Jurgen Bajorath

The CD40 ligand, gp39, is defective in activated T cells from patients with X-linked hyper-IgM syndrome.

The prominent role of the CD40 receptor in B cell responses led us to investigate the role of the gp39-CD40 interaction in a group of primary immunodeficient patients with defective antibody production. Here we report that patients with hyper-IgM syndrome (HIM) have a defective gp39-CD40 interaction. B cells from HIM patients express functional CD40, but their T cells do not bind CD40-Ig. These patients expressed normal levels of gp39 mRNA, but these mRNAs encode defective gp39 proteins owing to mutations in the extracellular domain of gp39. Soluble recombinant forms of gp39 containing these mutations were unable to bind CD40 and drive normal B cell proliferation. The gene encoding gp39 was mapped to Xq26, the X chromosome region where the gene responsible for HIM had previously been mapped. These data suggest that a defect in gp39 is the basis of X-linked HIM.

Human B7-1 (CD80) and B7-2 (CD86) bind with similar avidities but distinct kinetics to CD28 and CTLA-4 receptors

B7-0 or B7-2 (CD86) is a T cell costimulatory molecule that binds the same receptors (CD28 and CTLA-4) as B7-1 (CD80), but shares with it only approximately 25% sequence identity and is expressed earlier during an immune response. Here we show that human CD86 maintains similar (within approximately 2- to 3-fold) overall receptor binding and T cell costimulatory properties as CD80. However, CD80 and CD86 did not bind equivalently to CTLA-4: CD80 bound Y100A, a form of CTLA4lg with a mutation in the CDR3-like region, > 200-fold better than did CD86; inhibition of CD80-mediated cellular responses required approximately 100-fold lower CTLA4lg concentrations; and CD80-CTLA4lg complexes dissociated 5- to 8-fold more slowly, Thus, CD80 and CD86 utilize different binding determinants and have different kinetics of binding to CD28 and CTLA-4.

Rational Development of LEA29Y (belatacept), a High‐Affinity Variant of CTLA4‐Ig with Potent Immunosuppressive Properties

Current success in organ transplantation is dependent upon the use of calcineurin‐inhibitor‐based immunosuppressive regimens. Unfortunately, current immunotherapy targets molecules with ubiquitous expression resulting in devastating non‐immune side effects. T‐cell costimulation has been identified as a new potential immunosuppressive target. The best characterized pathway includes CD28, its homologue CTLA4 and their ligands CD80 and CD86. While an immunoglobulin fusion protein construct of CTLA4 suppressed rejection in rodents, it lacked efficacy in primate transplant models. In an attempt to increase the biologic potency of the parent molecule a novel, modified version of CTLA4‐Ig, LEA29Y (belatacept), was constructed. Two amino acid substitutions (L104E and A29Y) gave rise to slower dissociation rates for both CD86 and CD80. The increased avidity resulted in a 10‐fold increase in potency in vitro and significant prolongation of renal allograft survival in a pre‐clinical primate model. The use of immunoselective biologics may provide effective maintenance immunosuppression while avoiding the collateral toxicities associated with conventional immunsuppressants.

Both Extracellular Immunoglobin-like Domains of CD80 Contain Residues Critical for Binding T Cell Surface Receptors CTLA-4 and CD28

The B7-related molecules CD80 and CD86 are expressed on antigen-presenting cells, bind the homologous T cell receptors CD28 and CTLA-4, and trigger costimulatory signals important for optimal T cell activation. All four molecules are immunoglobulin superfamily members, each comprising an extracellular Ig variable-like (IgV) domain, with CD80 and CD86 containing an additional Ig constant-like (IgC) domain. Despite limited sequence identity, CD80 and CD86 share similar overall receptor binding properties and effector functions. We have identified, by site-directed mutagenesis of soluble forms of CD80 and CD86, residues in both the IgV and IgC domains that are important for CTLA4Ig and CD28Ig binding. Mutagenesis in the IgV domain of CD80 identified 11 amino acids that support receptor binding. Many of these residues are conserved in the B7 family, are hydrophobic, and approximately map to the GFCC′C″ β-sheet face of an IgV fold. Mutagenesis of corresponding residues in CD86 established that some, but not all, of these residues also played a role in CD86 receptor binding. In general, mutations had a similar effect on CTLA4Ig and CD28Ig binding, thereby indicating that both receptors bind to overlapping sites on CD80 and CD86. Further, mutagenesis of several conserved residues in the ABED β-sheet face of the IgC domain of CD80 completely ablated receptor binding. Point mutagenesis had a more pronounced effect than complete truncation of the IgC domain. Thus, full CTLA4Ig and CD28Ig binding to B7 molecules is dependent upon residues in the GFC′C″ face of the IgV domain and the ABED face of the IgC domain.

