The three-dimensional structure of a T-cell antigen receptor V alpha V beta heterodimer reveals a novel arrangement of the V beta domain.

The crystal structure of a mouse T-cell antigen receptor (TCR) Fv fragment complexed to the Fab fragment of a specific anti-clonotypic antibody has been determined to 2.6 A resolution. The polypeptide backbone of the TCR V alpha domain is very similar to those of other crystallographically determined V alphas, whereas the V beta structure is so far unique among TCR V beta domains in that it displays a switch of the c" strand from the inner to the outer beta-sheet. The beta chain variable region of this TCR antigen-binding site is characterized by a rather elongated third complementarity-determining region (CDR3beta) that packs tightly against the CDR3 loop of the alpha chain, without leaving any intervening hydrophobic pocket. Thus, the conformation of the CDR loops with the highest potential diversity distinguishes the structure of this TCR antigen-binding site from those for which crystallographic data are available. On the basis of all these results, we infer that a significant conformational change of the CDR3beta loop found in our TCR is required for binding to its cognate peptide-MHC ligand.     

Secretion of T cell receptor fragments from recombinant Escherichia coli cells.

This study describes the secretion and purification of T cell receptor (TCR) V alpha, V beta domains and single chain V alpha-V beta fragments (scTCRs) from recombinant Escherichia coli cells. The TCR V alpha and V beta genes are derived from a T cell hybridoma that is associated with disease pathogenesis in murine experimental allergic encephalomyelitis (EAE). Circular dichroism (c.d.) analyses of the single domains and the scTCR indicate that they are folded into beta-pleated sheet structures similar to those of immunoglobulin variable domains. The secreted TCR fragments can be purified in milligram quantities, and could therefore be used in high-resolution structural studies, in immunization to generate anti-clonotypic antibodies or in vaccination.     

Framework residue 71 is a major determinant of the position and conformation of the second hypervariable region in the VH domains of immunoglobulins

Analysis of the immunoglobulins of known structure reveals systematic differences in the position and main-chain conformation of the second hypervariable region of the VH domain (H2). We show that the major determinant of the position of H2 is the size of the residue at site 71, a site that is in the conserved framework of the VH domain. It is likely that for about two thirds of the known VH sequences the size of the residue at this site is also a major determinant of the conformation of H2. This effect can override the predisposition of the sequence, as in the case of the H2 loop of J539, which is an exception to the rules relating sequence and conformation of short hairpin loops. Understanding the relationship between the residue at position 71 and the position and conformation of H2 has applications to the prediction and engineering of antigen-binding sites of immunoglobulins.     

Novel arrangement of immunoglobulin variable domains: X-ray crystallographic analysis of the lambda-chain dimer Bence-Jones protein Loc

We have characterized and crystallized a human lambda I light-chain dimer, Bence-Jones protein Loc, which has variable (V) region antigenic determinants characteristic for the lambda I subgroup and constant (C) region determinants of the C lambda I gene Mcg. The crystal structure was determined to 3-A resolution; the R factor is 0.27. The angle formed by the twofold axes of the V and C domains, the "elbow bend", is 97 degrees, the smallest found so far for an antibody fragment. The antigen-binding site formed by the two V domains of the Loc light chain differs significantly from those of other immunoglobulin molecules (light-chain dimers and Fab fragments) for which X-ray crystallographic data are available. Whereas, in other antibody fragments, the V domains are related by a local twofold axis, a local twofold screw axis with a translational component of 3.5 A relates the V domains in protein Loc. In contrast to the classic antigen binding "pocket" formed by V domain interactions in the previously characterized antibody structures, the V region associations in protein Loc result in a central protrusion in the binding site, with grooves on two sides of the protrusion. The structure of protein Loc indicates that immunoglobulins are physically capable of forming a more diverse spectrum of antigen-binding sites than has been heretofore apparent. Moreover, the unusual protruding nature of the binding site may be analogous to structures required for some anti-idiotypic antibodies. Further, the complementarity-determining residues form parts of two independent grooves.(ABSTRACT TRUNCATED AT 250 WORDS)     

