conventional class I-peptide complexes. Our results also show that CD8αα binding is not dependent on either TL-associated peptide or TL glycosylation.

Thymic leukemia (TL)3 Ag, originally identified as a tumor Ag (1), is an MHC class Ib molecule encoded by T3b, T3d, and T18d as well as other closely related alleles in the T region of the murine MHC (2). The tissue distribution of TL is quite limited in comparison to class Ia molecules. TL is highly expressed on immature thymocytes in some strains of mice but not others. It is expressed on epithelial cells in the small intestine of all strains (3, 4), suggesting that this nonclassical class I molecule may play a specialized role in mucosal immunity. The TL gene products are closely related, with >95% sequence similarity and there is evidence for evolutionary selection in mice to maintain conservation in the α1 and α2 domains of these molecules (5). TL is expressed at the cell surface in association with β2-microglobulin (β2m) and it is predicted to fold with a conformation similar to other MHC class I proteins. The CD8 binding site is conserved and it has been demonstrated that CD8 can serve as a coreceptor for CTL recognition of a chimeric class I molecule containing the TL α3 domain (6). In addition, there is evidence that TL may serve as a restriction element for TCR αβ and γδ T cells (7, 8, 9, 10, 11, 12, 13).

Recent work demonstrates that soluble TL produced in insect cells binds to the homodimeric form of CD8, CD8αα, which is expressed on a large subpopulation of intraepithelial lymphocytes (IEL) (13, 14). TL (T18d) tetramers selectively bind CD8αα+ IEL and transfectants expressing CD8αα. There is no evidence for a role for TCR in this interaction. Binding experiments with rTL monomers generated using a baculovirus expression system demonstrate that TL binds CD8αα with an affinity at least 10-fold higher than the affinity for CD8αβ heterodimers, the usual form of CD8 on T cells (14). TL has a low affinity for CD8αβ, similar to that measured for H-2Kb with either CD8αα or CD8αβ (14, 15). Functional studies with primary T cells and T cell hybridomas demonstrate that the interaction of CD8αα on T cells with TL on APC can strikingly promote TCR-mediated cytokine production (14). By contrast, the CD8αα-TL interaction does not enhance other functions of IEL, including proliferation or cytotoxicity. Thus, TL appears to regulate TCR-mediated cytokine production by CD8αα+ IEL through a mechanism not requiring direct TCR recognition of TL.

The issue of whether TL can bind peptides remains an open question. Conserved residues that form hydrogen bonds with the N and C termini of bound peptides in class I molecules (16, 17) are retained in TL. However, TL is efficiently expressed in TAP-deficient cell lines (18, 19) and thymocytes (20). This observation does not necessarily rule out a role for peptide ligands in TL assembly and transport. For example, the class Ib molecule Qa-1 is expressed efficiently on the surface of TAP-deficient cells (21, 22) despite its usual assembly with a TAP-dependent peptide derived from the leader sequence of other class I molecules (23) and its capacity to bind diverse peptide sequences (24). Studies with transgenic mice expressing TL driven by the H-2Kb promoter demonstrate that TL can act as a transplantation Ag and that TL-specific CTL expressing TCRαβ can be induced (7, 8, 10, 11, 13). These CTL recognize TL expressed in TAP-deficient RMAS and insect cell lines (12), consistent with peptide-independent recognition or presentation of unconventional Ags loaded on TL through a TAP-independent pathway. The latter possibility has precedent because Qa-1 can present insulin to TCRαβ T cells through a TAP-independent endosomal pathway (25). Peptide elution experiments with TL isolated from a radiolabeled cell line demonstrated the apparent presence of TL-associated peptides but no sequence information was obtained by Edman degradation, possibly because of peptide N-terminal modifications (8). No peptides were identified in elution experiments with soluble TL expressed in insect cells (18).

In the current study, we demonstrate that TL, expressed as inclusion bodies in Escherichia coli, can efficiently fold in vitro with β2m in the absence of peptides or other potential ligands. The empty TL molecules are shown to form a stable conformation and to bind to rCD8αα with high affinity. Tetramers generated with this protein bind a large fraction of IEL and this binding is blocked with mAb to CD8α.

Materials and Methods

Generation of soluble class I Ags

A cDNA construct (18) was used as template to amplify a truncated form of the T18d H chain. The primers used were: sense, 5′-TACCATATGGGCTCACACTCGCTGAGGTACTTC-3′, and antisense, 5′-GCGGGATCCACGAACAGTGGTCCTGTTGG-3′, tagged with NdeI and BamHI sites (in bold italics), respectively. The PCR product was cloned inframe with a biotin-binding cassette (BSP) at the 3′ terminus into pET23a vector as previously described (26). The protein-BSP was expressed as inclusion bodies in E. coli BL21(DE3)pLys (Stratagene, La Jolla, CA) after induction with 0.4 M isopropyl-β-d-thiogalactopyranoside. T18d and human β2m (26) inclusion bodies were purified from bacterial lysates and solubilized in 8 M urea. Solubilized inclusion bodies were filtered through PD-10 columns (Amersham Pharmacia, Uppsala, Sweden) to remove low m.w. materials. After further dilution with buffered 6 M guanidine hydrochloride, 1 μM H chain, and 2 μM β2m were injected into a 200-ml refolding mix in the presence or absence of 20 mg of randomized 9-mer peptide libraries as described (26). The refolded T18d monomers were concentrated in an Amicon chamber (Beverly, MA) with a 10K-exclusion membrane and separated from aggregated H chain and free β2m on a S300 gel filtration column (Pharmacia Biotech, Peapack, NJ). Heterodimers were dialyzed against PBS at a neutral pH. Similarly, a truncated H chain of HLA-A2 was generated and refolded with or without peptide. HLA-A2-BSP and human β2m constructs have been described (26). All refolded proteins were analyzed by SDS-PAGE (12.5% gel) and visualized with Coomassie Blue.

