T Cell Recognition of Hapten
1999; Elsevier BV; Volume: 274; Issue: 6 Linguagem: Inglês
10.1074/jbc.274.6.3622
ISSN1083-351X
AutoresBenedikt M. Kessler, Olivier Michielin, Christopher Blanchard, Irina Apostolou, Christaiane Delarbre, Gabriel Gachelin, Claude Grégoire, Bernard Malissen, J.-C. Cerottini, Florian Μ. Wurm, Martin Karplus, Immanuel F. Luescher,
Tópico(s)Immunotherapy and Immune Responses
ResumoTo elucidate the structural basis of T cell recognition of hapten-modified antigenic peptides, we studied the interaction of the T1 T cell antigen receptor (TCR) with its ligand, the H-2Kd-bound Plasmodium bergheicircumsporozoite peptide 252–260 (SYIPSAEKI) containing photoreactive 4-azidobenzoic acid (ABA) on P. berghei circumsporozoite Lys259. The photoaffinity-labeled TCR residue(s) were mapped as Tyr48 and/or Tyr50 of complementary determining region 2β (CDR2β). Other TCR-ligand contacts were identified by mutational analysis. Molecular modeling, based on crystallographic coordinates of closely related TCR and major histocompatibility complex I molecules, indicated that ABA binds strongly and specifically in a cavity between CDR3α and CDR2β. We conclude that TCR expressing selective Vβ and CDR3α sequences form a binding domain between CDR3α and CDR2β that can accommodate nonpeptidic moieties conjugated at the C-terminal portion of peptides binding to major histocompatibility complex (MHC) encoded proteins. To elucidate the structural basis of T cell recognition of hapten-modified antigenic peptides, we studied the interaction of the T1 T cell antigen receptor (TCR) with its ligand, the H-2Kd-bound Plasmodium bergheicircumsporozoite peptide 252–260 (SYIPSAEKI) containing photoreactive 4-azidobenzoic acid (ABA) on P. berghei circumsporozoite Lys259. The photoaffinity-labeled TCR residue(s) were mapped as Tyr48 and/or Tyr50 of complementary determining region 2β (CDR2β). Other TCR-ligand contacts were identified by mutational analysis. Molecular modeling, based on crystallographic coordinates of closely related TCR and major histocompatibility complex I molecules, indicated that ABA binds strongly and specifically in a cavity between CDR3α and CDR2β. We conclude that TCR expressing selective Vβ and CDR3α sequences form a binding domain between CDR3α and CDR2β that can accommodate nonpeptidic moieties conjugated at the C-terminal portion of peptides binding to major histocompatibility complex (MHC) encoded proteins. CD8+ cytotoxic T lymphocytes (CTL) 1The abbreviations used are: CTL, cytotoxic T lymphocyte(s); ABA, 4-azidobenzoic acid; CDR, complementary determining region; MHC, major histocompatibility complex; TCR, T cell antigen receptor(s); PbCS, P. berghei circumsporozoite; HPLC, high pressure liquid chromatography; PAGE, polyacrylamide gel electrophoresis; PCR, polymerase chain reaction; LZ, leucine zipper; IASA, iodo-4-azidosalicylic acid. recognize by means of their T cell antigen receptor (TCR) antigenic peptides, usually 8–10 amino acids long, bound to major histocompatibility complex (MHC) class I molecules on target cells (1Matis L.A. Annu. Rev. Immunol. 1990; 8: 65-82Crossref PubMed Google Scholar, 2Bjorkman P.J. Cell. 1997; 89: 167-170Abstract Full Text Full Text PDF PubMed Scopus (81) Google Scholar, 3Fremont D.H. Stura E.A. Matsumura M. Peterson P.A. Wilson I.A. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 2479-2483Crossref PubMed Scopus (241) Google Scholar, 4Madden D.R. Garboczi D.N. Wiley D.C. Cell. 1993; 75: 693-708Abstract Full Text PDF PubMed Scopus (641) Google Scholar). However, CD8+ (and CD4+) T cells can also recognize antigenic peptides containing nonpeptidic moieties, such as carbohydrates or haptens, like trinitrophenyl, azobenzenearsonate, fluorescein, or phenylazides (5Nalefski E.A. Rao A. J. Immunol. 