Identification of the principal binding site for RGD-containing ligands in the alpha(V)beta(3) integrin: a photoaffinity cross-linking study

By superimposing data obtained by photo-cross-linking RGD-containing ligands to the human alpha(V)beta(3) integrin onto the crystal structure of the ectopic domain of this receptor (Xiong et al. (2001) Science 294, 339-345), we have identified the binding site for the RGD triad within this integrin. We synthesized three novel analogues of the 49-amino acid disintegrin, echistatin: [Bpa(21),Leu(28)]-, [Bpa(23),Leu(28)]-, and [Bpa(28)]echistatin. Each contains a photoreactive p-benzoyl-phenylalanyl (Bpa) residue in close proximity to the RGD motif which spans positions 24-26; together, the photoreactive positions flank the RGD motif. The analogues bind with high affinity to the purified recombinant alpha(V)beta(3) integrin, but very poorly to the closely related human alpha(IIb)beta(3) platelet integrin. While echistatin analogues containing Bpa in either position 23 or 28 cross-link specifically and almost exclusively to the beta(3) subunit of alpha(V)beta(3), [Bpa(21),Leu(28)]echistatin cross-links to both the alpha(V) and the beta(3) subunits, with cross-linking to the former favored. [Bpa(23),Leu(28)]echistatin cross-links 10-30 times more effectively than the other two analogues. We identified beta(3)[109-118] as the domain that encompasses the contact site for [Bpa(28)]echistatin. This domain is included in beta(3)[99-118] (Bitan et al. (2000) Biochemistry 39, 11014-11023), a previously identified contact domain for a cyclic RGD-containing heptapeptide with a benzophenone moiety in a position that is similar to the placement of the benzophenone in [Bpa(28)]echistatin relative to the RGD triad. Recently, we identified beta(3)[209-220] as the contact site for an echistatin analogue with a photoreactive group in position 45, near the C-terminus of echistatin (Scheibler et al. (2001) Biochemistry 40, 15117-14126). Taken together, these results support the hypothesis that the very high binding affinity of echistatin to alpha(V)beta(3) results from two distinct epitopes in the ligand, a site including the RGD triad and an auxiliary epitope at the C-terminus of echistatin. Combining our results from photoaffinity cross-linking studies with data now available from the recently published crystal structure of the ectopic domain of alpha(V)beta(3), we characterize the binding site for the RGD motif in this receptor.     

Structure of the alpha and beta heavy chains of the outer arm dynein from Chlamydomonas flagella. Masses of chains and sites of ultraviolet-induced vanadate-dependent cleavage

We report here on the UV-induced vanadate-dependent cleavage of the alpha and beta heavy chains of the outer arm dynein from Chlamydomonas flagella. Both polypeptides are cleaved at a single site (termed the V1 site) by UV irradiation in the presence of Mg2+, ATP, and vanadate. The alpha chain yields fragments of Mr 290,000 and 190,000. Fragments of Mr 255,000 and 185,000 are obtained from the beta chain. Ultraviolet irradiation of the alpha and beta chains in the presence of vanadate and Mn2+ (but no nucleotide) induces cleavage of both molecules at sites (termed the V2 sites) distinct from the V1 sites. The single V2 site within the beta chain is located 75,000 daltons from the site of V1 cleavage within the Mr 255,000 V1 fragment. The alpha chain contains three distinct sites of V2 cleavage; all are located within the Mr 290,000 V1 fragment, 60,000, 90,000, and 100,000 daltons from the site of V1 cleavage. From these studies, we estimate the masses of the alpha and beta heavy chains to be 480,000 and 440,000 daltons, respectively.     