Quantification of refolded T18d

Refolded heterodimers were quantified by a sandwich immunoassay. Briefly, 96-well Immulon II plates were coated overnight at 4°C with 5 μg/ml mAb to human β2m (Immunotech, Westbrook, ME) diluted in borate-buffered saline at pH 8.0. Plates were blocked 30 min at room temperature with 50 mM Tris, pH 7.5, containing 5% nonfat dry milk, 0.1% BSA, and 0.1% Tween 20 (MTB) and thoroughly washed in Tween 20 plus TBS (TTBS) (500 mM Tris-HCl, pH 7.5, and 0.1% Tween 20) before sample addition. Samples were diluted in MTB containing 0.2% Nonidet P-40 (MTBN) and incubated in the wells for 1 h at 4°C. The plates were washed again as above. HD168 rat mAb (27) diluted in MTBN to a final concentration of 20 μg/ml was added to the wells and incubated for 1 h at 4°C. After washing, 5 μg/ml biotin-labeled goat anti-mouse Ab (Southern Biotechnology Associates, Birmingham, AL) was added for 1 h at 4°C followed by a wash and the addition of 100 ng/ml europium-labeled streptavidin (Wallac, Gaithersburg, MD) as previously described (28). Detection occurred at 615 nm using a 1230 ARCUS time-resolved fluorometer (Wallac). Results are expressed as fluorescence counts per second (cps × 10−3).

Circular dichroism (CD) and thermal denaturation

CD spectra were measured on an AVIV 60DS spectropolarimeter (Aviv Associates, Lakewood, NJ) equipped with a thermoelectric cell holder. Refolded (T18d) samples were diluted in 0.15 M PBS, pH 7.0, to a final concentration of 250 μg/ml, determined by absorbance at 280 nm, using an extinction coefficient of 2.0. Far-UV spectra were recorded in a 0.1-cm path-length cuvette at 25°C using a step size of 1 nm, a bandwidth of 1.5 nm, and a time constant of 2.0 s. The spectra shown in Fig. 3⇓ represent the average of three experiments, each representing the average of three repetitive scans with a subtracted PBS background. Far-UV spectra are given as the mean residue ellipticity [θ]r. Standard heating/cooling cycling of the samples was used to demonstrate the reversibility of the thermal unfolding. The CD spectra of each sample were first measured at 25°C, then the sample was heated to 79°C (above the midpoint temperature of the thermal unfolding transition (Tm) of T18d), immediately cooled to 25°C, and spectra were recorded again after an equilibration time of 1 h. The far-UV spectra recorded at high temperatures were measured separately to avoid permanent denaturation of the heterodimers due to long incubation times at high temperatures.

Thermal denaturation curves were recorded by monitoring the CD signal at 223 nm as a function of temperature in a 0.1-cm path-length cuvette. The temperature was raised from 25°C to 79°C in 2°C increments, with each increment followed by an equilibration time of 30 s. The recording time was 30 s. Each thermal melting curve represents the average of three independent experiments.

Averaged spectra and thermal denaturation curves were smoothed by a least squares polynomial fit using the Aviv 60DS software, v4.1t (polynomial order of 5, sliding window of 5). The Tm was determined from the interpolated maximum of a plot of dθ/dT vs T (θ is ellipticity).

Peptide libraries

Random peptide libraries were assembled by Fmoc-based solid-phase synthesis at the 50 μmol per channel scale on a Wang resin using an Advanced ChemTech model 3480 multiple peptide synthesizer (Louisville, KY). Acylation of the resin was affected using a 6-fold molar excess of preformed hydroxybenzotriazole esters of all coded F-moc-protected amino acids (excluding cysteine) as described previously (29). A 50% molar excess of F-moc-Val, Fmoc-Arg(Mtr), and F-moc-Ile over other Fmoc-amino acids was used to compensate for their lower acylation rates (29, 30). After cleavage and deprotection (Reagent K, 2 h, 25°C), the peptides were precipitated with diethyl ether (−20°C, 4 h). The precipitate was collected by centrifugation and solubilized in neat trifluoroacetic acid (TFA), diluted with H2O, and lyophilized. These libraries were obtained in the form of their TFA salts. Quantitative amino acid analysis and N-terminal sequence analysis confirmed their expected composition and the randomness of their sequences.

Pool peptide sequencing

T18d and HLA-A2 were refolded together with human β2m in the presence or absence of 9-mer peptide libraries as described above. PD-10 columns were used to eliminate low m.w. contaminants of bacterial origin present in the inclusion body preparations. Refolded complexes of T18d and HLA-A2 (500 μg each) were concentrated in Amicon chambers with a 10K exclusion membrane. Peptide was eluted by diluting the samples in 2% TFA/10% acetonitrile and separated from H chains and β2m using a Centricon 3 (Amicon) concentrator pretreated with 2% TFA/10% acetonitrile. Eluted peptides were dried by speed vacuum centrifugation. Edman degradation of T18d and HLA-A2-bound peptides was performed in a mode Procise cLC peptide sequencer (PE Biosystems, Wellesley, MA) operated in the pulsed liquid mode as previously described (29, 31). Eluted peptides were analyzed by microbore reverse-phase HPLC and separated using a Zorbax SB-C18 column (Frankfurt, Germany) as previously described (24). Briefly, the column eluate was monitored at 210 and 280 nm, the fractions absorbing at 210 nm were manually collected, and matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) mass spectrometry analysis was performed as described (24).