1993; 150: 3806-3816PubMed Google Scholar, 6Nalefski E.A. Kasibhatla S. Rao A. J. Exp. Med. 1992; 175: 1553-1563Crossref PubMed Scopus (36) Google Scholar, 7Siliciano R.F. Hemesath T.J. Pratt J.C. Dintzis R.Z. Dintzis H.M. Acuto O. Shin H.S. Reinherz E.L. Cell. 1986; 47: 161-171Abstract Full Text PDF PubMed Scopus (60) Google Scholar, 8Harding C.V. Kihlberg J. Elofsson M. Magnusson G. Unanue E.R. J. Immunol. 1993; 151: 1425-2419Google Scholar, 9Jensen T. Hansen P. Galli-Stampino L. Mouritsen S. Frische K. Meinjohanns E. Meldal M. Werdelin O. J. Immunol. 1997; 158: 3769-3778PubMed Google Scholar, 10Kohler J. Hartmann U. Grimm R. Pflugfelder U. Weltzien H.U. J. Immunol. 1997; 158: 591-597PubMed Google Scholar, 11Martin S. von Bonin A. Fessler C. Pflugfelder U. Weltzien H.U. J. Immunol. 1993; 151: 678-687PubMed Google Scholar, 12Martin S. Ortmann B. Pflugfelder U. Birsner U. Wetzien H.U. J. Immunol. 1992; 149: 2569-2575PubMed Google Scholar, 13Anjuère F. Kuznetsov D. Romero P. Cerottini J.-C. Jongeneel V. Luescher I.F. J. Biol. Chem. 1997; 272: 8505-8514Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar, 14Romero P. Casanova J.-L. Cerottini J.-C. Maryanski J.L. Luescher I.F. J. Exp. Med. 1993; 177: 1245-1256Crossref Scopus (21) Google Scholar, 15Luescher I.F. Anjuère F. Peitsch M.C. Jongeneel V. Cerottini J.-C. Romero P. Immunity. 1995; 3: 51-63Abstract Full Text PDF PubMed Scopus (50) Google Scholar). Such T cells can be readily elicited and play a role in diseases, e.g. allergies, contact dermatitis, and eczema (16Hess D.A. Rieder M.J. Ann. Pharmacother. 1997; 31: 1378-1387Crossref PubMed Scopus (46) Google Scholar). The recognition of modified peptides is highly specific, and even small changes in the hapten or carbohydrate moiety can dramatically affect antigen recognition (13Anjuère F. Kuznetsov D. Romero P. Cerottini J.-C. Jongeneel V. Luescher I.F. J. Biol. Chem. 1997; 272: 8505-8514Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar, 14Romero P. Casanova J.-L. Cerottini J.-C. Maryanski J.L. Luescher I.F. J. Exp. Med. 1993; 177: 1245-1256Crossref Scopus (21) Google Scholar, 15Luescher I.F. Anjuère F. Peitsch M.C. Jongeneel V. Cerottini J.-C. Romero P. Immunity. 1995; 3: 51-63Abstract Full Text PDF PubMed Scopus (50) Google Scholar, 16Hess D.A. Rieder M.J. Ann. Pharmacother. 1997; 31: 1378-1387Crossref PubMed Scopus (46) Google Scholar, 17Kessler B.M. Bassanini P. Cerottini J.-C. Luescher I.F. J. Exp. Med. 1997; 185: 629-640Crossref PubMed Scopus (47) Google Scholar, 18Preckel T. Gimm R. Martin S. Weltzien H.U. J. Exp. Med. 1997; 185: 1803-1813Crossref PubMed Scopus (42) Google Scholar). This is reminiscent of immunoglobulins, which can be raised against and specifically bind such structures (19Bedzyk W.D. Herron J.N. Edmundson A.B. Voss Jr., E.W. J. Biol. Chem. 1990; 265: 133-138Abstract Full Text PDF PubMed Google Scholar, 20Strong R.K. Campbell R. Rose D.R. Petsko G.A. Sharon J. Margolies M.N. Biochemistry. 1991; 30: 3739-3748Crossref PubMed Scopus (86) Google Scholar). While x-ray crystallographic studies have revealed how antibodies bind haptens, little is known about how TCR do this. This is of particular interest, because TCR genes, unlike immunoglobulin genes, have no somatic mutations allowing affinity maturation. Moreover, TCR need to recognize hapten or carbohydrate moieties in the context of an MHC-peptide complex in a predefined orientation (21Garboczi D.N. Ghosh P. Utz U. Fan Q.R. Biddison W.E. Wiley D.C. Nature. 1996; 384: 134-141Crossref PubMed Scopus (1212) Google Scholar, 22Garcia K.C. Degano M. Stanfield R.L. Brunmark A. Jackson M.R. Peterson P.A. Teyton L. Wilson L.A. Science. 1996; 274: 209-219Crossref PubMed Scopus (1067) Google Scholar, 23Garcia K.C. Degano M. Pease L.R. Huang M. Peterson P.A. Teyton L. Wilson I.A. Science. 