Crystal structure of a human rhinovirus neutralizing antibody complexed with a peptide derived from viral capsid protein VP2.

The three-dimensional structure of the complex between the Fab fragment of an anti-human rhinovirus neutralizing antibody (8F5) and a cross-reactive synthetic peptide from the viral capsid protein VP2 has been determined at 2.5 A resolution by crystallographic methods. The refinement is presently at an R factor of 0.18 and the antigen-binding site and viral peptide are well defined. The peptide antigen adopts a compact fold by two tight turns and interacts through hydrogen bonds, some with ionic character, and van der Waals contacts with antibody residues from the six hypervariable loops as well as several framework amino acids. The conformation adopted by the peptide is closely related to the corresponding region of the viral protein VP2 on the surface of human rhinovirus 1A whose three-dimensional structure is known. Implications for the cross-reactivity between peptides and the viral capsid are discussed. The peptide-antibody interactions, together with the analysis of mutant viruses that escape neutralization by 8F5 suggest two different mechanisms for viral escape. The comparison between the complexed and uncomplexed antibody structures shows important conformational rearrangements, especially in the hypervariable loops of the heavy chain. Thus, it constitutes a clear example of the 'induced fit' molecular recognition mechanism.     

Testing the procedure of simulated annealing by refining homologous immunoglobulin light-chain dimers

The refinement of antigen-binding fragment structures by the method of simulated annealing was tested. Using the program X-PLOR, we refined the structure of one immunoglobulin light-chain dimer against 2.8 A diffraction data collected for a homologous light-chain dimer. The refinement proceeded smoothly; alpha-carbons of the conserved segments of the domain moved to the positions in the reference structure solved independently. An average movement of approximately 1.5 A for atoms in the variable domains (half of the molecule) was observed. Though the final R-factors and energy terms of the reference and test structures were very similar, some of the chain segments of the hypervariable loops (HVR3s) and the ends of some side chains did not converge to the positions in the reference structure. Therefore, although globally the refinement worked very well, positions of the loops and the side chains that are critical for immunoglobulin function have to be carefully examined by difference Fourier techniques.     

Crystal structure of a chimeric Fab' fragment of an antibody binding tumour cells.

The crystal structure of a chimeric Fab' fragment of a monoclonal antibody is presented. The Fab' comprises the murine light chain and heavy chain variable domains of the carcinoma-binding antibody B72.3 fused to the constant domain of human kappa, and the first constant domain and hinge domain of human gamma 4, respectively. A model for the Fab' has been determined by molecular replacement and refined to a resolution of 3.1 A with an R-factor of 17.6%. The additional residues that distinguish a Fab' from a Fab fragment are seen to be disordered in the crystals. The H3 hypervariable loop is short and adopts a sharp hairpin turn in a conformation that results from an interaction between the lysine side-chain of H93 and the main-chain carbonyl group of H96. The remaining hypervariable loops display conformations similar to those predicted from the canonical structures approach, although loop H2 is apparently displaced by a salt-bridge formed between H55 Asp and the neighbouring H73 Lys. These and other features of the structure likely to be important in grafting the hypervariable loops to an otherwise human framework are discussed.     

Modeling the antigen combining site of an anti-dinitrophenyl antibody, ANO2.

A model structure has been constructed for a monoclonal anti-dinitrophenyl antibody. The antibody, ANO2, has been sequenced and cloned (Anglister, J., Frey, T., & McConnell, H.M., 1984, Biochemistry 23, 1138-1142). Its amino acid sequence shows striking homology with the anti-lysozyme Fab fragments HyHel5 (83%) and HyHel10 (73%). Based on this homology, a model for the ANO2 variable heavy and variable light chain framework was constructed using a hybrid of the HyHel5 light chain and the HyHel10 heavy chain backbone, omitting the hypervariable loops. These coordinates were used as scaffolds for the model building of ANO2. The CONGEN conformational sampling algorithm (Bruccoleri, R.E. & Karplus, M., 1987, Biopolymers 26, 127-196) was used to model the six hypervariable loops that contain the antigen-combining site. All the possible conformations of the loop backbones were constructed and the best loop structures were selected using a combination of the CHARMM potential energy function and evaluation of the solvent-accessible surface area of the conformers. The order in which the loops were searched was carried out based on the relative locations of the loops with reference to the framework of the beta-barrel, namely, L2-H1-L3-H2-H3-L1. The model structures thus obtained were compared to the high resolution X-ray structure (Brünger, A.T., Leahy, D.J., Hynes, T.R., & Fox, R.O., 1991, J. Mol. Biol. 221, 239-256).     

