The three-dimensional structure of a phosphorylcholine-binding mouse immunoglobulin Fab and the nature of the antigen binding site

Monoclonal phosphorylcholine antibody binds to beta-lipoprotein from different animal species

Previously published observations have demonstrated that murine antibody with specificity for phosphorylcholine (PC) binds to a variety of PC-containing polysaccharides and to human beta-lipoprotein. In this investigation it was found that monoclonal anti-PC antibody binds not only to human beta-lipoprotein, but also to beta-lipoprotein from the serum of all vertebrate species examined. beta-Lipoprotein from mouse reacted with autologous antibody; therefore, this antibody must be an autoantibody. This raises the question whether PC antibodies may exert unwanted effects after having been passively transferred or induced by PC-containing antigens, e.g., C-polysaccharide from pneumococci.

Subgroups of variable region genes of beta chains of T-cell receptors for antigen

The beta-chain variable regions of 6 human, 21 mouse, and 1 rabbit T-cell receptors for antigen, including two pairs of duplicates, have been classified into two subgroups. Subgroup I has an invariant Phe at residue 65 and a salt bridge can be formed between the Asp at position 86 and the 11 Arg, 2 Lys, and 1 His residues at position 64. In one instance Tyr replaces Asp at position 86 so that a salt bridge cannot be formed with the Lys at position 64. Subgroup II has an invariant Tyr at residue 65; 8 chains have Gly and 3 have Ala at position 63. These 3 have Asp at position 64. Those with Gly-63 cannot form a salt bridge in this region of the beta chain. The 3 with Asp-64 have Arg at position 86, can form a reverse salt bridge, and could constitute a third subgroup. The relation of these findings to functional parameters in beta chains of the T-cell receptors for antigen deserves further study. The use of variability plots on the assumption that the beta chains are a single population may be misleading for the variable region. Most of the variability in the expressed beta chains would be created by variable-diversity-joining assortment, junctional diversity, and the possible insertion of N sequences.

Replacing the complementarity-determining regions in a human antibody with those from a mouse

The variable domains of an antibody consist of a beta-sheet framework with hypervariable regions (or complementarity-determining regions--CDRs) which fashion the antigen-binding site. Here we attempted to determine whether the antigen-binding site could be transplanted from one framework to another by grafting the CDRs. We substituted the CDRs from the heavy-chain variable region of mouse antibody B1-8, which binds the hapten NP-cap (4-hydroxy-3-nitrophenacetyl caproic acid; KNP-cap = 1.2 microM), for the corresponding CDRs of a human myeloma protein. We report that in combination with the B1-8 mouse light chain, the new antibody has acquired the hapten affinity of the B1-8 antibody (KNP-cap = 1.9 microM). Such 'CDR replacement' may offer a means of constructing human monoclonal antibodies from the corresponding mouse monoclonal antibodies.

Immunoglobulin with complementary paratope and idiotope

A hybridoma antibody (11E7-1) was isolated from a myeloma fusion with nu/nu BALB/c immunized against the T15 idiotype. This IgM antibody exhibited a dual specificity, binding both to PC and to anti-PC antibodies from two idiotype families. Binding to PC and anti-PC antibodies are completely inhibited by PC analogs. Furthermore, the hybridoma antibody binds to itself. Self-binding is also inhibited by PC analogs. From these data, we suggest that 11E7-1 hybridoma antibody has a PC-specific paratope site, and at same time expresses the internal PC antigen idiotope. The term autobody is proposed to signify its self-binding and potential role in autoimmunity. Autobodies may have a unique role in the network of immune system. Furthermore, it may be a model for designing idiotype vaccines.

Secondary, tertiary, and quaternary structure of T-cell-specific immunoglobulin-like polypeptide chains

To explore the possibility that the difference in antigen recognition between B and T cells derives from a structural difference in their respective antigen-specific receptors (immunoglobulins on B cells and immunoglobulin-like molecules on T cells), we compared the extracellular segments of the T-cell receptor alpha, beta, and gamma polypeptide chains and the N-terminal segment of the T-cell T8 (Lyt-2) antigen chain with the corresponding regions of immunoglobulins whose three-dimensional structures are known. The results indicate that the four T-cell polypeptide chains are organized into immunoglobulin-like domains consisting of multistranded antiparallel beta-sheet bilayers. Invariant amino acid side chains that are conserved in diverse immunoglobulins, including those that mediate domain-domain interactions and form a constant scaffold for antibody binding sites, are also conserved in the chains encoded by the T-cell receptor genes and in the N-terminal domain of T8 (Lyt-2). It appears that the binding sites of the antigen-specific T-cell alpha beta-chain receptors and of antibodies are very similar in their overall dimensions and geometry: a T-cell alpha beta receptor molecule probably has an antigen-specific binding site that is fundamentally no different than the conventional binding site of an antibody.

Modelling of the combining sites of three anti-lysozyme monoclonal antibodies and of the complex between one of the antibodies and its epitope

Models of the antigen combining sites of three monoclonal antibodies, which recognise different but overlapping epitopes within the 'loop' region of hen egg lysozyme (HEL), have been generated from the cDNA sequences of their Fv regions (the VL and VH domains) and the known crystal structures of immunoglobulin fragments. The alpha-carbon backbone of the structurally conserved framework region has been derived from the IgG myeloma protein NEW, and models for the hypervariable loop regions have been selected on the basis of length and maximum sequence homology. The model structures have been refined by energy minimisation. Both the size and chemical nature of the predicted combining site models correlate broadly with the epitope boundaries previously determined by affinity studies. A model of the complex formed between one antibody and the corresponding lysozyme epitope is described, and contact residues are identified for subsequent testing by oligonucleotide-directed site-specific mutagenesis.