Tetramer staining

Refolded T18d and HLA-A2 monomers were biotinylated enzymatically by incubation with purified BirA for 12 h at 25°C as described (26). T18d tetramers were prepared by mixing the biotinylated protein with streptavidin-PE) at a molar ratio of 4:1. Tetramerization was confirmed by gel electrophoresis. IELs, isolated as previously described (32), and inguinal lymph node CD8 T cells were isolated from C57BL/6 mice and incubated with FcR mAb 2.4G2 to prevent nonspecific staining. Approximately 105 IELs or lymph node cells were stained with 1 μg of PE-conjugated tetramer protein and FITC-labeled anti-CD8 mAb, 53-6.7 (BD PharMingen, San Diego, CA), and were analyzed on a FACScan (BD Biosciences, San Jose, CA). For blocking of tetramer staining, Rat-anti-CD8α mAb RM2200 (Caltag Laboratories, San Francisco, CA) was used at 1 μg/50 μl of resuspended cells. The cells were preincubated with the mAb for 10 min before tetramer staining.

Surface plasmon resonance binding studies

Binding measurements were conducted on a Biacore X instrument (Biacore International, Uppsala, Sweden) at 25°C at 5 μl/min. Capture of soluble CD8αα and CD8αβ molecules on the flow cell has been described (14). The sensograms represent subtracted data (experimental flow cell−control cell). Steady state binding (Req) was determined by averaging the plateau response phase of the binding curve. To determine Kd, Req data were plotted against T18d concentration.

Results

T18d folds efficiently with β2m in the absence of peptide.

The question of whether TL can bind peptide Ags has not been resolved (8, 18). We addressed this issue by generating recombinant soluble T18d H chain molecules in E. coli and refolding this protein with β2m in the presence or absence of a fully random nonameric peptide library containing all natural amino acids except Cys. This approach was successfully used to investigate the peptide binding motif of another class Ib molecule, Qa-1 (24), in addition to class Ia molecules (33, 34). In Fig. 1⇓, we show chromatographs of the components present in the refolding reactions resolved by S-300 gel filtration. Misfolded protein aggregates remaining in the sample after concentration are eluted in the first peak; the second peak, indicated by an arrow, represents the properly folded class I molecules, eluting as a 40–45 kDa complex. Excess free β2m elutes in the third peak. Peptides are eliminated from the samples during concentration of the folding mix through a 10-kDa exclusion membrane. The data are representative of several refolding experiments. Similar amounts of folded T18d protein complexes were obtained, as measured by UV absorbance at 280 nm, whether proteins were folded in the presence or absence of the peptide library (Fig. 1⇓A). By contrast, peptide was required for in vitro folding of HLA-A2 (Fig. 1⇓B). The yield of folded T18d generated in the absence of peptide was similar to that obtained with HLA-A2 folded in the presence of the peptide library. The yield was ∼8–10% of total protein added to the folding reaction. We routinely obtained 5–7 mg of refolded protein per liter of solution.

FIGURE 1.
FIGURE 1.

Chromatograms of refolded class I proteins resolved by S-300 gel filtration. A, Chromatograms of concentrated refolding reactions (400 ml) of T18d in the presence (solid line) or absence (dotted line) of a fully random 9-mer peptide library. Misfolded proteins either precipitate and are eliminated from the reaction by centrifugation or form soluble aggregates migrating in the void volume (first peak). The arrow indicates the peak of class I heterodimer (40–45 kDa). Excess of β2m is present in the third peak. B, Chromatograms of HLA-A2 concentrated in the same conditions as above in the presence (solid line) or absence (dotted line) of the same peptide library.

Refolded T18d and HLA-A2 proteins were analyzed by SDS-PAGE (Fig. 2⇓). The proteins were resolved into two bands: one band migrating at ∼30 kDa represents the truncated, unglycosylated form of H chain and a second band migrating at 10 kDa corresponds to β2m. There was no difference in the stoichiometry between these complexes. Samples were analyzed in immunoassays using mAb to β2m to capture refolded class I molecules and mAb HD168, which recognized native T18d, for detection. Similar signals were observed with T18d that had been folded in the presence or absence of peptide, whereas only background was observed with control HLA-A2 and unfolded T18d controls (Fig. 2⇓). These results demonstrated that the T18d H chain assembles with β2m and that the refolded protein adopts a native conformation as defined by reactivity with the HD168 mAb. We concluded that T18d folds efficiently in vitro in the absence of peptide or other potential ligands and that nonameric peptides do not further promote the folding reaction.

FIGURE 2.
FIGURE 2.

T18d H chains assemble with β2m and refold into a native conformation. A, Purified heterodimers were analyzed by SDS-PAGE under reducing conditions. The 30 and 10 kDa bands represent unglycosylated H chains and β2m, respectively. B, Immunoassay detecting native T18d. Heterodimers were captured in 96-well plates precoated with mAb to β2m. Biotinylated HD168, mAb specific for native T18d, was added and detected by europium-avidin fluorescence. HD168 does not recognize HLA-A2 or denatured T18d protein in inclusion bodies.