VL:VH domain rotations in engineered antibodies: Crystal structures of the Fab fragments from two murine antitumor antibodies and their engineered human constructs

The crystal structures of two pairs of Fab fragments have been determined. The pairs comprise both a murine and an engineered human form, each derived from the antitumor antibodies A5B7 and CTM01. Although antigen specificity is maintained within the pairs, antigen affinity varies. A comparison of the hypervariable loops for each pair of antibodies shows their structure has been well maintained in grafting, supporting the canonical loop model. Detailed structural analysis of the binding sites and domain arrangements for these antibodies suggests the differences in antigen affinity observed are likely to be due to inherent flexibility of the hypervariable loops and movements at the V_L:V_H domain interface. The four structures provide the first opportunity to study in detail the effects of protein engineering on specific antibodies. Proteins 29:161–171, 1997. © 1997 Wiley-Liss, Inc.     

Antigen specificity and high affinity binding provided by one single loop of a camel single-domain antibody

Detailed knowledge on antibody-antigen recognition is scarce given the unlimited antibody specificities of which only few have been investigated at an atomic level. We report the crystal structures of an antibody fragment derived from a camel heavy chain antibody against carbonic anhydrase, free and in complex with antigen. Surprisingly, this single-domain antibody interacts with nanomolar affinity with the antigen through its third hypervariable loop (19 amino acids long), providing a flat interacting surface of 620 A(2). For the first time, a single-domain antibody is observed with its first hypervariable loop adopting a type-1 canonical structure. The second hypervariable loop, of unique size due to a somatic mutation, reveals a regular beta-turn. The third hypervariable loop covers the remaining hypervariable loops and the side of the domain that normally interacts with the variable domain of the light chain. Specific amino acid substitutions and reoriented side chains reshape this side of the domain and increase its hydrophilicity. Of interest is the substitution of the conserved Trp-103 by Arg because it opens new perspectives to 'humanize' a camel variable domain of heavy chain of heavy chain antibody (VHH) or to 'camelize' a human or a mouse variable domain of heavy chain of conventional antibody (VH).     

Canonical antigen-binding loop structures in immunoglobulins: more structures, more canonical classes?

Grafting the antigen-binding loops onto a human antibody scaffold is a widely used technique to humanise murine antibodies. The success of this approach depends largely on the observation that the antigen-binding loops adopt only a limited number of canonical structures. Identification of the correct canonical structure is therefore essential. Algorithms that predict the main-chain conformation of the hypervariable loops using only the amino acid sequence often provide this information. Here, we describe new canonical loop conformations for the hypervariable regions H1 and H2 as found in single-domain antibody fragments of dromedaries or llama. Although the occurrence of these new loop conformations was not predicted by the algorithms used, it seems that they could occur in human or mouse antigen-binding loops. Their discovery indicates that the currently used set of canonical structures is incomplete and that the prediction algorithms should be extended to include these new structures.     

Haloperidol binding to monoclonal antibodies. Hypervariable region amino acid sequence determination

The primary sequences of five monoclonal antibodies (mAbs A-E) which bind with various affinities (Kd = 4-810 nM) to the D-2 dopaminergic antagonist, haloperidol, have been determined. Immunoglobulin light and heavy chain mRNA was isolated and gene sequence determined by primer extension in the presence of dideoxynucleotides. The pattern of insertions and deletions found within the hypervariable regions produce loops which differ in length from one antibody to another, and are directly responsible for establishing the gross architecture of the combining site. Two of the anti-haloperidol mAbs have long hypervariable loops which form a pocket-shaped combining site. Three other mAbs have deletions of 3 or 4 amino acids in the third heavy chain complementarity producing region which result in a groove-like combining site as determined by computer based molecular modeling. A discussion of the probable mechanism by which the given sequences were generated from various gene segments is also presented.     