Solvation energy in protein folding and binding

We have developed a method for calculating the stability in water of protein structures, starting from their atomic coordinates. The contribution of each protein atom to the solvation free energy is estimated as the product of the accessibility of the atom to solvent and its atomic solvation parameter. Applications of the method include estimates of the relative stability of different protein conformations, estimates of the free energy of binding of ligands to proteins and atomic-level descriptions of hydrophobicity and amphiphilicity.

Recognition and interactions controlling the assemblies of beta barrel domains

We present a qualitative computer graphics approach to the characterization of forces important to the assembly of beta domains that should have general utility for examining protein interactions and assembly. In our approach, the nature of the molecular surface buried by the domain contacts, the specificity of the residue-to-residue interactions, and the identity of electrostatic, hydrophobic, and hydrophilic interactions are elucidated. These techniques are applied to the beta barrel domains of Cu, Zn superoxide dismutase (SOD), immunoglobulin Fab, and tomato bushy stunt virus coat protein (TBSV), a plant viral capsid protein. By looking at a set of proteins having different numbers of interacting beta domains, we have been able to see some of the variety and also some of the patterns common to these assembled domains. Strong beta domain interactions (identified by their biochemical integrity) are apparently due to chemical, electrostatic, and shape complementarity of the molecular surfaces buried from interaction with solvent molecules. Although the amount of hydrophobic buried surface area appears to correlate with the strength of the interaction, electrostatic forces appear to be important in both stabilizing and destabilizing specific contacts. In TBSV, analysis of electrostatic interactions may help explain mechanisms of subunit accommodation to different environments, particle expansion, and pathways of assembly. The possible molecular basis for observed differences in the stability and flexibility of the domain complexes is discussed.

Hydrophobicity and amphiphilicity in protein structure

The importance of the hydrophobic interaction in stabilizing native protein structure has long been appreciated. However, more than other component forces, this one has resisted quantitative description. We present two approximate methods of assessing the hydrophobic component to the free energy of protein folding. Both are expressed in terms of what can be called hydrophobic moments of the protein. The first method is intended to yield an approximate value for the hydrophobic energy. This energy is calculated from a set of atomic coordinates in terms of the hydrophobicity (or 0th hydrophobic moment) of each amino acid residue and its accessibility or lack of it to aqueous solvent. The second method considers the first moment of the hydrophobicity of a group of residues, the hydrophobic moment. Segments of secondary structure in folded proteins tend to have hydrophobic moments that oppose each other. For example, alpha-helices on the protein surface tend to have one hydrophobic face and one hydrophilic face, with the hydrophilic face out towards the solvent. This pattern of organization is often apparent from a computer model of the protein that shows the magnitude and direction of the hydrophobic moment of each segment of secondary structure. Examples are given for the incorrectly folded structures of Novotný et al [J Mol Biol 177:787, 1984] and for the correct structures to which they correspond.

cDNA of the immunoglobulin kappa chain of an Epstein-Barr virus-transformed human lymphoid cell line: partial sequence determination and bacterial expression

We report the isolation, nucleotide sequence determination, and bacterial expression of a partial cDNA for the immunoglobulin kappa chain from the Epstein-Barr virus-transformed human lymphoid cell line GM131. The cDNA, cloned in pBR322 by use of oligo(dG) X oligo(dC) tails, yields two Pst I fragments of 250 and 600 base pairs (bp). Various restriction enzyme fragments of the cDNA were subcloned in the vectors M13 mp10 and M13 mp11 for sequence analysis. As a result of instability of the 250-bp M13 subclones, the base sequence of the 250-bp Pst I fragment could not be determined. The 600-bp Pst I fragment contains coding sequences for part of the variable (V) region (residues 78-95) and all of the joining (J) (residues 96-108) and constant (C) regions (residues 109-212) and extends 148 bp into the 3' flanking region. Although the C- and 3'-flanking-region sequences are identical to germ-line sequences, the J-region sequence does not correspond to any of the five human germ-line J regions. The sequence is most similar to that of J4, with three base changes resulting in one silent mutation and two amino acid substitutions, at residues 103 (Lys----Tyr) and 106 (Ile---Met). The silent mutation appears to be the result of RNA splicing between the J and the C regions. The V-region sequence differs from published V-region germ-line sequences at several codons and from the more common amino acid sequences at two positions, residues 91 and 93. At these positions, histidine residues are found in place of the more common tyrosine and serine, respectively. None of the four amino acid substitutions observed for the GM131 kappa-chain are unique, suggesting that the changes, which most likely contribute to antigenic specificity, are compatible with antibody structure and function. The 600-bp Pst I fragment was subcloned in two prokaryotic expression vectors, pATH11 and pUC8. In both instances, a kappa-chain fusion protein detectable by immunoblotting was produced.