1998; 279: 1166-1172Crossref PubMed Scopus (592) Google Scholar, 24Ding Y.H. Smith K.J. Garboczi D.N. Utz U. Biddison W.E. Wiley D.C. Immunity. 1998; 8: 403-411Abstract Full Text Full Text PDF PubMed Scopus (393) Google Scholar, 25Teng M.K. Smolyar A. Tse A.G.D. Liu J.H. Liu J. Hussey R.E. Nathenson S.G. Chang H.C. Reinherz E.L. Wang J.H. Curr. Biol. 1998; 8: 409-410Abstract Full Text Full Text PDF PubMed Google Scholar). Available three-dimensional structures of TCR-ligand complexes revealed a consensus “diagonal” TCR-ligand orientation, in which the MHC-bound peptide runs diagonally between the CDR3 loops, extending from CDR1α to CDR1β (21Garboczi D.N. Ghosh P. Utz U. Fan Q.R. Biddison W.E. Wiley D.C. Nature. 1996; 384: 134-141Crossref PubMed Scopus (1212) Google Scholar, 23Garcia K.C. Degano M. Pease L.R. Huang M. Peterson P.A. Teyton L. Wilson I.A. Science. 1998; 279: 1166-1172Crossref PubMed Scopus (592) Google Scholar, 24Ding Y.H. Smith K.J. Garboczi D.N. Utz U. Biddison W.E. Wiley D.C. Immunity. 1998; 8: 403-411Abstract Full Text Full Text PDF PubMed Scopus (393) Google Scholar, 25Teng M.K. Smolyar A. Tse A.G.D. Liu J.H. Liu J. Hussey R.E. Nathenson S.G. Chang H.C. Reinherz E.L. Wang J.H. Curr. Biol. 1998; 8: 409-410Abstract Full Text Full Text PDF PubMed Google Scholar). In this orientation, the CDR3 loops can interact extensively with peptide side chains, which are mainly located in the center of MHC molecules, as well as with residues of the MHC α-helices. The α-helices of MHC class I molecules are elevated at the N-terminal portions; therefore, the approximately planar surface of the TCR ligand binding site can realize the best contact with the ligand in a diagonal orientation (21Garboczi D.N. Ghosh P. Utz U. Fan Q.R. Biddison W.E. Wiley D.C. Nature. 1996; 384: 134-141Crossref PubMed Scopus (1212) Google Scholar). Hapten or carbohydrates conjugated with antigenic peptides are part of the epitope recognized by TCR (5Nalefski E.A. Rao A. J. Immunol. 1993; 150: 3806-3816PubMed Google Scholar, 8Harding C.V. Kihlberg J. Elofsson M. Magnusson G. Unanue E.R. J. Immunol. 1993; 151: 1425-2419Google Scholar, 9Jensen T. Hansen P. Galli-Stampino L. Mouritsen S. Frische K. Meinjohanns E. Meldal M. Werdelin O. J. Immunol. 1997; 158: 3769-3778PubMed Google Scholar, 10Kohler J. Hartmann U. Grimm R. Pflugfelder U. Weltzien H.U. J. Immunol. 1997; 158: 591-597PubMed Google Scholar, 11Martin S. von Bonin A. Fessler C. Pflugfelder U. Weltzien H.U. J. Immunol. 1993; 151: 678-687PubMed Google Scholar, 12Martin S. Ortmann B. Pflugfelder U. Birsner U. Wetzien H.U. J. Immunol. 1992; 149: 2569-2575PubMed Google Scholar, 13Anjuère F. Kuznetsov D. Romero P. Cerottini J.-C. Jongeneel V. Luescher I.F. J. Biol. Chem. 1997; 272: 8505-8514Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar, 14Romero P. Casanova J.-L. Cerottini J.-C. Maryanski J.L. Luescher I.F. J. Exp. Med. 1993; 177: 1245-1256Crossref Scopus (21) Google Scholar, 15Luescher I.F. Anjuère F. Peitsch M.C. Jongeneel V. Cerottini J.-C. Romero P. Immunity. 1995; 3: 51-63Abstract Full Text PDF PubMed Scopus (50) Google Scholar). TCR specific for hapten-modified antigenic peptides typically exhibit preferential usage of certain Vβ/Jα, and/or specific junctional sequences (13Anjuère F. Kuznetsov D. Romero P. Cerottini J.-C. Jongeneel V. Luescher I.F. J. Biol. Chem. 1997; 272: 8505-8514Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar, 15Luescher I.F. Anjuère F. Peitsch M.C. Jongeneel V. Cerottini J.-C. Romero P. Immunity. 