Crystallographic refinement and atomic models of the intact immunoglobulin molecule Kol and its antigen-binding fragment at 3.0 Å and 1.9 Å resolution

The crystal structures of the intact immunoglobulin G1, (λ) Kol and its Fab² fragment were crystallographically refined at 3.0 Å and 1.9 Å resolution, respectively. The methods used were real space refinement (RLSP) energy and residual refinement (EREF), phase combination, constrained rigid body refinement (CORELS) and difference and Fourier map inspection. The final R-values are 0.24 and 0.26. These analyses allowed the construction of atomic models of parts not seen in detail in the previous analyses at 5 Å and 3 Å resolution, respectively (Colman et al., 1976; Matsushima et al., 1978): i.e. the hinge segment, the hypervariable segments and their intimate interaction with the hinge segment of a crystallographically related molecule.The hinge segment forms a short poly-l-proline double helix from Cys527 to Cys530 (Eu numbering 226 to 230). The preceding segment forms an open turn of helix. This segment and the segment following the poly-l-proline part, which was found to be flexible in Fc fragment crystals (Deisenhofer et al., 1976) probably allow arm and stem movement of the antibody molecule. The combining site of Kol is compared with the combining site of Fab New (Saul et al., 1978). The narrow cleft formed by the hypervariable loops in Kol is filled with aromatic amino acid side-chains. In the crystal, the hypervariable loops contact the hinge and adjacent segments of a related molecule accompanied by a substantial loss in accessible surface area. This contact is preserved in Kol Fab crystals and presumably occurs in the Kol cryoprecipitate. A comparison of the quaternary structures of intact Kol and Fab New showed, in addition to the large change in elbow angle (Colman et al., 1976), changes in lateral domain association. These are discussed in the context of a possible signal transmission from the combining site to the distal end. An attempt was made to model build the IgG3 hinge segment, which is quadruplicated with respect to IgG1 (Michaelsen et al., 1977), on the basis of the Kol hinge structure. A polyproline double helix appeared to be the most plausible model. The Fc part was found to be disordered in intact Kol crystals (Colman et al., 1976). Refinement has reduced the electron density further in the crystal space, where the Fc parts must be located. Disorder, if static, must be fourfold or more in the crystalline state.Intensity measurements on Kol F(ab′)₂ and their comparison with intact Kol crystals provide evidence that the disorder is predominantly of a static nature.     

Structural repertoire of the human VH segments.

The VH gene segments produce the part of the VH domains of antibodies that contains the first two hypervariable regions. The sequences of 83 human VH segments with open reading frames, from several individuals, are currently known. It has been shown that these sequences are likely to form a high proportion of the total human repertoire and that an individual's gene repertoire produces about 50 VH segments with different protein sequences. In this paper we present a structural analysis of the amino acid sequences produced by the 83 segments. Particular residue patterns in the sequences of V domains imply particular main-chain conformations, canonical structures, for the hypervariable regions. We show that, in almost all cases, the residue patterns in the VH segments imply that the first hypervariable regions have one of three different canonical structures and that the second hypervariable regions have one of five different canonical structures. The different observed combinations of the canonical structures in the first and second regions means that almost all sequences have one of seven main-chain folds. We describe, in outline, structures of the antigen binding site loops produced by nearly all the VH segments. The exact specificity of the loops is produced by (1) sequence differences in their surface residues, particularly at sites near the centre of the combining site, and (2) sequence differences in the hypervariable and framework regions that modulate the relative positions of the loops.     

A single-domain antibody fragment in complex with RNase A: non-canonical loop structures and nanomolar affinity using two CDR loops