Secondary structure of the immunoglobulin J chain

J chain is a 137-residue polypeptide that is covalently linked to polymeric immunoglobulins and participates in their synthesis and transport to external secretions. To clarify these roles, the secondary structure of J chain was characterized by computer-assisted analyses of human and mouse sequences and by circular dichroism measurements of the isolated J chain. The secondary-structure profiles obtained were very similar to those of superoxide dismutase or immunoglobulin light chain variable domains, suggesting that the J chain folds into an eight-stranded antiparallel beta-barrel and should contain approximately 37% beta-sheet conformation, with the rest of the structure existing as reverse turns (random coil). The circular dichroism measurements indicated that the conformation of denatured, S-carboxymethylated or S-sulfonated J chain consists of 75% random coil and 25% beta-structure. Upon reformation of disulfide bonds the percentage of beta-structure in the air-oxidized J chain increased to 34%, a value that is in good agreement with the secondary-structure analysis. Two alternative models of J-chain structure, a two-domain model [Cann, G., Zaritsky, A. & Koshland, M.E. (1982) Proc. Natl. Acad. Sci. USA 79, 6656-6660] and a single-domain antiparallel beta-sheet bilayer model (proposed in this paper), are compared.

Diversity and structure of genes of the alpha family of mouse T-cell antigen receptor

We have analysed 19 complementary DNA clones encoding the alpha-chain of the T-cell antigen receptor derived from thymic transcripts, and find that 15 of them contain partial or complete variable (V alpha) genes. Seven of these genes cross-hybridize to over 40 germline V alpha gene segments in Southern blot analyses. Of the 19 joining (J alpha) sequences examined, 18 seem to be encoded by distinct gene segments, hence the repertoire of J alpha gene segments is much larger than those of the immunoglobulin or T-cell receptor beta-chain gene families. We suggest that the variable domains of immunoglobulins and T-cell antigen receptors are similar in structure.

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)

Structural invariants of antigen binding: comparison of immunoglobulin VL-VH and VL-VL domain dimers

Antigen-combining site arises by noncovalent association of the variable domain of the immunoglobulin heavy chain (VH) with that of the light chain (VL). To analyze the invariant features of the binding region (VL-VH domain interface), we compared the known immunoglobulin three-dimensional structures by a variety of methods. The interface forms a close-packed, twisted, prism-shaped "beta-barrel" characterized by cross-sectional dimensions 1.04 X 0.66 nm and a top-to-bottom twist angle of 212 degrees. The geometry of the interface is preserved via invariance of some 15 side chains, both inside the domains and on their surface. Buried polar residues form a conserved hydrogen-bonding network that has a similar topological connectivity in the two domain types; two hydrogen bonds contributed by invariant side chains extend across the interface and anchor the beta-sheets in their relative orientation. Invariant aromatic residues close-pack at the bottom of the binding-site beta-barrel with their ring planes oriented perpendicularly in the characteristic "herringbone" packing mode. Electrostatic computations that implicitly include solvent effects show the domains to be stabilized by large electrostatic forces. However, structures that were crystallized at lower pH have their electrostatic energies appropriately lowered, implying that full ionization of carboxyl side chains is essential for efficient electrostatic stabilization. The unusual mode of domain-domain association in the VL-VL dimer RHE correlates with its overall repulsive electrostatic energy (+54 kJ/mol), as opposed to negative (i.e., stabilizing) energy values (-263 to -543 kJ/mol) found in the domains of the other structures. The VL-VL dimer REI mimics closely the interface geometry of VL-VH dimers although its domain-domain contact area is lower by 18%.

Specificity of mouse hybridoma antibodies to DNA. III. Antigenic determinants of nucleic acids recognized by poly(dG) specific monoclonal antibodies

Immunochemical properties of monoclonal autoantibodies to single stranded DNA (ssDNA), which were derived from systemic lupus erythematosus (SLE) prone mice and were specific for poly(dG), were studied using a hapten inhibition assay with oligo- and mononucleotides. Deoxyguanosine monophosphate (dGMP) completely inhibited the ssDNA binding of poly(dG) specific monoclonal antibodies. The affinity to deoxyguanosine was approximately 70% of dGMP, and to guanosine monophosphate (GMP) and to guanosine was 38% and 24%, respectively. In contrast, other purine monomers such as deoxyadenosine monophosphate (dAMP), adenosine monophosphate (AMP) and adenosine exhibited virtually no inhibitory effect on the ssDNA binding of these monoclonal hybridoma autoantibodies. By the same competitive inhibition enzyme immunoassay using deoxyguanosine oligonucleotides of various chain lengths, the minimal antigenic size of ssDNA encompassed by the combining site of the Fab portion of anti-ssDNA monoclonal autoantibodies was determined to be the tetra- or pentanucleotide.

On the specificity of antibody/antigen interactions: phosphocholine binding to McPC603 and the correlation of three-dimensional structure and sequence data

Refined three-dimensional structures of McPC603 Fab and the complex with phosphocholine permit a detailed assessment of the residues crucial to determining the antibody specificity. Correlation with sequence data suggests that the structure of the binding site is highly conserved in immunoglobulins with phosphocholine-binding specificity. There is suggestive evidence that coupling of somatic mutations occurs to preserve antigen-binding specificity. The immune response is characterized by specificity and diversity. While each antibody appears to be specific for a single antigen, the immune response can generate up to 10(9) different specificities. In order to understand, at the molecular level, the nature of the interaction between antibody and antigen, it is necessary to have a high-resolution three-dimensional picture of the complex. Today it is possible to investigate antibody/antigen interactions directly by the crystallographic analysis of hybridoma products [5, 10]; in the past, structural studies were limited to myeloma proteins which, in some cases, could be shown to complex to certain haptens. Of the four Fab structures that have been determined by X-ray diffraction, only two have been demonstrated to bind hapten in the crystal. They are Fab NEW, which was shown to bind a vitamin K1 derivative [1] and McPC603, which binds to phosphocholine [6,9]. During the last few years, the McPC603 Fab structure has been refined at 2.7 A resolution and the complex of McPC603 Fab with phosphocholine has been refined independently at 3.1 A. In this communication, we make a comparative analysis of the sequences of a number of mouse phosphocholine-binding immunoglobulins based on the refined structure of the phosphocholine-binding site in McPC603.