Thermal stability of T18d

Far-UV CD spectroscopy has been used to analyze the structure and thermal stability of a variety of classical and nonclassical MHC molecules (17, 35, 36, 37, 38, 39, 40). The far-UV spectra of T18d are shown in Fig. 3⇓. The spectra are nearly identical for T18d folded in the presence or absence of peptide and they are very similar to spectra previously reported for other MHC class I molecules. These results confirm that T18d folded in the absence of peptide assumes a class I-like conformation with similar secondary structure. Thermal denaturation experiments were performed to analyze the stability of empty T18d and determine whether protein folded in the presence of peptide acquires increased stability. When spectra were measured at 79°C, there was a loss of CD signal corresponding to a change of conformation due to heat denaturation (Fig. 3⇓). After heating to 79°C, the samples were cooled rapidly and allowed to renature slowly over a 1-h incubation at room temperature. The subsequent spectra taken at 25°C are similar to those obtained with unheated samples, demonstrating that thermal denaturation is largely reversible under the conditions used. Thermal denaturation profiles demonstrated a Tm of 54°C, reflecting dissociation and unfolding of the H chain (35). The Tm is identical for T18d folded in the presence or absence of the nonameric peptide library. A second transition with Tm >60°C characterized by a sign reversal of the CD signal probably represents unfolding of free β2m (35, 40). The 54°C Tm of empty T18d is similar to that reported for H-2Kd peptide complexes (54–57°C) (35, 38) and greater than that reported for the class Ib protein T10 (49°C) (40), which does not bind peptides. Thus empty T18d molecules are relatively stable and folding the protein in the presence of peptide does not increase its stability.

FIGURE 3.
FIGURE 3.

Far-UV CD spectra and thermal denaturation of T18d/β2m (0.25 mg/ml, pH 7.0) refolded in the presence or absence of a 9-mer random peptide library. Spectra include scans of native, unfolded, and renatured T18d/β2m. Thermal denaturation curves were obtained by monitoring the change in CD signal at 223 nm in the range of 25–79°C, in 2°C increments.

Analysis of peptides eluted from HLA-A2 and T18d

Although T18d folded in the presence or absence of peptide exhibit identical thermal stability, we further analyzed the protein to confirm the absence of bound peptides. Low m.w. material acid eluted from HLA-A2 and T18d, folded in the presence of peptide library, was analyzed by MALDI-TOF mass spectrometry. A m/z profile consistent with a highly complex mixture of nonameric peptides was obtained with acid-eluted material from HLA-A2 (Fig. 4⇓). This material was subjected to several cycles of Edman degradation, revealing a significant enrichment in Met and Leu at position 2, consistent with the HLA-A2 peptide-binding motif. By contrast, very few species were identified in the acid eluate of T18d that could be derived from the input peptide library (Fig. 4⇓). This material was fractionated by capillary reverse-phase HPLC and one peak was successfully sequenced (GSLHHILD). However, this peptide does not appear to bind T18d (data not shown). These results support the conclusion that peptides from the nonamer library do not bind to T18d during folding in vitro.

FIGURE 4.
FIGURE 4.

Acid eluates from refolded class I heterodimers were analyzed by MALDI-TOF mass spectrometry. Delayed extraction reflectron MALDI-TOF mass spectra of the eluates of HLA-A2 + random 9-mer peptide library (top panel) and T18d + same peptide library (bottom panel). Two-hundred fifty microliters of each eluate was concentrated using a SpeedVac evaporator, loaded on C18 reversed phase ZipTip in 0.1% TFA and eluted with 70% acetonitrile/0.1% aqueous TFA. The eluates were analyzed as described in Materials and Methods.

Peptide-free T18d molecules bind CD8αα

Recent experiments demonstrate that tetramers generated from soluble T18d molecules expressed in insect cells bind a large fraction of IEL through a TCR-independent interaction with CD8αα homodimers expressed on IEL (14). Although it seems likely that the insect cell-derived protein does not contain bound peptides (18), the presence in this material of peptides loaded through a TAP-independent mechanism or other Ags such as glycolipids cannot be excluded. To definitively address this issue, we generated tetramers with T18d molecules from bacterial inclusion bodies refolded in the absence of peptides or other potential ligands. The bacteria-derived tetramers were observed to stain a large fraction of both TCRαβ+ and TCRγδ+ IEL isolated from B6 mice (Fig. 5⇓, A and B), but not lymph node T cells (Fig. 5⇓E), which express exclusively CD8αβ rather than CD8αα. Indeed, almost all IEL that express CD8α bind the tetramer. T18d tetramer staining of IEL was completely inhibited with an appropriate blocking CD8α-specific mAb (Fig. 5⇓, C and D). The mAb used for costaining CD8α (53-6.7) does not block tetramer staining.

FIGURE 5.
FIGURE 5.

Flow cytometry analysis of IELs (A and B) and lymph node cells (E) from a B6 mouse with streptavidin-PE-labeled T18d tetramers together with FITC-coupled CD8α Ab 53-6.7. A blocking Ab specific for CD8α completely inhibits tetramer staining (C and D).

To further analyze the binding of ligand-free T18d molecules to CD8αα, direct binding studies were performed by surface plasmon resonance. Bacteria-derived T18d monomer binding to CD8αα immobilized on a biosensor chip showed fast association and dissociation rates, with an equilibrium Kd of 11 μM (Fig. 6⇓). This value is very similar to the value (Kd = 12 μM) previously obtained with insect cell-derived TL monomers (14). Thus, we conclude that neither peptides nor other ligands associated with the peptide-binding groove of T18d are required for binding to CD8αα molecules.