The Amino-terminal Immunoglobulin-like Domain of Activated Leukocyte Cell Adhesion Molecule Binds Specifically to the Membrane-proximal Scavenger Receptor Cysteine-rich Domain of CD6 with a 1:1 Stoichiometry

Activated leukocyte cell adhesion molecule (ALCAM) was recently identified as a ligand for CD6, a signaling receptor expressed on T cells, a subset of B cells, and some cells in the brain. Receptor-ligand binding assays, antibody blocking experiments, and examination of the tissue distribution of these two cell surface proteins suggest that CD6-ALCAM interactions play an important role in mediating the binding of thymocytes to thymic epithelial cells and of T cells to activated leukocytes. Presently, the details of CD6-ALCAM interactions and of signaling through CD6 are unknown. A series of truncated human ALCAM and CD6 immunoglobulin fusion proteins were produced and tested in different binding assays to analyze ALCAM-CD6 interactions in more detail. In this study, we report that the amino-terminal Ig-like domain of human ALCAM specifically binds to the third membrane-proximal scavenger receptor cysteine-rich (SRCR) domain of human CD6. Using thrombin-cleaved Ig fusion proteins containing single or multiple ALCAM or CD6 domains, we were able to determine that the stoichiometry of the interaction between the amino-terminal ALCAM domains and the membrane-proximal CD6 SRCR domain is 1:1. These results provide the first example of an Ig-like domain mediating an interaction with an SRCR domain.

Analysis of 4-1BBL and Laminin Binding to Murine 4-1BB, a Member of the Tumor Necrosis Factor Receptor Superfamily, and Comparison with Human 4-1BB

The T cell activation antigen 4-1BB (CDw137) is a distantly related member of the tumor necrosis factor receptor family of cell surface receptors. We previously reported that murine 4-1BB (m4-1BB) bound to extracellular matrix (ECM) proteins. Recently, a tumor necrosis factor-like ligand of m4-1BB, m4-1BBL, as well as the human counterparts of 4-1BB (ILA) and 4-1BBL (h4-1BB and h4-1BBL, respectively) have been cloned. No information is currently available on how binding of m4-1BB to ECM proteins affects its binding to m4-1BBL and vice versa and if the ability of m4-1BB to bind ECM proteins is conserved across species. We report that binding of m4-1BBL to m4-1BB blocked its ability to bind laminin (LN), while binding of m4-1BB to LN did not block its ability to bind m4-1BBL. Furthermore, binding of m4-1BBL to the m4-1BB·LN complex did not displace LN. These findings suggest the two ligands bind to proximal but distinct sites on m4-1BB. This is supported by the observation that six of eight anti-m4-1BB monoclonal antibodies blocked the interaction between 4-1BB and 4-1BBL, while seven blocked LN binding. Ligand and monoclonal antibody binding studies with a truncated protein lacking the amino-terminal LN-homologous domain of m4-1BB demonstrated that regions downstream of the LN-homologous domain participate in LN binding and that the intact protein is required for m4-1BBL binding. Studies with h4-1BB showed that h4-1BB only bound h4-1BBL, indicating that the ECM binding activity of 4-1BB is not conserved across species. This finding allowed the construction of murine/human 4-1BB chimeras, which permitted further dissection of the regions of 4-1BB involved in LN and 4-1BBL binding and suggests that sequence differences in the LN-homologous domain of h4-1BB in part account for the inability of h4-1BB to bind ECM proteins.

Construction and analysis of a detailed three-dimensional model of the ligand binding domain of the human B cell receptor CD40.

The interaction between the human B cell receptor CD40 and its ligand on T cells is critical for B cell proliferation and the regulation of humoral immune responses. CD40 is a member of the tumor necrosis factor receptor (TNFR) family. We report here the construction and analysis of a detailed three‐dimensional model of the TNFR‐homologous extracellular region of CD40. This study provides an example for structure‐based model building in the presence of low sequence similarity. The assessment of model quality and sequence‐structure compatibility is emphasized, and limitations of the model are discussed. The current CD40 model predicts structural details beyond the backbone level. Features of the CD40 ligand binding site are discussed in conjunction with the results of a previous mutagenesis study. Proteins 27:59–70

Comparison of an antibody model with an X‐ray structure: The variable fragment of BR96

A model of the BR96 antibody variable regions is compared to two X‐ray structures of a BR96–carbohydrate complex, independently determined after the model was built and analyzed. The comparison illustrates the opportunities and limitations of antibody modeling. Encouraging results were obtained for the prediction of single CDR loop conformations and for the outline of the BR96 antigen binding site. The comparison of CDR loop conformations in the two X‐ray structures provides a realistic reference frame for the CDR loop predictions. CDR loop prediction accuracy is lower when not only conformational, but also positional criteria are taken into account.