Conformation of human IgG subclasses in solution. Small-angle X-ray scattering and hydrodynamic studies

The structure of six human myeloma proteins: IgG1(Bal), IgG2(Klu), IgG3(Bak), IgG3(Het), IgG4(Kov) and IgG4(Pol), was studied in solution using small-angle X-ray scattering and hydrodynamic methods. For IgG1(Bal) and IgG3(Het) the experimental data, including radius of gyration (Rg degree), radii of gyration of the cross-section (Rq1, Rq2), intrinsic viscosity [eta], sedimentation coefficient (S degree 20,w) and molecular mass, were interpreted in terms of structural models based on the Fab and Fc conformations, observed in crystal, by varying the relative positions of the Fab and Fc parts, i.e. their relative angles and distances. The values Rg degree = (6.00 +/- 0.05) nm, S degree 20,w = (6.81 +/- 0.10) S and [eta] = 0.0062 +/- 0.0005 cm3/mg obtained for IgG1(Bal) are compatible with a planar model in which the angle between the Fab arms is about 120 degrees. For IgG3(Het) the following data were obtained: Rg degree = (4.90 +/- 0.05) nm, S degree 20,w = (6.32 +/- 0.01) S and [eta] = (0.0065 +/- 0.0005) cm3/mg. The apparent contradiction between the higher molecular mass and lower Rg degree and S degree 20,w values for IgG3(Het) in comparison to IgG1(Bal) can be resolved by proposing a 'non-planar' (tetrahedral) molecular shape, in which the long hinge peptide is in a folded conformation and the two Fab and Fc parts are in a closely packed arrangement. In this model the angle between the two Fab arms is about 90 degrees, in the average position. The X-ray scattering and hydrodynamic behaviour of the IgG2 and IgG4 types of antibodies appeared to be similar to IgG1(Bal). The parameters of the two IgG3 proteins are similar while they are different to the others.

Nucleotide sequences of five anti-lysozyme monoclonal antibodies

The nucleotide sequences of the heavy and light chain immunoglobulin mRNAs derived from five hybridomas (Gloop 1-5) secreting IgGs specific for the loop region of hen egg lysozyme were determined. These monoclonal antibodies recognise three distinct but overlapping epitopes within the loop region. The sequences of two pairs of antibodies with indistinguishable fine specificities were similar in both chains whereas the sequences of antibodies of non-identical specificities were very different. It is proposed that the D-segments expressed in two of the antibodies (Gloop3 and Gloop4) are the products of one, or perhaps two, previously unidentified germ line D-genes. Gloop1 and Gloop2 use a D-segment previously identified in antibodies specific for the hapten 2-phenyloxazolone; however it is recombined in a different reading frame in the anti-lysozyme antibodies, producing a different amino acid sequence.

Three distinct epitopes within the loop region of hen egg lysozyme defined with monoclonal antibodies

Five monoclonal antibodies specific for the loop region of hen egg lysozyme were prepared by immunisation with a synthetic conjugate of a proteolytic fragment of lysozyme coupled to bovine serum albumin. Their fine specificities were investigated using a panel of variant lysozymes and peptide fragments of lysozyme in a quantitative radio-immunoassay procedure. Knowledge of the structure of hen lysozyme to high resolution and the use of computer graphics enables the localisation of the epitopes recognised by the antibodies with some precision. The antibodies were shown to define three distinct, overlapping epitopes within what was previously considered to be a single antigenic site. These results are discussed in relation to current ideas of the antigenic nature of proteins and other recent studies in which anti-protein antibodies have been elicited by immunisation with small peptides.

Three-dimensional structure of an antigen-antibody complex at 6 A resolution

Present understanding of the three-dimensional structure of antibody combining sites is based on X-ray diffraction studies of myeloma immunoglobulins. The structures of the antigen-binding fragment (Fab) complexes of two of these immunoglobulins with small ligands have also been determined. However, there is no crystallographic information concerning the interactions of an antibody with an antigen, nor do we know the precise structure of antigenic determinants on protein molecules. We now report the first structure determination of an antigen-antibody complex at 6 A resolution. The structure of the complex between hen egg-white lysozyme and the Fab of a monoclonal anti-lysozyme antibody (D1.3) shows that the combining site of antibodies is not merely a cleft delineated by the complementarity-determining regions of the variable regions of the light and heavy chains, but is a larger area extending beyond it. A correspondingly large area of the antigen makes close contacts with the antibody, in agreement with the notion of a 'topographical' rather than 'sequential' antigenic determinant. The structural basis of cross-reactivities of an antibody with heterologous antigens and the effect of a single amino acid substitution on antigenic specificity can thus be visualized in the structural model presented here.