FIGURE 6.
FIGURE 6.

T18d binding to immobilized CD8αα measured by surface plasmon resonance. Binding sensograms are shown on the left and a plot of equilibrium binding as a function of T18d concentration is shown on the right. Binding of 1.5, 3.1, 6.2, 10.3, 15.5, and 31 μM TL to immobilized CD8αα was analyzed. Each curve represents the binding obtained with increasing concentrations of TL protein. Each measurement is representative of at least two independent experiments. I, injection; D, dissociation phase.

Discussion

The results presented in this study demonstrate that the nonclassical class I molecule TL (T18d) spontaneously folds in vitro in the absence of peptides or other potential ligands such as glycolipids. In contrast to classical class I molecules and the T23-encoded class Ib molecule Qa-1, folding efficiency is not enhanced by including a random nonameric peptide library as a source of peptides to stabilize the peptide-binding groove. The peptide-free TL molecules react with a mAb that recognizes a conformational epitope in TL, and far-UV CD spectroscopy analysis demonstrates that the secondary structure of empty TL molecules is similar to other MHC class I molecules. Thus, neither peptides nor other ligands are required to generate properly folded TL molecules.

Thermal melting experiments demonstrate a Tm of 54°C for TL generated in folding reactions in the absence of peptides, identical to the value obtained with TL generated in the presence of a peptide library. Analysis of low m.w. material acid-eluted from TL demonstrated only a few candidate peptides. One of these was sequenced and we could not confirm that a synthetic version of this peptide binds TL. By contrast, a diverse mixture of peptides was present in acid eluates of control HLA-A2 molecules generated with the same random nonamer peptide library. Thus, we conclude that empty TL molecules are generated even in folding reactions that contain a source of peptides. We cannot rule out the possibility that TL can bind shorter or longer peptides, or peptides with specific modifications not represented in the synthetic peptide library. The latter is true for the class Ib molecule, H-2 M3, which selectively binds N-formylated peptides (41, 42). Despite this formal possibility, it is clear that empty TL molecules are quite stable.

The Tm value of 54°C is similar to the values (54 and 57°C) reported for H-2Kd bound to specific peptide ligands (35, 38). Most class Ia molecules, like HLA-A2 in the present study, fail to assemble in the absence of appropriate peptides, indicating that peptide is an integral component of the properly folded molecules. H-2Kd is unusual in that it can be generated as an empty molecule. However, the Tm of empty H-2Kd (45°C) (35) is considerably lower than the value for peptide-free TL. Thermal denaturation studies have been performed with one additional nonclassical class I molecule, T10 (40). T10 and the closely related T22 (94% identity) are recognized by a subpopulation of γδ T cells in mice without a requirement for bound peptide or other components (40, 43). The crystal structure of T22 demonstrates a severely modified MHC-like fold, lacking a classical peptide-binding groove (44). The thermal stability of T10 (Tm = 49°C) is significantly lower than that of TL. These comparisons support the hypothesis that TL functions not as an Ag-presenting molecule, but rather as an invariant ligand that regulates lymphocyte function.

Recent work has demonstrated that TL binds selectively to CD8αα homodimers expressed on IEL, promoting cytokine production induced by TCR signaling (14). This function is independent of direct TCR recognition of TL. Given its selective localization on intestinal epithelial cells, one can conclude that TL acts in a site-specific manner to regulate mucosal immune responses. Data presented in this study formally demonstrate that TL molecules, lacking bound peptide, glycosylation, other posttranslational modifications specific to eukaryotic cells, or other associated proteins, bind CD8αα with the same affinity as TL molecules generated in insect cells. Previous studies with transgenic mice expressing TL (T3b) driven by an H-2Kb promoter demonstrated that TL can act as a transplantation Ag and that both αβ and γδ CTL can recognize TL (7, 8, 9, 10, 11, 12, 13). It is unlikely that these previous results can be explained by expansion of a population of CD8αα+ T cells that are stimulated by TL through a TCR-independent mechanism. It was reported that Abs to either αβ TCR or TL block the killing of TL-expressing target cells by CD8α+ CTL (10). In addition, TL-specific CD4−CD8α− CTL can be generated in CD8 T cell-depleted mice transplanted with skin from TL transgenic donors (10). Thus, it appears that under certain experimental conditions, T cells with true specificity for TL can be generated and that TL recognition does not require expression of CD8αα.

The potential requirement for TL-bound peptides, lipids, or other antigenic moieties in TCR-mediated recognition of TL remains to be established. It has been demonstrated that TL-specific CTL clones kill TAP-deficient RMAS and insect cells (12). In addition, TL tetramers generated using a baculovirus expression system were used to enrich TL-specific CTL from polyclonal populations generated in mixed leukocyte cultures (13). These findings are most consistent with the idea that T cells recognize empty TL molecules. Alternatively, some ubiquitous unconventional ligand could be involved. It is possible that there is a population of “natural” T cells with specificity for TL, in analogy to NK-T cells selected by interaction with CD1d expressed on thymocytes. Joyce et al. (20) reported a numerical increase in TCRαβ+NK1.1+ in thymi of mice expressing TL in the thymus. However, specificity for TL was not demonstrated. It is clear that the vast majority of NK T cells recognize CD1d (45, 46). Nevertheless, the possibility that a dedicated population of T cells exists with specificity for TL needs further investigation.