A alpha and A beta class II I-A determinants of antigen-specific T-helper factor and its antigen-nonbinding chain

Antigen-specific T-helper factor (ThF) of CBA (H-2k) origin in the picryl (TNP) contact sensitivity system (Mr 60-70 kDa) was reduced with dithiothreitol under mild conditions. Affinity chromatography on antigen yielded an antigen-binding chain (Mr 20-30 kDa) and an antigen-nonbinding chain (Mr 40-50 kDa). Both chains were glycoproteins and were bound by lentil lectin. Affinity chromatography on anti-I-A monoclonal antibodies showed that I-A determinants occurred on the complete molecule and on the antigen-nonbinding, but not on the antigen-binding, chain. In contrast, five different monoclonal antibodies to I-E alpha failed to absorb ThF. Moreover, the complete molecule and the I-A+ antigen-nonbinding chains had determinants of the alpha and beta chains of I-A and conformational determinants which are based on both chains. Sequential absorption and elution showed that A alpha and A beta determinants occurred on the same molecular complex. These data suggest a minimal model of ThF as a two-chain disulfide-bonded structure with an antigen-binding chain and a separate I-A+ antigen-nonbinding chain which behaves as a single unit in phosphate-buffered saline and has elements of both A alpha and A beta.     

The association of heavy and light chain variable domains in antibodies: implications for antigen specificity

The antigen-binding site of immunoglobulins is formed by six regions, three from the light and three from the heavy chain variable domains, which, on association of the two chains, form the conventional antigen-binding site of the antibody. The mode of interaction between the heavy and light chain variable domains affects the relative position of the antigen-binding loops and therefore has an effect on the overall conformation of the binding site. In this article, we analyze the structure of the interface between the heavy and light chain variable domains and show that there are essentially two different modes for their interaction that can be identified by the presence of key amino acids in specific positions of the antibody sequences. We also show that the different packing modes are related to the type of recognized antigen.     

Structural,antigenic and evolutionary analyses of immunoglobulins and T cell receptors

We have had the pleasure of collaborating with Allen Edmundson for the past 15 years on the structure, binding properties and evolution of immunoglobulins and T cell receptors. Among the most significant contributions of our joint efforts were: (1) the predictive use of structural features of immunoglobulin domains to model the three-dimensional structures of the immunoglobulin domains of human T-cell receptor alpha and beta chains as well as shark light chains and V(H) domains; (2) the finding that normal humans and other vertebrates express autoantibodies against combining site epitopes of their own T cell receptors; (3) the mapping of the peptide autoepitopes recognized in health, autoimmunity and retroviral infection; and (4) the determination that epitope recognition promiscuity is a characteristic property of the combining sites of IgM immunoglobulins ranging from those of sharks to those of humans. We briefly review the salient findings and status of these studies and indicate the future directions that we will pursue in their continuation.     

Three-dimensional structure of the tryptophan synthase alpha 2 beta 2 multienzyme complex from Salmonella typhimurium

The three-dimensional structure of the alpha 2 beta 2 complex of tryptophan synthase from Salmonella typhimurium has been determined by x-ray crystallography at 2.5 A resolution. The four polypeptide chains are arranged nearly linearly in an alpha beta beta alpha order forming a complex 150 A long. The overall polypeptide fold of the smaller alpha subunit, which cleaves indole glycerol phosphate, is that of an 8-fold alpha/beta barrel. The alpha subunit active site has been located by difference Fourier analysis of the binding of indole propanol phosphate, a competitive inhibitor of the alpha subunit and a close structural analog of the natural substrate. The larger pyridoxal phosphate-dependent beta subunit contains two domains of nearly equal size, folded into similar helix/sheet/helix structures. The binding site for the coenzyme pyridoxal phosphate lies deep within the interface between the two beta subunit domains. The active sites of neighboring alpha and beta subunits are separated by a distance of about 25 A. A tunnel with a diameter matching that of the intermediate substrate indole connects these active sites. The tunnel is believed to facilitate the diffusion of indole from its point of production in the alpha subunit active site to the site of tryptophan synthesis in the beta active site and thereby prevent its escape to the solvent during catalysis.     