BACKGROUND: Camelid serum contains a large fraction of functional heavy-chain antibodies - homodimers of heavy chains without light chains. The variable domains of these heavy-chain antibodies (VHH) have a long complementarity determining region 3 (CDR3) loop that compensates for the absence of the antigen-binding loops of the variable light chains (VL). In the case of the VHH fragment cAb-Lys3, part of the 24 amino acid long CDR3 loop protrudes from the antigen-binding surface and inserts into the active-site cleft of its antigen, rendering cAb-Lys3 a competitive enzyme inhibitor. RESULTS: A dromedary VHH with specificity for bovine RNase A, cAb-RN05, has a short CDR3 loop of 12 amino acids and is not a competitive enzyme inhibitor. The structure of the cAb-RN05-RNase A complex has been solved at 2.8 A. The VHH scaffold architecture is close to that of a human VH (variable heavy chain). The structure of the antigen-binding hypervariable 1 loop (H1) of both cAb-RN05 and cAb-Lys3 differ from the known canonical structures; in addition these H1 loops resemble each other. The CDR3 provides an antigen-binding surface and shields the face of the domain that interacts with VL in conventional antibodies. CONCLUSIONS: VHHs adopt the common immunoglobulin fold of variable domains, but the antigen-binding loops deviate from the predicted canonical structure. We define a new canonical structure for the H1 loop of immunoglobulins, with cAb-RN05 and cAb-Lys3 as reference structures. This new loop structure might also occur in human or mouse VH domains. Surprisingly, only two loops are involved in antigen recognition; the CDR2 does not participate. Nevertheless, the antigen binding occurs with nanomolar affinities because of a preferential usage of mainchain atoms for antigen interaction.     

Modeling large regions in proteins: applications to loops, termini, and folding

Template-based methods for predicting protein structure provide models for a significant portion of the protein but often contain insertions or chain ends (InsEnds) of indeterminate conformation. The local structure prediction "problem" entails modeling the InsEnds onto the rest of the protein. A well-known limit involves predicting loops of ≤12 residues in crystal structures. However, InsEnds may contain as many as ~50 amino acids, and the template-based model of the protein itself may be imperfect. To address these challenges, we present a free modeling method for predicting the local structure of loops and large InsEnds in both crystal structures and template-based models. The approach uses single amino acid torsional angle "pivot" moves of the protein backbone with a C(β) level representation. Nevertheless, our accuracy for loops is comparable to existing methods. We also apply a more stringent test, the blind structure prediction and refinement categories of the CASP9 tournament, where we improve the quality of several homology based models by modeling InsEnds as long as 45 amino acids, sizes generally inaccessible to existing loop prediction methods. Our approach ranks as one of the best in the CASP9 refinement category that involves improving template-based models so that they can function as molecular replacement models to solve the phase problem for crystallographic structure determination.     

Modifications to canonical structure sequence patterns: analysis for L1 and L3

The conformation of five of the six hypervariable loops that form the antigen-binding site of antibodies is limited to a small set of structures designated as canonical structures. The canonical structure model has been constituted as a fundamental tool for the modeling of antibodies. The detailed study of tens of crystallographic structures of antibodies has shown the validity of this model in the great majority of cases. The robustness of the forecast capacity of this model depends fundamentally on the precision with which the sequence patterns that characterize each canonical structure form can be defined. Nevertheless, due to the enormous quantity of structural information about antibodies generated during the last decade, it is difficult to avoid mistakes or confusion in the model. In the present work, we propose some corrections to the model for loops L1 and L3 that permit defining sequence patterns that avoid confusion and make better forecasting of the canonical structure model possible.     

Refined structure of baboon alpha-lactalbumin at 1.7 A resolution. Comparison with C-type lysozyme

The solution of the structure of alpha-lactalbumin from baboon milk (Papio cynocephalus) at 4.5 A resolution using the isomorphous replacement method has been reported previously. Initial refinement on the basis of these low-resolution studies was not successful because of the poor isomorphism of the best heavy-atom derivative. Because of the striking similarity between the structure of lysozyme and alpha-lactalbumin, a more cautious molecular replacement approach was tried to refine the model. Using hen egg-white lysozyme as the starting model, preliminary refinement was performed using heavily constrained least-squares minimization in reciprocal space. The model was further refined using stereochemical restraints at 1.7 A resolution to a conventional crystallographic residual of 0.22 for 1141 protein atoms. In the final model, the root-mean-square deviation from ideality for bond distances is 0.015 A, and for angle distances it is 0.027 A. The refinement was carried out using the human alpha-lactalbumin sequence and "omit maps" calculated during the course of refinement indicated eight possible sequence changes in the baboon alpha-lactalbumin X-ray sequence. During the refinement, a tightly bound calcium ion and 150 water molecules, of which four are internal, have been located. Some of the water molecules were modelled for disordered side-chains. The co-ordination around the calcium is a slightly distorted pentagonal bipyramid. The Ca-O distances vary from 2.2 A to 2.6 A, representing a tight calcium-binding loop in the structure. The calcium-binding fold only superficially resembles the "EF-hand" and presumably has no evolutionary relationship with other EF-hand structures. The overall structure of alpha-lactalbumin is very similar to that of lysozyme. All large deviations occur in the loops where all sequence deletions and insertions are found. The C terminus appears to be rather flexible in alpha-lactalbumin compared to lysozyme. The experimental evidence supports the earlier predictions for the alpha-lactalbumin structure that were based upon the assumption that alpha-lactalbumin and lysozyme have similar three-dimensional structures, with minimal deletions and insertions. A detailed comparison of the two structures shows striking features as well as throwing some light on the evolution of these two proteins from a common precursor.     