Diversity at the variable-joining region boundary of lambda light chains has a pronounced effect on immunoglobulin ligand-binding activity

By recombining lambda light (L) chains having known variable (V) region amino acid or nucleotide sequences with a heavy (H) chain from a myeloma protein or a monoclonal antibody, we obtained reconstituted Igs that differed from each other in sequence by only one or a few amino acid substitutions at known L chain positions. Differences in affinity of the reconstituted Igs for 2,4-dinitrophenyl (DNP) ligands revealed a pronounced effect on Ig binding activity of amino acids at the V-J boundary of the lambda chains. In one instance, two reconstituted Igs that differed about 1000-fold in affinity for epsilon-DNP-aminocaproate differed in primary structure by only a single tyrosine-phenylalanine substitution at the V-J junction (position 98) of their lambda 2 chains--i.e., by only one out of approximately 660 amino acid residues (L + H chains). By focusing on affinity changes, chains with unusual V lambda-J lambda junctional residues were identified. It is possible that because of a critical effect on tertiary structure junctional amino acid variations arising from gene segment assembly (V/J and perhaps V/D/J) constitute an important source of ligand-binding diversity of antibodies.

Antisera specificities to beta-D-galactopyranoside cluster ligands

Mono-, di-, and tri-beta-D-galactopyranosides of 2-(5-hydrazinocarbonylpentanamido)-2-(hydroxymethyl)-1,3-propan edi ol [(Gal)n-TA] have been conjugated to bovine serum albumin (BSA), and used to study the binding specificities to the Gal receptors of liver parenchymal cells. In this study, rabbit antisera produced to the (Gal)n-TA-BSA were characterized by using an enzyme-linked, immunosorbent assay under conditions that allow only the antibodies directed to the carbohydrate part of the antigen to react with the solid-phase (Gal)n-TA-BSA antigens. Inhibition assays using (Gal)n-TA-BSA conjugates showed a relative specificity of the antisera for the number of Gal residues on the TA bridging group to the BSA carrier-protein, indicating that antibodies having specificities to oligosaccharide branch points can be produced. Inhibition assays with (Gal)n-TA haptens, Gal, and methyl beta-D-Gal indicated that the antibody combining-sites interact mainly with the Gal units; no inhibition was observed with the TA bridging group used as a hapten inhibitor. The spatial distances of the Gal units were apparently important for interaction with the anti-(Gal)n-TA-BSA antibody-combining-sites, as (Lac)3-TA-BSA and (Lac)3-TA exhibited relatively little inhibitory activity.

Chemical synthesis of idiotopes. Evidence that antisera to the same JH1 peptide detect multiple binding site-associated idiotopes

In an attempt to better understand the molecular basis of idiotypy, we have generated several site-specific antisera through immunization of animals with synthetic peptides corresponding to the (JH1) heavy chain joining segment 1 of the mouse heavy chain variable (VH) region. These anti-peptide sera identify several idiotypic determinants present on intact hybridoma and myeloma immunoglobulins. Expression of at least three of these idiotopes is correlated with the antigen specificity of the family of immunoglobulins bearing the determinant. Use of synthetic peptides may prove a powerful technique in the generation of molecularly defined antiidiotypic reagents.

Kappa-chain allotypes and isotypes in the rabbit: cDNA sequences of clones encoding b9 suggest an evolutionary pathway and possible role of the interdomain disulfide bond in quantitative allotype expression

The constant regions of rabbit kappa light chains are unusual because the sequences of the allotypic forms can differ more from each other than do some variable regions with which they associate. We report the nucleic acid sequence of a full-length cDNA clone of b9 allotype and show comparisons to available sequences of the rabbit kappa allotypes b4, b5, and bas-N4. Our analyses suggest that the primordial rabbit kappa gene encoded a bas-like sequence. They also reveal a surprising difference in the position of the variable region cysteine that forms the interdomain disulfide bond that is unique to most rabbit kappa chains. One b9 cDNA sequence lacks the usual cysteine-80 and instead encodes cysteine-108, which in three-dimensional models appears capable of forming the interdomain disulfide bond with cysteine-171 in the constant region. A partial sequence of a second b9 clone encodes both cysteine-80 and cysteine-108; the translation product of this clone could have a free reactive sulfhydryl group that might lead to an unstable nonfunctional Ig molecule. The fact that pre-B cells with b9 kappa chains do not differentiate and expand into productive Ig-producing cells with frequencies comparable to the other allotypes may be explained if a substantial proportion of the gene products have a free sulfhydryl group. Our sequence results suggest that in cells differentiating to produce kappa light chains of b9 allotype the number and location of the cysteines influence immunoglobulin expression.

Structural and functional aspects of domain motions in proteins

Three distinct categories of large-scale flexibility in proteins have been documented by single-crystal X-ray diffraction studies: the relatively free movement of essentially rigid globular domains that are connected by a flexible segment of polypeptide, the reorientation of essentially rigid domains among a few distinct conformations, and the concerted transition of a contiguous region of the surface of a protein from a disordered state to an ordered state. In a number of examples, well-defined functions can be assigned to these large-scale structural changes. The occurrence of such motions in proteins of known structure is reviewed, and the best-studied examples are discussed in detail to allow a critical evaluation of the methods used to identify and study these motions.

Idiotypes of anti-phosphocholine antibodies: structural correlates

Phosphocholine-specific antibodies in mice are composed of three families of antibodies, the T15, the M603 and the M511, which are constructed by combinatorial association of a VH4 H chain and one of three different L chains, VK22, VK8 and VK24, respectively. Antiidiotypic antisera can be generated which (1) recognize all members of a family, i. e. anti-T15 IdX, anti-M603 IdX and anti-M511 IdX, (2) distinguish between members of a family, e. g. A/J anti-T15, or (3) recognize determinants on only a single anti-PC antibody. Consideration of protein sequences and L-chain polymorphisms and analysis of chain recombination experiments among Id-positive and -negative antibodies revealed that IdX are determined by the L chain. A similar approach demonstrated that the Id determinant defined by the A/J anti-T15 is dependent upon the L chain and the D region of the H chain.