1995; 3: 51-63Abstract Full Text PDF PubMed Scopus (50) Google Scholar,18Preckel T. Gimm R. Martin S. Weltzien H.U. J. Exp. Med. 1997; 185: 1803-1813Crossref PubMed Scopus (42) Google Scholar). We used as hapten photoreactive 4-azidobenzoic acid (ABA). This allowed assessment of TCR-ligand binding by TCR photoaffinity labeling and identification of the photoaffinity-labeled site(s),i.e. the contact(s) of the hapten with the TCR (13Anjuère F. Kuznetsov D. Romero P. Cerottini J.-C. Jongeneel V. Luescher I.F. J. Biol. Chem. 1997; 272: 8505-8514Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar, 15Luescher I.F. Anjuère F. Peitsch M.C. Jongeneel V. Cerottini J.-C. Romero P. Immunity. 1995; 3: 51-63Abstract Full Text PDF PubMed Scopus (50) Google Scholar). We have previously generated and characterized two families of H-2Kd-restricted CTL clones, specific for two different photoreactive derivatives of the Plasmodium bergheicircumsporozoite peptide PbCS-(252–260) (SYIPSAEKI) (13Anjuère F. Kuznetsov D. Romero P. Cerottini J.-C. Jongeneel V. Luescher I.F. J. Biol. Chem. 1997; 272: 8505-8514Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar, 15Luescher I.F. Anjuère F. Peitsch M.C. Jongeneel V. Cerottini J.-C. Romero P. Immunity. 1995; 3: 51-63Abstract Full Text PDF PubMed Scopus (50) Google Scholar). In one peptide derivative, ABA was conjugated with PbCS Lys259, whereas P-255 was replaced by Lys(ABA) in the other. In addition, to prevent Kd-peptide derivative complex dissociation, PbCS Ser252 was replaced with iodo-4-azidosalicylic acid (IASA), which upon selective photoactivation permitted covalent attachment of the peptide derivative to Kd (26Luescher I.F. Cerottini J.-C. Romero P. J. Biol. Chem. 1994; 269: 5574-5582Abstract Full Text PDF PubMed Google Scholar). The ABA, but not the IASA group, was part of the epitope recognized by these CTL. The two families of CTL clones were non-cross-reactive, and exhibited different TCR sequences (13Anjuère F. Kuznetsov D. Romero P. Cerottini J.-C. Jongeneel V. Luescher I.F. J. Biol. Chem. 1997; 272: 8505-8514Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar, 15Luescher I.F. Anjuère F. Peitsch M.C. Jongeneel V. Cerottini J.-C. Romero P. Immunity. 1995; 3: 51-63Abstract Full Text PDF PubMed Scopus (50) Google Scholar). In this study, we describe the interaction of the TCR of the T1 CTL clones with its ligand, Kd-bound IASA-YIPSAEK(ABA)I. Using mutational analysis, mapping of the photoaffinity-labeled site(s) and molecular modeling, we identified a specific binding mode, how the T1 TCR binds the ABA group. We propose that this binding principle has universal aspects. Amino acids and other chemicals were obtained from Bachem Finechemicals AG (Bubendorf, Switzerland), Sigma Chemie (Buchs, Switzerland), and Neosystems (Strasbourg, France). Synthesis and characterization of peptide derivatives was performed as described previously (15Luescher I.F. Anjuère F. Peitsch M.C. Jongeneel V. Cerottini J.-C. Romero P. Immunity. 1995; 3: 51-63Abstract Full Text PDF PubMed Scopus (50) Google Scholar, 17Kessler B.M. Bassanini P. Cerottini J.-C. Luescher I.F. J. Exp. Med. 1997; 185: 629-640Crossref PubMed Scopus (47) Google Scholar, 26Luescher I.F. Cerottini J.-C. Romero P. J. Biol. Chem. 1994; 269: 5574-5582Abstract Full Text PDF PubMed Google Scholar). HPLC-purified peptide derivatives were reconstituted in PBS at 1 mm. The specific radioactivity of 125I conjugates was approximately 2000 Ci/mmol. For the cytolytic assay,51Cr-labeled P815 cells (5 × 103cells/well) were incubated for 1 h at 37 °C in medium containing 10-fold dilutions of peptide derivatives, followed by UV irradiation at ≥350 nm. Cloned T1 CTL (1.5 × 104cells/well) were added, and after 4 h of incubation at 37 °C, released 51Cr was determined. The specific lysis was calculated as 100 × ((experimental − spontaneous release)/(total − spontaneous release)). The relative antigenic activities were calculated by dividing the concentration of IASA-YIPSAEK(ABA)I required for half-maximal lysis by that required for the variant peptide derivatives. These values were normalized by division with the corresponding relative Kd competitor activities (14Romero P. Casanova J.