Acknowledgments

We thank Dr. Keith Wilkinson, Department of Biochemistry, Emory University (Atlanta, GA), for his help and useful discussions regarding circular dichroism measurements, Julie Jun for excellent technical support, Olga Stuchlick for technical assistance with MALDI-TOF mass spectrometry analyses of eluted peptides, and Olga Naidenko for her expertise in surface plasmon resonance analysis.

Footnotes

  • ↵1 This work was supported by National Institutes of Health Grants AI33614 and AI30554 (to P.E.J.), AI50263 (to H.C.), AI45022 (to E.L.R.), and a grant from the Swiss National Fund (to A.A.).

  • ↵2 Address correspondence and reprint requests to Dr. Peter E. Jensen, Department of Pathology and Laboratory Medicine, School of Medicine, Emory University, Woodruff Memorial Building, Room 7-309, 1639 Pierce Drive, Atlanta, GA 30322. E-mail address: pjensen{at}emory.edu

  • ↵3 Abbreviations used in this paper: TL, thymic leukemia; β2m, β2-microglobulin; IEL, intestinal epithelial lymphocyte; CD, circular dichroism; MALDI-TOF, matrix-assisted laser desorption ionization time-of-flight.

  • Received August 1, 2002.
  • Accepted September 11, 2002.