Structure of the alpha and beta heavy chains of the outer arm dynein from Chlamydomonas flagella. Nucleotide binding sites

The photoaffinity analogs 2-azidoadenosine 5'-tri(di)-phosphate (2-N3AT(D)P) and 8-azidoadenosine 5'-triphosphate (8-N3ATP) have been used to probe the substructural organization of the nucleotide binding pockets within the alpha and beta heavy chains of the outer arm dynein from Chlamydomonas flagella. Both 2-N3ATP and 8-N3ATP are competitive inhibitors of dynein ATP hydrolysis, and both analogs are themselves hydrolyzed by the alpha-beta dimer. Following vanadate-dependent photolysis at the V1 site (by UV irradiation in the presence of Mg2+, ATP, and vanadate), both probes exclusively labeled the larger fragment from the alpha chain. In contrast, within the beta chain the predominant insertion sites for the two analogs were located on opposite sides of the V1 site. Therefore, the hydrolytic pockets of these two molecules have different substructures. Vanadate-dependent photolysis of the alpha and beta chains at the V2 sites (by UV irradiation in the presence of vanadate and Mn2+) profoundly affected the predominant modification sites; for example, following photolysis at the V2a site neither fragment of the alpha chain was photolabeled by 2-N3ATP or 8-N3ATP. Based on the photolabeling patterns obtained, the single V2 site within the beta chain is predicted to be analogous to the V2b site within the alpha chain. The results support the hypothesis that the V2 sites occur within the ATP binding pockets, and indicate that these functional domains are composed of portions of the heavy chains which are linearly separated by up to at least 100,000 daltons. Thus, the central region of each dynein heavy chain must be extensively folded so as to bring the widely separated photocleavage and photolabeling sites together within a single catalytic unit.     

Refined structure of the human histocompatibility antigen HLA-A2 at 2.6 A resolution

The three-dimensional structure of the human histocompatibility antigen HLA-A2 was determined at 3.5 A resolution by a combination of isomorphous replacement and iterative real-space averaging of two crystal forms. The monoclinic crystal form has now been refined by least-squares methods to an R-factor of 0.169 for data from 6 to 2.6 A resolution. A superposition of the structurally similar domains found in the heterodimer, alpha 1 onto alpha 2 and alpha 3 onto beta 2m, as well as the latter pair onto the ancestrally related immunoglobulin constant domain, reveals that differences are mainly in the turn regions. Structural features of the alpha 1 and alpha 2 domains, such as conserved salt-bridges that contribute to stability, specific loops that form contacts with other domains, and the antigen-binding groove formed from two adjacent helical regions on top of an eight-stranded beta-sheet, are analyzed. The interfaces between the domains, especially those between beta 2m and the HLA heavy chain presumably involved in beta 2m exchange and heterodimer assembly, are described in detail. A detailed examination of the binding groove confirms that the solvent-accessible amino acid side-chains that are most polymorphic in mouse and human alleles fill up the central and widest portion of the binding groove, while conserved side-chains are clustered at the narrower ends of the groove. Six pockets or sub-sites in the antigen-binding groove, of diverse shape and composition, appear suited for binding side-chains from antigenic peptides. Three pockets contain predominantly non-polar atoms; but others, especially those at the extreme ends of the groove, have clusters of polar atoms in close proximity to the "extra" electron density in the binding site. A possible role for beta 2m in stabilizing permissible peptide complexes during folding and assembly is presented.     

Ligand-integrin alpha v beta 3 interaction determined by photoaffinity cross-linking: a challenge to the prevailing model

Integrin alpha(V)beta(3) plays a crucial role in angiogenesis, apoptosis, and bone remodeling, mainly by interacting with matrix proteins through recognition of an Arg-Gly-Asp (RGD) motif. Recently, a small cyclic RGD-containing alpha(V)beta(3)-ligand possessing a C-terminal photoreactive group was photo-cross-linked within beta(3)[99-118], in the N-terminus of the beta(3) chain [Bitan G et al. (1999) Biochemistry 38, 3414-3420]. In this paper, a photoreactive group at the N-terminus of the RGD-ligand is shown to interact within beta(3)[167-171], approximately 60 residues C-terminal to the previously identified domain. On the basis of these findings, a model of the putative I-like domain of the beta(3) subunit, homologous to alpha(M)-, alpha(L)-, and alpha(2)-I-domains, reveals that the beta(3)[99-118] and beta(3)[167-171] contact sites are close to each other and are on the opposite side relative to the metal ion-dependent adhesion site (MIDAS) motif. These observations contradict the prevailing model that proposes proximity between metal- and RGD-binding sites on the I-like domain. Our data suggest that either the I-like domain structure predicted for beta(3) is incorrect, or there is no spatial proximity between the RGD-binding site and the MIDAS motif in the I-like domain. Our results indicate that the current models for ligand-receptor interaction should be revisited.     