-L. Cerottini J.-C. Maryanski J.L. Luescher I.F. J. Exp. Med. 1993; 177: 1245-1256Crossref Scopus (21) Google Scholar, 15Luescher I.F. Anjuère F. Peitsch M.C. Jongeneel V. Cerottini J.-C. Romero P. Immunity. 1995; 3: 51-63Abstract Full Text PDF PubMed Scopus (50) Google Scholar, 17Kessler B.M. Bassanini P. Cerottini J.-C. Luescher I.F. J. Exp. Med. 1997; 185: 629-640Crossref PubMed Scopus (47) Google Scholar). All photoaffinity labeling procedures were performed as described previously (15Luescher I.F. Anjuère F. Peitsch M.C. Jongeneel V. Cerottini J.-C. Romero P. Immunity. 1995; 3: 51-63Abstract Full Text PDF PubMed Scopus (50) Google Scholar, 17Kessler B.M. Bassanini P. Cerottini J.-C. Luescher I.F. J. Exp. Med. 1997; 185: 629-640Crossref PubMed Scopus (47) Google Scholar). In brief, for TCR photoaffinity labeling, 107 cpm of Kd-125IASA-YIPSAEK(ABA)I were incubated with 107 T1 CTL on ice for 3 h, followed by UV irradiation at 312 ± 40 nm. For peptide mapping, T1 CTL (4 × 107) were incubated likewise with Kd-SYIPSAEK(125IASA)I (1 mCi) in 2 ml of medium containing β2-microglobulin (2.5 μg/ml). After UV irradiation at ≥350 nm, cells were washed twice and lysed in phosphate-buffered saline (1 × 107 cells/ml) containing 0.7% Nonidet P-40, HEPES, phenylmethylsulfonyl fluoride, leupeptin, and iodoacetamide. The detergent-soluble fractions were subjected to immunoprecipitation with anti-TCR Cβ monoclonal antibody H57–597. The immunoprecipitates were analyzed by SDS-PAGE (10%, reducing conditions) and quantified using a PhosphorImager and the ImageQuant software (Molecular Dynamics, Inc., Sunnyvale, CA). S.D. values were calculated from 2–4 experiments. T1 α and β cDNAs extending from the 5′ terminus up to, but not including, bases encoding the extracellular membrane-proximal cysteine residues were generated by reverse transcription on total T1 CTL RNA followed by polymerase chain reaction (PCR) amplification. The DNA fragments encoding a linker sequence and leucine zipper (LZ) components were generated by using oligonucleotides and PCR on templates pACID and pBASE (27Chang H.C. Bao Z.Z. Yao Y. Tse A.G.D. Goyarts E.C. Madsen M. Kawasaki E. Brauer P.P. Sacchettini J.C. Nathenson S.G. Reinherz E.L. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 11408-11412Crossref PubMed Scopus (144) Google Scholar). The T1 TCR-leucine zipper cDNAs were prepared by using recombinant PCR on these templates. The T1αLZ and T1βLZ cDNA containing basic and acidic LZ, respectively, were cloned into pCR-script (Stratagene) and subcloned into the EcoRI site of the mammalian expression vector pCI-neo (Promega). All PCR amplifications were performed usingPfu DNA polymerase (Stratagene), and both strands of cloned inserts were sequenced and found to be error-free. TCR mutants were generated using the QuickChange site-directed mutagenesis kit (Stratagene) following the suppliers instructions. 293T cells (ATCC) were transfected with pT1αLZ and pT1βLZ DNA (1:2 ratio) for transient expression of soluble αβT1 TCR following published procedures (28Jordan M. Schallhorn A. Wurm F.M. Nucleic Acids Res. 1996; 24: 596-601Crossref PubMed Scopus (736) Google Scholar). After 2 days, supernatants were harvested, and T1 TCR concentrations were equalized. Preparation of T1 single chain Fv cDNA constructs and protein expression were performed as described previously (29Grégoire C. Lin S.Y. Mazza G. Rebai N. Luescher I.F. Malissen B. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 7184-7189Crossref PubMed Scopus (18) Google Scholar). A full-length Kd cDNA cloned in pKC expression vector (promotor SV40, Hanahan) was double-digested withHindIII and XbaI in order to excise the sequences encoding the cytoplasmic and transmembrane domains up to nucleotide 966. The vector was religated using aHindIII–XbaI linker containing a stop codon, giving rise to translated C terminus RWKLA-stop. This KdcDNA was cloned into the pcDNA3 expression vector (Invitrogen) under control of the cytomegalovirus promoter. Site-directed mutagenesis was performed using the TransformerTM kit (CLONTECH) according to the manufacturer's instructions. Full-length cDNA coding for β2m was prepared and inserted in the same vector using the proper linkers (30Godeau F. Casanova J.-L. Fairchild K.D. Saucier C. Delarbre C. Gachelin G. Kourilsky P. Res. Immunol. 1991; 142: 409-416Crossref PubMed Scopus (6) Google Scholar). All procedures have been described previously (13Anjuère F. Kuznetsov D. Romero P. Cerottini J.-C. Jongeneel V. Luescher I.F. J. Biol. Chem. 1997; 272: 8505-8514Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar, 15Luescher I.F. Anjuère F. Peitsch M.C. Jongeneel V. Cerottini J.-C. Romero P. Immunity. 1995; 3: 51-63Abstract Full Text PDF PubMed Scopus (50) Google Scholar). Enzymes were obtained from Boehringer Mannheim (Rotkreuz, Switzerland) and used as recommended (31Keesey J. Biochemicals for Protein Research. Boehringer Mannheim, Indianapolis, IN1987: 87-108Google Scholar). In brief, photoaffinity-labeled T1 TCR was reduced, alkylated, and reconstituted in 500 μl of 100 mm Tris, pH 8.0 (for tryptic digests) or 100 mm phosphate buffer, pH 7.8 (for V8 digests) containing 10% acetonitrile. Aliquots of enzymes (10 μg) were added in 12-h intervals, and after 48 h of incubation at 37 °C, the digests were subjected to reverse phase HPLC on an analytical C-18 column (4 × 250 mm, 5-μm particle size, Vydac, Hisperia, CA). The column was eluted with a linear gradient of acetonitrile in 0.1% trifluoroacetic acid, rising within 1 h from 0 to 75%. Elution of radioactivity was monitored by γ-counting of 1-ml fractions. For destructive digestion the double-labeled V8 and Asp-N digest fragment was reconstituted in 300 μl of citrate buffer (50 mm, pH 5.5) containing 50 mm NaCl and Nonidet P-40 (0.01%) and incubated at 37 °C for 36 h with cathepsin C and carboxypeptidases P. Enzymes (5 μg) were added every 12 h. For biosynthetic labeling of T1 TCR with [3H]tyrosine, CD8 α/β-transfected T1.4 T cell hybridomas (0.7 × 106) were incubated in tyrosine-deficient Dulbecco's modified Eagle's medium supplemented with fetal calf serum (5%) and 5 mCi of [3H]tyrosine (NEN Life Science Products; specific activity of 142 Ci/mmol) at 37 °C for 20 h. The washed cells were photoaffinity-labeled with Kd-SYIPSAEK(125IASA)I (specific radioactivity of 20 Ci/mmol). The M r values of the labeled digest fragments were assessed by SDS-PAGE as described (32Schragger A. von Jagow G. Anal. Biochem. 1987; 166: 368-379Crossref PubMed Scopus (10505) Google Scholar). A homology model of the T1 TCR and the Kd-SYIPSAEK(ABA)I complex was built using the MODELLER program (33Sali A. Blundell T.L. J. Mol. Biol. 1993; 234: 779-815Crossref PubMed Scopus (10636) Google Scholar) based on the crystal coordinates of TCR A6 (Vα2.3, Jα24; Vβ12.3, Jβ2.1)-HLA-A2-Tax peptide complex (21Garboczi D.N. Ghosh P. Utz U. Fan Q.R. Biddison W.E. Wiley D.C. Nature. 1996; 384: 134-141Crossref PubMed Scopus (1212) Google Scholar), TCR 2C (Vα3, Jα58; Vβ8.2, Jβ2.4) (22Garcia K.C. Degano M. Stanfield R.L. Brunmark A. Jackson M.R. Peterson P.A. Teyton L. Wilson L.A. Science. 1996; 274: 209-219Crossref PubMed Scopus (1067) Google Scholar, 23Garcia K.C. Degano M. Pease L.R. Huang M. Peterson P.A. Teyton L. Wilson I.A. Science. 