References

  1. ↵
    Old, L. J., E. A. Boyse, E. Stockert. 1963. Antigenic properties of experimental leukemias. I. Serological studies in vitro with spontaneous and radiation-induced leukemias. J. Natl. Cancer Inst. 31: 977
    OpenUrlAbstract/FREE Full Text
  2. ↵
    Cheroutre, H., H. R. Holcombe, S. Tangri, A. R. Castano, M. Teitell, J. E. Miller, S. Cardell, C. Benoist, D. Mathis, W. D. Huse, et al 1995. Antigen-presenting function of the TL antigen and mouse CD1 molecules. Immunol. Rev. 147: 31
    OpenUrlCrossRefPubMed
  3. ↵
    Hershberg, R., P. Eghtesady, B. Sydora, K. Brorson, H. Cheroutre, R. Modlin, M. Kronenberg. 1990. Expression of the thymus leukemia antigen in mouse intestinal epithelium. Proc. Natl. Acad. Sci. USA 87: 9727
    OpenUrlAbstract/FREE Full Text
  4. ↵
    Wu, M., L. van Kaer, S. Itohara, S. Tonegawa. 1991. Highly restricted expression of the thymus leukemia antigens on intestinal epithelial cells. J. Exp. Med. 174: 213
    OpenUrlAbstract/FREE Full Text
  5. ↵
    Obata, Y., Y. Satta, K. Moriwaki, T. Shiroishi, H. Hasegawa, T. Takahashi, N. Takahata. 1994. Structure, function, and evolution of mouse TL genes, nonclassical class I genes of the major histocompatibility complex. Proc. Natl. Acad. Sci. USA 91: 6589
    OpenUrlAbstract/FREE Full Text
  6. ↵
    Teitell, M., M. F. Mescher, C. A. Olson, D. R. Littman, M. Kronenberg. 1991. The thymus leukemia antigen binds human and mouse CD8. J. Exp. Med. 174: 1131
    OpenUrlAbstract/FREE Full Text
  7. ↵
    Morita, A., T. Takahashi, E. Stockert, E. Nakayama, T. Tsuji, Y. Matsudaira, L. J. Old, Y. Obata. 1994. TL antigen as a transplantation antigen recognized by TL-restricted cytotoxic T cells. J. Exp. Med. 179: 777
    OpenUrlAbstract/FREE Full Text
  8. ↵
    Sharma, P., S. Joyce, K. A. Chorney, J. W. Griffith, R. H. Bonneau, F. D. Wilson, C. A. Johnson, R. A. Flavell, M. J. Chorney. 1996. Thymus-leukemia antigen interacts with T cells and self-peptides. J. Immunol. 156: 987
    OpenUrlAbstract
  9. ↵
    Tsujimura, K., T. Takahashi, A. Morita, H. Hasegawa-Nishiwaki, S. Iwase, Y. Obata. 1996. Positive selection of γδ CTL by TL antigen expressed in the thymus. J. Exp. Med. 184: 2175
    OpenUrlAbstract/FREE Full Text
  10. ↵
    Tsuji, K., Y. Obata, T. Takahashi, J. Arata, E. Nakayama. 1997. Requirement of CD4 T cells for skin graft rejection against thymus leukemia (TL) antigen and multiple epitopes on the TL molecule recognized by CD4 T cells. J. Immunol. 159: 159
    OpenUrlAbstract
  11. ↵
    Tsujimura, K., T. Takahashi, S. Iwase, Y. Matsudaira, Y. Kaneko, H. Yagita, Y. Obata. 1998. Two types of anti-TL (thymus leukemia) CTL clones with distinct target specificities: differences in cytotoxic mechanisms and accessory molecule requirements. J. Immunol. 160: 5253
    OpenUrlAbstract/FREE Full Text
  12. ↵
    Tsujimura, K., Y. Obata, S. Iwase, Y. Matsudaira, S. Ozeki, T. Takahashi. 2000. The epitope detected by cytotoxic T lymphocytes against thymus leukemia (TL) antigen is TAP independent. Int. Immunol. 12: 1217
    OpenUrlAbstract/FREE Full Text
  13. ↵
    Tsujimura, K., Y. Obata, Y. Matsudaira, S. Ozeki, K. Yoshikawa, S. Saga, T. Takahashi. 2001. The binding of thymus leukemia (TL) antigen tetramers to normal intestinal intraepithelial lymphocytes and thymocytes. J. Immunol. 167: 759
    OpenUrlAbstract/FREE Full Text
  14. ↵
    Leishman, A. J., O. V. Naidenko, A. Attinger, F. Koning, C. J. Lena, Y. Xiong, H. C. Chang, E. Reinherz, M. Kronenberg, H. Cheroutre. 2001. T cell responses modulated through interaction between CD8αα and the nonclassical MHC class I molecule, TL. Science 294: 1936
    OpenUrlAbstract/FREE Full Text
  15. ↵
    Devine, L., L. Rogozinski, O. V. Naidenko, H. Cheroutre, P. B. Kavathas. 2002. The complementarity-determining region-like loops of CD8α interact differently with β2-microglobulin of the class I molecules H-2Kb and thymic leukemia antigen, while similarly with their α3 domains. J. Immunol. 168: 3881
    OpenUrlAbstract/FREE Full Text
  16. ↵
    Matsumura, M., D. H. Fremont, P. A. Peterson, I. A. Wilson. 1992. Emerging principles for the recognition of peptide antigens by MHC class I molecules. Science 257: 927
    OpenUrlAbstract/FREE Full Text
  17. ↵
    Bouvier, M., D. C. Wiley. 1994. Importance of peptide amino and carboxyl termini to the stability of MHC class I molecules. Science 265: 398
    OpenUrlAbstract/FREE Full Text
  18. ↵
    Holcombe, H. R., A. R. Castano, H. Cheroutre, M. Teitell, J. K. Maher, P. A. Peterson, M. Kronenberg. 1995. Nonclassical behavior of the thymus leukemia antigen: peptide transporter-independent expression of a nonclassical class I molecule. J. Exp. Med. 181: 1433
    OpenUrlAbstract/FREE Full Text
  19. ↵
    Rodgers, J. R., V. Mehta, R. G. Cook. 1995. Surface expression of β2-microglobulin-associated thymus-leukemia antigen is independent of TAP2. Eur. J. Immunol. 25: 1001
    OpenUrlCrossRefPubMed
  20. ↵
    Joyce, S., I. Negishi, A. Boesteanu, A. D. DeSilva, P. Sharma, M. J. Chorney, D. Y. Loh, L. Van Kaer. 1996. Expansion of natural (NK1+) T cells that express αβ T cell receptors in transporters associated with antigen presentation-1 null and thymus leukemia antigen positive mice. J. Exp. Med. 184: 1579
    OpenUrlAbstract/FREE Full Text
  21. ↵
    Robinson, P. J., P. J. Travers, A. Stackpoole, L. Flaherty, H. Djaballah. 1998. Maturation of Qa-1b class I molecules requires β2-microglobulin but is TAP independent. J. Immunol. 160: 3217
    OpenUrlAbstract/FREE Full Text
  22. ↵
    Vance, R. E., J. R. Kraft, J. D. Altman, P. E. Jensen, D. H. Raulet. 1998. Mouse CD94/NKG2A is a natural killer cell receptor for the nonclassical major histocompatibility complex (MHC) class I molecule Qa-1b. J. Exp. Med. 188: 1841
    OpenUrlAbstract/FREE Full Text
  23. ↵