On the use of minimization from many randomly generated loop structures in modeling antibody combining sites

We describe an approach to modeling the combining sites of antibodies that is based on energy minimization from many randomly generated starting conformations. This technique combines the Random Tweak method for loop backbone generation with a combinatorial approach to selecting side-chain conformations from site-specific customized rotamer libraries. Generated structures are subjected to screening for van der Waals overlaps, and acceptable conformations are refined using energy minimization. Low-energy loop structures are then evaluated using energetic, solvation, and knowledge-based criteria. The application of a solvent model that combines the continuum electrostatic program DelPhi with a surface area model for hydrophobicity is presented. Random Tweak-based modeling is compared with other systematic conformational search methods. Results obtained for two hypervariable loops in the antibody HyHel5 are reported. Random Tweak-based modeling is combined with knowledge-based predictions and loop-search techniques into an overall multideterminantal strategy for modeling novel antibody combining sites.

Conformational Similarity and Systematic Displacement of Complementarity Determining Region Loops in High Resolution Antibody X-ray Structures

Comparison of seven high resolution x-ray structures shows that the conformations of canonical complementarity determining region (CDR) loops, which are shared by these antibodies, are very similar. However, large spatial displacements (up to 2.7 Å) of the essentially identical CDR loops become evident when the antibody β-sheet frameworks, to which the loops are attached, are least-squares superposed. The loop displacements follow, and amplify, small positional differences in framework/loop splice points. Intradomain structural variability and, to a lesser extent, domain-domain orientation appear to cause the observed loop divergences. The results suggest that the selection of framework regions for loop grafting procedures is more critical than previously thought. Immunoglobulin variable domains, VL 1 and VH, associate noncovalently to form the Fv, a dimer of antiparallel, eight-stranded β-sandwiches(1 2). The VL and VH β-sheets from different antibodies are nearly identical in three dimensions. However, the six complementarity determining region (CDR) loops (L1-L3, H1-H3), which connect the β-strands of the conserved framework and encode antigenic specificities, are much more variable in both sequence and conformation (2 3). Chothia, Lesk, and colleagues (4 5) identified sets of similar “canonical” conformations for all CDR loops except H3. Between 50 and 95% of antibody sequences are consistent with the classified canonical conformations(4), which are determined by conserved interactions of only a few key residues (“structural determinants”) within the loop and/or the framework regions. Some differences in the position of canonical CDR loops relative to superposed framework regions by comparing x-ray structures (4) or x-ray and modeled structures (5) were previously observed. However, these effects were observed in structures determined at medium resolution, generally considered minor(4 5), and not systematically explored. We have compared, via least-squares superpositions, the Fvs of seven x-ray structures refined to high resolution (better than 2 Å) available in the Brookhaven Protein Data Bank(6). It was anticipated that a systematic comparison of structures determined to such high precision would (i) shed light on the general relation between CDR loop conformation and position and (ii) help to assess the limitations of a procedure widely used in comparative model building, i.e. splicing of loops from a known x-ray structure onto the conserved structural scaffold of an antibody model(7 8). Fig. 1 shows a comparison (9) of the amino acid sequences of the seven antibodies, which include various heavy and light chains (both κ and ) from free as well as antigen-complexed antibodies. The 4-4-20 Fv, the highest resolution structure with κ light chain, was used as the template on which backbone segments of the other Fvs were superposed. Cumulative backbone root mean square (rms) deviations of the β-strands were determined after superposing each of these antibodies on the 4-4-20 Fv. Two alternative least-squares superpositions were used, employing different pairs of equivalence residues. The S1 set of residues consisted of only the most conserved regions of the Fvs, i.e. the four short 4-residue segments (10) of the central β-sheets. The S2 set consisted of a more extended set of residues and included the majority of the β-sheet framework (Fig. 1) akin to Stanfield et al.(11). As can be seen in Table 1 the cumulative backbone rms deviations were small and the results obtained with the two superposition sets were similar.