1998; 279: 1166-1172Crossref PubMed Scopus (592) Google Scholar), TCR 14.3 β-chain (Vβ8.2, Jβ2.1) (34Bentley G.A. Boulot G. Karjalainen K. Mariuzza R.A. Science. 1995; 267: 1984-1987Crossref PubMed Scopus (263) Google Scholar), TCR 1934.4 Vα (Vα4.2) (35Fields B.A. Ober B. Malchiodi E.L. Lebedeva M.I. Braden B.C. Ysern X. Kim J.K. Shao X. Ward E.S. Mariuzza R.A. Science. 1995; 270: 1821-1824Crossref PubMed Scopus (179) Google Scholar), and H2-Kb (3Fremont D.H. Stura E.A. Matsumura M. Peterson P.A. Wilson I.A. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 2479-2483Crossref PubMed Scopus (241) Google Scholar). The related sequences of corresponding chains were aligned using a dynamic programming method implemented in the MODELLER program (33Sali A. Blundell T.L. J. Mol. Biol. 1993; 234: 779-815Crossref PubMed Scopus (10636) Google Scholar). An all atom model of the complex was built using MODELLER by satisfaction of spatial restraints obtained from the alignment and parameters in the program. A distance restraint was introduced initially between the phenyl rings of ABA and β-Tyr48 and β-Tyr50, respectively. Side chain orientations were optimized using a backbone-dependent rotamer library (36Dunbrack R.L. Karplus M. Nat. Struct. Biol. 1994; 1: 334-340Crossref PubMed Scopus (297) Google Scholar, 37Bower J.M. Cohen F.E. Dunbrack R.L. J. Mol. Biol. 1997; 267: 1268-1282Crossref PubMed Scopus (488) Google Scholar). CDR1β and CDR2β loops were not subsequently refined, since their conformation was modeled from the TCR 2C, which has the same Vβ8.2 as TCR T1. For the other CDR loops, the conformations with the low energies were identified by simulated annealing with the rest of the structure fixed. From these, the final loop orientations were selected by using data from the mutation experiments. The resulting structure was refined with 500 steps of steepest descent energy minimization using the CHARMM (version 25) program (38Brooks B.R. Bruccoleri R.E. Olasfon B.D. States D.J. Swaminathan S. Karplus M. J. Comp. Chem. 1983; 4: 187-217Crossref Scopus (14019) Google Scholar) with the all-atom PARAM 22 parameter set (39MacKerell Jr., A.D. Bashford D. Bellott M. Dunbrack Jr., R.L. Karplus M. J. Phys. Chem. 1998; 102: 3586-3616Crossref PubMed Scopus (11819) Google Scholar). No significant violation of spatial restraints was found for β-Tyr48, β-Tyr50, and K(ABA)after optimization, indicating that the imposed distance restraint does not imply a distortion of the structure. Details concerning the modeling will be presented separately. 2O. Michielin and M. Karplus, manuscript in preparation. To obtain information on PbCS(ABA) contacts with T1 TCR, several PbCS(ABA) variants were assessed by TCR photoaffinity labeling with soluble Kd-125IASA-YIPSAEK(ABA)I complexes (Fig.1 A). The replacement of PbCS Pro255 by Ala, Asp, or Ser reduced T1 TCR labeling by 10-, 100-, and 17-fold, respectively, whereas replacement by Leu increased it 2-fold, suggesting that voluminous aliphatic residues in this position stabilize, and polar ones destabilize, T1 TCR-ligand binding. Alanine substitution of PbCS Ser256 impaired T1 TCR photoaffinity labeling by 95%. Substitution of PbCS Glu258with alanine or glutamine obliterated detectable T1 TCR labeling, indicating that Glu258 forms a polar contact with T1 TCR. Shortening of PbCS Lys259 by one methylene group (YIPSAEOrn(ABA)I) also abolished T1 TCR labeling, indicating that the full spacer length was required. To define the interaction of ABA with T1 TCR, PbCS(ABA) variants with modified ABA were examined. These nonphotoreactive compounds were assessed in a cytolytic assay as derivatives of SYIPSAEK(ABA)I (Fig.1 B). Cloned T1 CTL killed target cells sensitized with SYIPSAEK(benzoic acid)I approximately 100-fold less efficiently than those sensitized with