Amino acid sequence of the Fv region of a human monoclonal IgM (protein WEA) with antibody activity against 3,4-pyruvylated galactose in Klebsiella polysaccharides K30 and K33

We have determined the amino acid sequence of the Fv [variable heavy (VH) and variable light (VL)] region of a human monoclonal IgM-kappa with antibody activity against 3,4-pyruvylated galactose, isolated from the plasma of patient WEA with Waldenström macroglobulinemia. The VH region has 114 residues, belongs to subgroup III, and has a very short third complementarity-determining region (CDR3), probably due to a small D segment/or an unusual D-J rearrangement (D, diversity; J, joining). The VL region has 108 residues and belongs to subgroup V kappa I. Compared to other members of the human VHIII and V kappa I families, WEA Fv does not appear to have significant differences within the framework residues but has unique CDRs that might be responsible for the particular antibody activity. Another IgM-kappa (GAL), which has an as-yet-undetermined antibody activity, shares a striking homology in V kappa with WEA, including an identical CDR1.

Complete nucleotide sequence of an immunoglobulin VH gene homologue from Caiman, a phylogenetically ancient reptile

Immunoglobulin variable (V) gene regions typify extensive multigenic families in terms of overall size, chromosomal arrangement and presence of large numbers of apparent pseudogenes. A unique mechanism of somatic reorganization involving recombination of VH, D and JH or VL and JL segments accompanies the differentiation of lymphoid cells and together with somatic mutation and other types of recombination accounts for V-region diversity. Although these processes have been well characterized in higher mammals, little is known concerning their origin and diversification during phylogenetic time. Previously, we described the blot-hybridization characteristics of murine VHIII probes with restriction enzyme-digested genomic DNA isolated from several phylogenetically critical species, including Caiman crocodylus, a modern representative of an ancient reptilian subclass. Here we have used a murine probe, S107V, to select homologous clones from a library of Caiman genomic DNA constructed in a lambda bacteriophage. The complete nucleotide sequence of a Caiman gene homologous to the murine VH gene and its adjacent 5' and 3' region is described. Comparison of the sequence with mammalian prototypes shows evidence of considerable organizational and structural homology extending outside the presumed VH-coding region and including elements believed to be involved in somatic recombination. Inferences about the evolution of this multigenic family can now be extended to the level of phylogenetic class.

Structure of the catalytic and antigenic sites in influenza virus neuraminidase

The catalytic sites of influenza virus neuraminidase are located on the upper corners of the box-shaped tetramer that forms the head of the molecule. Antigenic determinants form a nearly-continuous surface across the top of the monomer encircling the catalytic site. Approximately the same number of amino acid sequence changes occurred in these determinants between the years 1968 and 1975 as occurred in the antigenic sites of influenza virus haemagglutinin in the same period.

Structural correlates of immunoglobulin diversity

VL and VH domains, from different species and with widely different primary structures, interact with each other in the same way to create the globular FV region. Much of the FV is a highly conserved framework structure that is probably common to most, if not all, mammalian FV regions. The extensive contoured frontal surface of the FV is composed of highly variable polypeptide segments (Wu-Kabat complementarity-determining regions). These segments are derived from parts of VL, VH, JH gene products and most of the D gene product. This surface is currently considered to be the most likely location of the antigen-binding sites. The firm immunochemical data based on identification of contacting amino acids supporting this location are still, however, very fragmentary. VL and VH gene products form a large part of the potential antigen-reactive surface. Hence, combinations of different VL and VH gene products are the largest source of structural diversity. The JL and JH gene products have chiefly structural functions in maintaining domain architecture and controlling some interactions between VL and VH domains. The VL-J junction amino acid can provide unique structural properties in the deeper accesses of the potential antigen reaction surface. The VHD-JH junction is more superficial and could be, but has not yet been, directly implicated in antigen binding. The D gene product and the additional amino acids associated with the (VH-D-JH) rearrangement process do determine a substructural part of the potential antigen reactive surface. The D gene product (a connecting segment between two beta strands) can have many different secondary structures. Functionally, the D region product could interact with VL-CDR-1 amino acids to create a specific contour of the antigen reaction surface. Curiously, primary structural variations in H3 have not yet been directly implicated in antigen binding. Much remains to be learned about the role of VH-D-JH rearrangement in antibody diversity. The major genetic factors in creating structural diversity are the multiple VL and VH gene libraries. The gene rearrangement process provides a further amplification. Somatic mutations are yet another additional mechanism.

An idiotypic determinant formed by both immunoglobulin constant and variable regions

Immunoglobulin idiotypes are serologically defined determinants associated with the variable (V) region of antibody molecules (reviewed in refs 1-4). One of the best defined idiotype systems is that borne by the phosphorylcholine (PC)-binding IgA proteins TEPC15 (T15) and HOPC8 (H8). The T15 idiotype, defined by sera raised in A strain mice, or in rabbits, is considered identical to that expressed by the majority of BALB/c anti-PC antibodies. To define the idiotypic determinants (idiotopes) of which the T15 idiotype is comprised, monoclonal anti-T15 antibodies were used here to examine both serum and monoclonal anti-PC antibodies. The latter were found to differ from T15 with respect to the idiotope defined by the monoclonal anti-idiotope antibody, 21A5, in that the '21A5 idiotope' was absent from anti-PC sera; of the monoclonal anti-PC antibodies examined, only those which were both T15+ and of the IgA isotype seemed to express this idiotype fully. This result suggests that not only the V region, but also the constant (C) region, of the immunoglobulin molecule can contribute to the formation of an idiotypic determinant. Isotype-restricted idiotopes may be involved in the regulation of antibody responses of particular classes.