Disparate degrees of hypervariable loop flexibility control T-cell receptor cross-reactivity, specificity, and binding mechanism

αβ T-cell receptors (TCRs) recognize multiple antigenic peptides bound and presented by major histocompatibility complex molecules. TCR cross-reactivity has been attributed in part to the flexibility of TCR complementarity-determining region (CDR) loops, yet there have been limited direct studies of loop dynamics to determine the extent of its role. Here we studied the flexibility of the binding loops of the αβ TCR A6 using crystallographic, spectroscopic, and computational methods. A significant role for flexibility in binding and cross-reactivity was indicated only for the CDR3α and CDR3β hypervariable loops. Examination of the energy landscapes of these two loops indicated that CDR3β possesses a broad, smooth energy landscape, leading to rapid sampling in the free TCR of a range of conformations compatible with different ligands. The landscape for CDR3α is more rugged, resulting in more limited conformational sampling that leads to specificity for a reduced set of peptides as well as the major histocompatibility complex protein. In addition to informing on the mechanisms of cross-reactivity and specificity, the energy landscapes of the two loops indicate a complex mechanism for TCR binding, incorporating elements of both conformational selection and induced fit in a manner that blends features of popular models for TCR recognition.     

Length of the antibody heavy chain complementarity determining region 3 as a specificity-determining factor

The antigen binding site of an antibody is made up of residues residing in six hypervariable loops of the heavy and light chains. In most cases several or all of these loops are required for the establishment of the antigen-binding surface. Five of these loops display a limited diversity in length and sequence while the third complementarity determining region (CDR) of the heavy chain is highly different between antibodies not only with respect to sequence but also with respect to length. Its extensive diversity is a key component in the establishment of binding sites allowing for the recognition of essentially any antigen by humoral immunity. The relative importance of its sequence vs its length diversity in this context is however, not very well established. To investigate this matter further we have used an approach employing combinatorial antibody libraries and antigen-specific selection in the search for CDRH3 length and sequence diversity compatible with a given antigen specificity, the major antigenic determinant on the tumour-associated antigen mucin-1. In this way we have now defined heavy chain CDR3 length as a critical parameter in the creation of an antigen-specific binding site. We also propose that this may reflect a dependence of a particular structure of this hypervariable loop, the major carrier of diversity in the binding site, for establishment of a given specificity.     

Functional consequences of insertions and deletions in the complementarity-determining regions of human antibodies

Insertions and deletions of nucleotides in the genes encoding the variable domains of antibodies are natural components of the hypermutation process, which may expand the available repertoire of hypervariable loop lengths and conformations. Although insertion of amino acids has also been utilized in antibody engineering, little is known about the functional consequences of such modifications. To investigate this further, we have introduced single-codon insertions and deletions as well as more complex modifications in the complementarity-determining regions of human antibody fragments with different specificities. Our results demonstrate that single amino acid insertions and deletions are generally well tolerated and permit production of stably folded proteins, often with retained antigen recognition, despite the fact that the thus modified loops carry amino acids that are disallowed at key residue positions in canonical loops of the corresponding length or are of a length not associated with a known canonical structure. We have thus shown that single-codon insertions and deletions can efficiently be utilized to expand structure and sequence space of the antigen-binding site beyond what is encoded by the germline gene repertoire.     