Photosensitized cleavage of dynein heavy chains. Cleavage at the V2 site by irradiation at 365 NM in the presence of oligovanadate

Irradiation of the outer-arm dynein ATPase from sea urchin sperm flagella at 365 nm in the presence of 50-200 microM vanadate (Vi) and 1 mM manganese acetate, in the absence of ATP, cleaves the alpha and beta heavy chains at a specific site, termed the V2 site, to form discrete peptides of Mr approximately 260,000 and 170,000 from the alpha chain and of Mr approximately 255,000 and 175,000 from the beta chain, with a yield of 80%. This cleavage at the V2 site is not correlated with any direct effect on the dynein ATPase activity. In the presence of 100 microM Vi, the half-times for cleavage of the alpha and beta chains are about 12 and 50 min, respectively. The rate of heavy chain cleavage shows a sigmoidal dependence upon Vi concentration, with half-maximal rate occurring at 58 +/- 7 microM, consistent with the chromophore responsible for cleavage being tri-vanadate. Addition of 10 microM ATP or ADP, or of 100 microM CTP or UTP, to the irradiation medium inhibits cleavage at the V2 site, and results in a slow cleavage occurring at the V1 site described previously. The peptides produced by sequential cleavage at the V2 and then the V1 sites indicate that the sites are separated by about 100,000 Da along the length of each heavy chain. Photoaffinity labeling with [alpha-32P] 8-azidoadenosine 5'-triphosphate (8-N3ATP) gives specific incorporation of 32P into both the Mr 255,000 and 175,000 peptides of the beta chain but into only the Mr 260,000 peptide of the alpha chain. These results suggest that V2 cleavage occurs on a loop of the heavy chain that forms part of the ATP-binding site, close to the locus of 8-N3ATP attachment.     

Molecular basis of ligand recognition by integrin alpha 5beta 1. I. Specificity of ligand binding is determined by amino acid sequences in the second and third NH2-terminal repeats of the alpha subunit

The NH(2)-terminal portion (putative ligand-binding domain) of alpha subunits contains 7 homologous repeats, the last 3 or 4 of which possess divalent cation binding sequences. These repeats are predicted to form a seven-bladed beta-propeller structure. To map ligand recognition sites on the alpha(5) subunit we have taken the approach of constructing and expressing alpha(V)/alpha(5) chimeras. Although the NH(2)-terminal repeats of alpha(5) and alpha(V) are >50% identical at the amino acid level, alpha(5)beta(1) and alpha(V)beta(1) show marked differences in their ligand binding specificities. Thus: (i) although both integrins recognize the Arg-Gly-Asp (RGD) sequence in fibronectin, the interaction of alpha(5)beta(1) but not of alpha(V)beta(1) with fibronectin is strongly dependent on the "synergy" sequence Pro-His-Ser-Arg-Asn; (ii) alpha(5)beta(1) binds preferentially to RGD peptides in which RGD is followed by Gly-Trp (GW) whereas alpha(V)beta(1) has a broader specificity; (iii) only alpha(5)beta(1) recognizes peptides containing the sequence Arg-Arg-Glu-Thr-Ala-Trp-Ala (RRETAWA). Therefore, amino acid residues involved in ligand recognition by alpha(5)beta(1) can potentially be identified in gain-of-function experiments by their ability to switch the ligand binding properties of alpha(V)beta(1) to those of alpha(5)beta(1). By introducing appropriate restriction enzyme sites, or using site-directed mutagenesis, parts of the NH(2)-terminal repeats of alpha(V) were replaced with the corresponding regions of the alpha(5) subunit. Chimeric subunits were expressed on the surface of Chinese hamster ovary-B2 cells (which lack endogenous alpha(5)) as heterodimers with hamster beta(1). Stable cell lines were generated and tested for their ability to attach to alpha(5)beta(1)-selective ligands. Our results demonstrate that: (a) the first three NH(2)-terminal repeats contain the amino acid sequences that determine ligand binding specificity and the same repeats include the epitopes of function blocking anti-alpha subunit mAbs; (b) the divalent cation-binding sites (in repeats 4-7) do not confer alpha(5)beta(1)- or alpha(V)beta(1)-specific ligand recognition; (c) amino acid residues Ala(107)-Tyr(226) of alpha(5) (corresponding approximately to repeats 2 and 3) are sufficient to change all the ligand binding properties of alpha(V)beta(1) to those of alpha(5)beta(1); (d) swapping a small part of a predicted loop region of alpha(V) with the corresponding region of alpha(5) (Asp(154)-Ala(159)) is sufficient to confer selectivity for RGDGW and the ability to recognize RRETAWA.     