Polymorphism in anti-phosphocholine antibodies reflecting evolution of immunoglobulin families

Complete variable (V) region amino acid sequences were determined for four heavy (H) and one light (L) chain from C57BL phosphocholine (PC)-binding monoclonal antibodies. Additional NH2-terminal sequences were obtained from H and L chains of C57BL and CBA/J origin. When these V regions were compared with previously reported anti-PC sequences, a number of observations could be made regarding the function and evolution of L and H chain segments used in these antibodies. (a) L and H chain V segments are remarkably conserved in these inbred strains, although there has been an accumulation of point mutations identifying apparently allelic forms of VK and VH. (b) Mice of each genotype use the same three VK segments in combination with a single VH segment to produce most anti-PC antibodies. An exception has been noted that indicates the occasional use of a second VH gene segment. (c) Multiple, different DH regions are used by mice of each strain, which suggests that the DH segment sequence plays no critical role in either antigen binding or VH-VL pairing. Furthermore, the DH segments and their corresponding gene families appear to be highly conserved in the inbred strains studied. (d) Most PC-binding antibodies use the JH1 joining segment. All JH1 sequences from C57BL mice differ from the BALB/c JH1 at position 105, which identifies allelic forms of the JH1 region. These studies are a first assessment of the nature of mutational events associated with the evolution of specific multigene immunoglobulin families and indicate that homologous VH, DH, JH, VK, and JK genes are similarly assembled and expressed in PC antibodies from three diverse genotypes.

A combined proton and phosphorus-31 nuclear magnetic resonance investigation of the combining site of M603, a phosphocholine-binding myeloma protein

Phosphorus-31 nuclear magnetic resonance (NMR) studies on the two phosphorus nuclei of the phosphonium analogue (Me3P+CH2CH2OPO3(2-)) of phosphocholine are used to monitor the charged subsites in the phosphocholine-binding immunoglobulin A mouse myeloma M603. Comparison of the 270-MHz 1H NMR difference spectrum on addition of either this analogue or phosphocholine to M603 and the almost identical changes in the pKa values of the phosphate groups on binding to M603 confirm that the analogue is a good model for phosphocholine. The pKa of the phosphate groups is decreased by 0.5 unit on binding to M603, which is consistent with the phosphate group being hydrogen bonding to Tyr-33H and Arg-95L, as suggested from the X-ray structure, and also implies that the binding energies for the mono- and dianion are similar. The P+Me3 moiety is used to probe the electrostatic interactions in the choline subsite. Titration of the chemical shift of the phosphonium phosphorus reflects a group on the protein that has a pKa value of less than or equal to 5, which from the refined X-ray structure (D.R. Davies, personal communication) of the site is assigned to Asp-97L. The choline subsite is monitored by using 1H NMR difference spectra, which indicates that the subsite is highly aromatic as expected from the crystal structure that places Trp-107H and Tyr-100L in this subsite. The ring current interactions from these rings can account for the 1H NMR chemical shift data on choline.

A small hypervariable segment in the variable domain of an immunoglobulin light chain stimulates formation of anti-idiotypic suppressor T cells

The induction in BALB/c mice of suppressor T cells that block a delayed-type hypersensitivity (DTH) response to the idiotype of M315, a myeloma protein of BALB/c origin, was examined with a variety of immunoglobulin chains and fragments whose amino acid sequences are known. Normal BALB/c mice receiving either the light chain of M315 (L315, lambda 2 isotype) or the variable (V) domain of this chain prior to sensitization with M315 showed marked suppression of DTH to the M315 idiotype. In contrast, neither the heavy chain nor the variable domain of the heavy chain of M315 affected the DTH response. Two other lambda 2 chains were tested and they also failed to suppress DTH to M315. Comparison of amino acid sequences in the three lambda 2 chains indicates that in L315 at most four V region amino acid substitutions (each resulting from a somatic mutation in the V lambda 2 germ-line gene) determine the specificity of the T-cell suppressor pathway. One of the four is in the framework and probably of negligible importance; the other three, however, are all clustered in the third hypervariable loop of the L315 V domain. The tertiary structure of L315 may also be essential, because disruption of intrachain disulfide bonds abolished the ability of the chain to induce suppression.

A thermodynamic study of cooperative structures in rabbit immunoglobulin G

The intramolecular melting of rabbit immunoglobulin G and its proteolytic fragments has been studied by scanning microcalorimetry in the acidic pH region. By thermodynamic analysis of the heat capacity function the following conclusions have been made. 1. The Fab and Fc fragments in IgG molecules are thermodynamically practically independent subunits in the pH region studied. 2. The Fab fragment exhibits three cooperative transitions at melting. Two of these transitions are explained assuming that equivalent domains in this fragment are paired, forming two cooperative blocks, while the third transition seems to correspond to the disruption of contacts between these cooperative blocks. 3. Fc and pFc' fragments exhibit four and two cooperative transitions, respectively. These transitions can be explained only by assuming that each of the structural domains of these fragments consists of two subparts and that the pairs of equivalent subparts are merged into the cooperative blocks thus forming four cooperative blocks in the Fc fragment.