Structure determination of a monoclonal Fab fragment specific for histidine-containing protein of the phosphoenolpyruvate: sugar phosphotransferase system of Escherichia coli

Jel 42 is a monoclonal antibody specific for histidine-containing protein, a small phosphocarrier protein required for sugar transport in Escherichia coli. Fab fragments prepared from this antibody by papain digestion consisted of three major isoelectric forms which were separated on a chromatofocusing column. Two of these forms produced large crystals in space group P21 and unit cell dimensions a = 117.48 A, b = 66.56 A, c = 67.31 A, and beta = 118.7 degrees, with two Fab fragments per asymmetric unit. Data were collected to 3.5-A resolution. The structure of Fab Jel 42 was solved by the Molecular Replacement method (least-squares refined to R = 0.282) using the known structure of Fab HED 10 (12) as the search model; the amino acid residues of the hypervariable and elbow regions of Fab HED 10 were omitted from the starting model. A Fourier map calculated at this stage revealed electron density which corresponded to the hypervariable loops forming the antigen-binding crevice and the elbow region of Fab Jel 42. The elbow angles for the two independent Fab molecules are 159 and 167 degrees, similar to that of the Fab HED 10 search model which has an elbow angle of 162 degrees. There is no local noncrystallographic axis of symmetry relating the two molecules in the asymmetric unit.     

Monte Carlo loop refinement and virtual screening of the thyroid-stimulating hormone receptor transmembrane domain

Metropolis Monte Carlo (MMC) loop refinement has been performed on the three extracellular loops (ECLs) of rhodopsin and opsin-based homology models of the thyroid-stimulating hormone receptor transmembrane domain, a class A type G protein-coupled receptor. The Monte Carlo sampling technique, employing torsion angles of amino acid side chains and local moves for the six consecutive backbone torsion angles, has previously reproduced the conformation of several loops with known crystal structures with accuracy consistently less than 2?Å. A grid-based potential map, which includes van der Waals, electrostatics, hydrophobic as well as hydrogen-bond potentials for bulk protein environment and the solvation effect, has been used to significantly reduce the computational cost of energy evaluation. A modified sigmoidal distance-dependent dielectric function has been implemented in conjunction with the desolvation and hydrogen-bonding terms. A long high-temperature simulation with 2?kcal/mol repulsion potential resulted in extensive sampling of the conformational space. The slow annealing leading to the low-energy structures predicted secondary structure by the MMC technique. Molecular docking with the reported agonist reproduced the binding site within 1.5?Å. Virtual screening performed on the three lowest structures showed that the ligand-binding mode in the inter-helical region is dependent on the ECL conformations.     

Three-dimensional structure of an Fv from a human IgM immunoglobulin.

An IgM(kappa) immunoglobulin from a patient (Pot) with Waldenstrom's macroglobulinemia was hydrolyzed with pepsin to release a fragment consisting of the 'variable' (V) domains of the light and heavy chains plus eight residue 'tails' from the 'constant' (C) domains. The crystal structure of this fragment was determined at 2.3 A resolution by molecular replacement and crystallographic refinement methods. When examined separately, the light chain component closely resembles another human kappa chain (Rei) in both the beta-pleated sheet regions and the 'hypervariable' loops. The conserved pleated sheets in the heavy chain are similar to those in the human Kol IgG1 protein, but the third hypervariable loop in particular is different from that in any immunoglobulin structure described to date. As in the Kol protein, this loop blocks the access to any internal active site along the light-heavy chain interface. Unlike the Kol protein, however, the loop does not protrude beyond the boundaries of a conventional antigen combining site. Instead, it forms a very compact structure, which fills almost all residual space between the domains. This is an example of one dominant complementarity-determining region (CDR) essentially negating the diversity possible with five other CDRs in the two chains. Ordered water molecules are associated with light chain constituents along the interface, but not with CDR3 of the heavy chain. In screening exercises the Pot IgM failed to bind a wide variety of peptides. Together, the results suggest that ligand binding can only occur on external surfaces of the protein. These surfaces carry a limited number of side chains usually assigned to CDRs in more typical antibodies.