SYIPSAEK(ABA)I. Replacement of the phenylazide by a methyl group (SYIPSAEK(Ac)I) obliterated detectable antigen recognition, while introduction of an iodine and hydroxy substituent in ABA (YIPSAEK(IASA)I) reduced the efficiency of antigen recognition by 8-fold. These results indicate that the phenylazide of the ABA moiety was essential for antigen recognition and that changes of substituents predictably affected the efficiency of recognition. To identify Kd-TCR contacts, we prepared soluble Kd and 12 Kd mutants containing single alanine substitutions on the surface of the α1 or α2 helices (Fig.2). After photo-cross-linking with radiolabeled 125IASA-YIPSAEK(ABA)I, TCR-ligand binding was assessed by T1 TCR photoaffinity labeling, as described above. Six of the Kd mutations impaired T1 TCR-ligand binding by ≥50%. Two were on Kdα1 (E62A and Q72A), and four were on Kdα2 (Q149A, D152A, Y155A, and E166A). Some Kd mutations increased T1 TCR photoaffinity labeling by up to 20% (S69A, R79A, and E163A). To define ligand contact residues, soluble T1 TCR and 31 mutants were prepared and tested by T1 TCR photoaffinity labeling with soluble Kd-125IASA-YIPSAEK(ABA)I (Fig.3). Seven of these mutations reduced TCR photoaffinity labeling by ≥90%. Three of these were in CDR3 loops, three others were in CDR1, and one was in CDR2α. In addition, six other mutations reduced T1 TCR photoaffinity labeling by ≥50%. Four of these were in CDR2α, one was in CDR2β, and one was in CDR3α. Several mutations increased TCR photoaffinity labeling by up to 60% (e.g. αN27A, βY50F, and βT55A). Mutants labeled with an asterisk were T1 TCR Fv single chain constructs (29Grégoire C. Lin S.Y. Mazza G. Rebai N. Luescher I.F. Malissen B. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 7184-7189Crossref PubMed Scopus (18) Google Scholar). To localize the photoaffinity-labeled site(s), T1 TCR was photoaffinity-labeled with SYIPSAEK(125IASA)I, a derivative that was efficiently recognized by T1 CTL (Fig. 1 B) and, being monovalent, precluded cross-linking with Kd. Since T1 TCR was photoaffinity-labeled exclusively at the β-chain (15Luescher I.F. Anjuère F. Peitsch M.C. Jongeneel V. Cerottini J.-C. Romero P. Immunity. 1995; 3: 51-63Abstract Full Text PDF PubMed Scopus (50) Google Scholar), the photoaffinity-labeled TCR was directly subjected to peptide mapping. After extensive digestion with trypsin, the resulting digest fragments were separated by reverse phase HPLC. The major labeled digest fragment eluted from the C-18 column after 30 min and according to SDS-PAGE was homogenous and had an apparent M r of approximately 3000 (Fig. 4,lane 2). When protease V8 was used instead of trypsin, a major labeled material eluted from the column after 32–33 min, which migrated on SDS-PAGE with an apparentM r of approximately 8300 (Fig. 4,lane 1). The size of this fragment was bigger than any theoretical V8 fragment of the variable domain of the T1 TCR β-chain (Fig. 5 B), suggesting that protease V8, even after extensive digestion, omitted a cleavage site. Since this enzyme primarily cleaves C-terminal to Glu and hence may fail to cleave after Asp (31Keesey J. Biochemicals for Protein Research. Boehringer Mannheim, Indianapolis, IN1987: 87-108Google Scholar), the labeled V8 digest product was digested with protease Asp-N. Essentially the same HPLC profile was observed; however, on SDS-PAGE this material migrated with an apparent M r of approximately 2000 (Fig. 4, lane 3), indicating that V8 failed to cleave at an aspartic acid. This residue probably was β-Asp38, because the big reduction in size (approximately 6300 Da) may correspond to the size of the segment 2–37, which contains the glycosylation site β-Asn24 (Fig. 5 B and Ref. 34Bentley G.A. Boulot G. Karjalainen K.
Referência(s)