    Aldrich, C. J., A. DeCloux, A. S. Woods, R. J. Cotter, M. J. Soloski, J. Forman. 1994. Identification of a Tap-dependent leader peptide recognized by alloreactive T cells specific for a class Ib antigen. Cell 79: 649
    OpenUrlCrossRefPubMed
  24. ↵
    Kraft, J. R., R. E. Vance, J. Pohl, A. M. Martin, D. H. Raulet, P. E. Jensen. 2000. Analysis of Qa-1b peptide binding specificity and the capacity of CD94/NKG2A to discriminate between Qa-1-peptide complexes. J. Exp. Med. 192: 613
    OpenUrlAbstract/FREE Full Text
  25. ↵
    Tompkins, S. M., J. R. Kraft, C. T. Dao, M. J. Soloski, P. E. Jensen. 1998. Transporters associated with antigen processing (TAP)-independent presentation of soluble insulin to α/β T cells by the class Ib gene product, Qa-1b. J. Exp. Med. 188: 961
    OpenUrlAbstract/FREE Full Text
  26. ↵
    Altman, J. D., P. A. H. Moss, P. J. R. Goulder, D. H. Barouch, M. G. McHeyzer-Williams, J. I. Bell, A. J. McMichael, M. M. Davis. 1996. Phenotypic analysis of antigen-specific T lymphocytes. Science 274: 94
    OpenUrlAbstract/FREE Full Text
  27. ↵
    Obata, Y., Y. T. Chen, E. Stockert, L. J. Old. 1985. Structural analysis of TL genes of the mouse. Proc. Natl. Acad. Sci. USA 82: 5475
    OpenUrlAbstract/FREE Full Text
  28. ↵
    Tompkins, S. M., P. A. Rota, J. C. Moore, P. E. Jensen. 1993. A europium fluoroimmunoassay for measuring binding of antigen to class II MHC glycoproteins. J. Immunol. Methods 163: 209
    OpenUrlCrossRefPubMed
  29. ↵
    Flynn, G. C., J. Pohl, M. T. Flocco, J. E. Rothman. 1991. Peptide-binding specificity of the molecular chaperone BiP. Nature 353: 726
    OpenUrlCrossRefPubMed
  30. ↵
    Lebl, M., V. Krchnak. 1997. Synthetic peptide libraries. Methods Enzymol. 289: 336
    OpenUrlCrossRefPubMed
  31. ↵
    Pohl, J.. 1994. Sequence analysis of peptide resins from Boc/benzyl solid-phase synthesis. Methods Mol. Biol. 36: 107
    OpenUrlPubMed
  32. ↵
    Leishman, A. J., L. Gapin, M. Capone, E. Palmer, H. R. MacDonald, M. Kronenberg, H. Cheroutre. 2002. Precursors of functional MHC class I- or class II-restricted CD8αα+ T cells are positively selected in the thymus by agonist self-peptides. Immunity 16: 355
    OpenUrlCrossRefPubMed
  33. ↵
    Davenport, M. P., K. J. Smith, D. Barouch, S. W. Reid, W. M. Bodnar, A. C. Willis, D. F. Hunt, A. V. Hill. 1997. HLA class I binding motifs derived from random peptide libraries differ at the COOH terminus from those of eluted peptides. J. Exp. Med. 185: 367
    OpenUrlAbstract/FREE Full Text
  34. ↵
    Stevens, J., K. H. Wiesmuller, P. J. Barker, P. Walden, G. W. Butcher, E. Joly. 1998. Efficient generation of major histocompatibility complex class I-peptide complexes using synthetic peptide libraries. J. Biol. Chem. 273: 2874
    OpenUrlAbstract/FREE Full Text
  35. ↵
    Fahnestock, M. L., I. Tamir, L. Narhi, P. J. Bjorkman. 1992. Thermal stability comparison of purified empty and peptide-filled forms of a class I MHC molecule. Science 258: 1658
    OpenUrlAbstract/FREE Full Text
  36. ↵
    Weiss, G. A., R. J. Valentekovich, E. J. Collins, D. N. Garboczi, W. S. Lane, S. L. Schreiber, D. C. Wiley. 1996. Covalent HLA-B27/peptide complex induced by specific recognition of an aziridine mimic of arginine. Proc. Natl. Acad. Sci. USA 93: 10945
    OpenUrlAbstract/FREE Full Text
  37. ↵
    Dedier, S., S. Reinelt, T. Reitinger, G. Folkers, D. Rognan. 2000. Thermodynamic stability of HLA-B*2705 peptide complexes: effect of peptide and major histocompatibility complex protein mutations. J. Biol. Chem. 275: 27055
    OpenUrlAbstract/FREE Full Text
  38. ↵
    Reich, Z., J. D. Altman, J. J. Boniface, D. S. Lyons, H. Kozono, G. Ogg, C. Morgan, M. M. Davis. 1997. Stability of empty and peptide-loaded class II major histocompatibility complex molecules at neutral and endosomal pH: comparison to class I proteins. Proc. Natl. Acad. Sci. USA 94: 2495
    OpenUrlAbstract/FREE Full Text
  39. ↵
    Raghavan, M., L. N. Gastinel, P. J. Bjorkman. 1993. The class I major histocompatibility complex related Fc receptor shows pH-dependent stability differences correlating with immunoglobulin binding and release. Biochemistry 32: 8654
    OpenUrlCrossRefPubMed
  40. ↵
    Crowley, M. P., Z. Reich, N. Mavaddat, J. D. Altman, Y. Chien. 1997. The recognition of the nonclassical major histocompatibility complex (MHC) class I molecule, T10, by the γδ T cell, G8. J. Exp. Med. 185: 1223
    OpenUrlAbstract/FREE Full Text
  41. ↵
    Shawar, S. M., J. M. Vyas, J. R. Rodgers, R. G. Cook, R. R. Rich. 1991. Specialized functions of major histocompatibility complex class I molecules. II. Hmt binds N-formylated peptides of mitochondrial and prokaryotic origin. J. Exp. Med. 174: 941
    OpenUrlAbstract/FREE Full Text
  42. ↵
    Wang, C. R., A. R. Castano, P. A. Peterson, C. Slaughter, K. F. Lindahl, J. Deisenhofer. 1995. Nonclassical binding of formylated peptide in crystal structure of the MHC class Ib molecule H2–M3. Cell 82: 655
    OpenUrlCrossRefPubMed
  43. ↵
    Crowley, M. P., A. M. Fahrer, N. Baumgarth, J. Hampl, I. Gutgemann, L. Teyton, Y. Chien. 2000. A population of murine γδ T cells that recognize an inducible MHC class Ib molecule. Science 287: 314
    OpenUrlAbstract/FREE Full Text
  44. ↵
    Wingren, C., M. P. Crowley, M. Degano, Y. Chien, I. A. Wilson. 2000. Crystal structure of a γδ T cell receptor ligand T22: a truncated MHC-like fold. Science 287: 310
    OpenUrlAbstract/FREE Full Text
  45. ↵
    Mendiratta, S. K., W. D. Martin, S. Hong, A. Boesteanu, S. Joyce, L. Van Kaer. 1997. CD1d1 mutant mice are deficient in natural T cells that promptly produce IL-4. Immunity 6: 469
    OpenUrlCrossRefPubMed
  46. ↵
    Chen, Y. H., N. M. Chiu, M. Mandal, N. Wang, C. R. Wang. 1997. Impaired NK1+ T cell development and early IL-4 production in CD1-deficient mice. Immunity 6: 459
    OpenUrlCrossRefPubMed
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section