Common features of the conformations of antigen-binding loops in immunoglobulins and application to modeling loop conformations.

Using database screening techniques we have examined the relationship between antigen-binding loops in immunoglobulins, and regions of similar conformation in other protein families. The conformations of most antigen-binding loops are not unique to immunoglobulins. But in many cases, the geometrical relationship between the loop and the peptides flanking it differs between the immunoglobulins and other structures with the same loop. We assess model building by data base screening, compared with that based on canonical structures.     

A comparison of the amino acid sequences of the extracellular domains of the immunoglobulin superfamily. Possible correlations between conservancy and conformation

Amino acid sequences of immunoglobulins, histocompatibility antigens, Thy-1 antigens, polyimmunoglobulin receptor and T-lymphocyte receptor have been compared and conserved residues correlated with features on the atomic models of immunoglobulin fragments. The peptide chains are apparently folded into globular domains closely related in three-dimensional structure to immunoglobulin V or C domains. Exceptions are histocompatibility class I alpha 2 and class II beta 1 domains, distantly related, and class I and II alpha 1 domains, no apparent relationship to immunoglobulin domains.     

The evolution of vertebrate antigen receptors: a phylogenetic approach

Classical T cells, those with alpha beta T-cell receptors (TCRs), are an important component of the dominant paradigm for self-nonself immune recognition in vertebrates. alpha beta T cells recognize foreign peptide antigens when they are bound to MHC molecules on the surfaces of antigen-presenting cells. gamma delta T cells bear a similar receptor, and it is often assumed that these T cells also require specialized antigen-presenting molecules for immune recognition, which we term "indirect antigen recognition." B-cell receptors, or immunoglobulins, bind directly to antigens without the help of a specialized antigen-presenting molecule. Phylogenetically, it has been assumed that T-cell receptors and the genes that encode them are a monophyletic group, and that "indirect" antigen recognition evolved before the split into two types of TCR. Recently, however, it has been proposed that gamma delta-TCRs bind directly to antigens, as do immunoglobulins (Ig's). This calls into question the null hypothesis that indirect antigen recognition is a common characteristic of TCRs and, by extension, the hypothesis that all TCR gene sequences form a monophyletic group. To determine whether alternative explanations for antigen recognition and other historical relationships among TCR genes might be possible, we performed phylogenetic analyses on amino acid sequences of the constant and variable regions which encode the basic subunits of TCR and Ig molecules. We used both maximum-parsimony and genetic distance-based methods and could find no strong support for the hypothesis of TCR monophyly. Analyses of the constant region suggest that TCR gamma or delta sequences are the most ancient, implying that the ancestral immune cell was like a modern gamma delta T cell. From this gamma delta-like ancestor arose alpha beta T cells and B cells, implying that indirect antigen recognition is indeed a derived property of alpha beta-TCRs. Analyses of the variable regions are complicated by strong selection on antigen-binding sequences, but imply that direct antigen binding is the ancestral condition.