Solid phase synthesis of the protected 43-55 tridecapeptide of the heavy chain of myeloma immunoglobin M603, employing cyclohexyl ester protection for glutamic acid

A protected tridecapeptide of the sequence Boc-Lys(2CIZ)-Arg(Tos)-Leu-Glu (OcHex)-Trp(For)-Ile-Ala-Ala-Ser(Bzl)-Arg(Tos)-Asn-Lys(2CIZ)-Gly-OH, representing residues 43-55 of the variable region of the heavy chain of mouse myeloma protein M603, was synthesized. It was assembled by a stepwise solid phase method designed to give a fully protected peptide in high yield and purity with minimal side reactions. Thus, the peptide chain was attached as an alpha-methyl phenacyl ester to a 2-bromopropionyl-resin. After the synthesis the protected peptide fragment was obtained in 89% yield by photolytic cleavage from the resin. The peptide was purified by multiple precipitation and column chromatography. It was shown to be homogeneous by reverse phase high pressure liquid chromatography, and it had the correct amino acid composition and sequence. In the course of this work it was shown that tert.-butyloxycarbonyl-amino acids caused the formation of significant amounts of pyrrolidone carboxylic acid residues during the coupling reaction when a gamma-benzyl glutamyl residue was NH2-terminal. Other weak-acid additives also caused this chain terminating side reaction. The cyclization was markedly suppressed by protection of the glutamyl side chain as a cyclohexyl ester. With this protecting group, no evidence of pyrrolidone carboxylic acid formation could be detected in the tridecapeptide 43-55.

Single amino acid substitution altering antigen-binding specificity

S107, a phosphocholine-binding myeloma protein, has been cloned in soft agar, and an antigen-binding variant has been isolated and characterized. The variant does not bind phosphocholine attached to carrier or as free hapten in solution but does retain antigenic determinants (idiotypes) of the parent. Chain recombination experiments suggest that the defect in binding is entirely in the heavy chain. Amino acid sequence analysis showed a single substitution--glutamic acid to alanine at position 35--in the first hypervariable or complementarity-determining region. In terms of the three-dimensional model of the phosphocholine-binding site, glutamic acid-35 provides a hydrogen bond to tyrosine-94 of the light chain that appears to be critical for stability of this portion of the binding site. The removal of this bond and the presence of the smaller alanine side chain is thus consistent with the loss in binding activity. These results suggest that small numbers of substitutions in antibodies, such as those presumably introduced by somatic mutation, may in some situations be effective in altering antigen-binding specificity.

Participation of the major histocompatibility complex in antibody recognition of viral antigens expressed on infected cells

The capacity of the antibody repertoire to recognize complex antigens on viral-infected cells was investigated at the level of monoclonal B cell responses. A majority of primary B cells responsive to PR8(H1N1)-infected H-2 syngeneic cells produced antibody that bound viral determinants only in the context of infected cells and not the isolated virion. An examination of the fine specificity of such antibodies revealed that most could be distinguished by a panel of cells infected with closely related heterologous H1 influenza strains. Indeed, most antibodies bound hemagglutinin determinants of PR8 exclusively, and few were broadly cross-reactive. An examination of the same antibodies for their recognition of cell surface antigens revealed that the majority recognized MHC-encoded antigenic determinants. Thus, most BALB.K (H-2k) primary B cells responsive to PR8-L919 (H-2k) cells produced monoclonal antibodies that bound PR8-antigens only in the context of H-2Dk-infected cells. Most C57BL/10 (H-2b) B cells responsive to PR8-EL4 (H-2b) cells produced monoclonal antibodies that bound PR8 antigens only in the context of H-2Kb-infected cells. These latter antibodies were further shown to recognize that H-2Kb molecule by virtue of their capacity to be discriminated by a panel of PR8-infected H-2Kb mutant cells. These findings demonstrate that much of the antibody repertoire is capable of highly specific complex recognition of viral antigenic determinants in the context of the appropriate MHC alloantigen.

A monoclonal antibody that defines an idiotope with two subsites in galactan-binding myeloma proteins

An IgG1 monoclonal antibody HyX24-14 was derived from A/J mice that were immunized with the IgA XRPC24 (X24) galactan binding myeloma protein (GalBMP) of BALB/c origin by the Kohler-Milstein hybridoma technology. HyX24-14 specifically binds some but all GalBMP. Different patterns of binding using a panel of nine Gal BMP were found, depending upon the concentration of antibody and the antigenic target. From molecular models and amino acid sequence data, ti was proposed that the idiotope defined by HyX24-14 had two subsites, each of which appeared to be able to bind independently to the antibody.

A survey of x-ray diffraction studies of enzyme-other molecule interactions as possible models for receptor sites

In the absence of a structure for a hormone-receptor complex, one may ask what systems of known structure are most likely to provide information about hormone interactions. Here, I discussed enzyme-substrate, enzyme-(protein) inhibitor, enzyme fragment (S peptide), and antibody-hapten (or antigen) interactions as possible models. Following a study of a fairly inflexible hormone (insulin) and of a flexible hormone (glucagon), I commented on probable binding regions. Finally, my conclusion was that, at present, allosteric enzymes have many of the characteristics thought to be present in those hormone-receptor interactions which activate enzymes. This model does not necessarily apply in detail to examples of hormone interactions that affect permeability of the cell wall or activate genetic processes.