Critical contribution of VH-VL interaction to reshaping of an antibody: the case of humanization of anti-lysozyme antibody, HyHEL-10

To clarify the effects of humanizing a murine antibody on its specificity and affinity for its target, we examined the interaction between hen egg white lysozyme (HEL) and its antibody, HyHEL-10 variable domain fragment (Fv). We selected a human antibody framework sequence with high homology, grafted sequences of six complementarity-determining regions of murine HyHEL-10 onto the framework, and investigated the interactions between the mutant Fvs and HEL. Isothermal titration calorimetry indicated that the humanization led to 10-fold reduced affinity of the antibody for its target, due to an unfavorable entropy change. Two mutations together into the interface of the variable domains, however, led to complete recovery of antibody affinity and specificity for the target, due to reduction of the unfavorable entropy change. X-ray crystallography of the complex of humanized antibodies, including two mutants, with HEL demonstrated that the complexes had almost identical structures and also paratope and epitope residues were almost conserved, except for complementary association of variable domains. We conclude that adjustment of the interfacial structures of variable domains can contribute to the reversal of losses of affinity or specificity caused by humanization of murine antibodies, suggesting that appropriate association of variable domains is critical for humanization of murine antibodies without loss of function.     

Crystal structure of a polymeric immunoglobulin binding fragment of the human polymeric immunoglobulin receptor

The polymeric immunoglobulin receptor (pIgR) is a type I transmembrane protein that delivers dimeric IgA (dIgA) and pentameric IgM to mucosal secretions. Here, we report the 1.9 A resolution X-ray crystal structure of the N-terminal domain of human pIgR, which binds dIgA in the absence of other pIgR domains with an equilibrium dissociation constant of 300 nM. The structure of pIgR domain 1 reveals a folding topology similar to immunoglobulin variable domains, but with differences in the counterparts of the complementarity determining regions (CDRs), including a helical turn in CDR1 and a CDR3 loop that points away from the other CDRs. The unusual CDR3 loop position prevents dimerization analogous to the pairing of antibody variable heavy and variable light domains. The pIgR domain 1 structure allows interpretation of previous mutagenesis results and structure-based comparisons between pIgR and other IgA receptors.     

Stabilization of antibody VH-domains by proteolytic selection

Single variable domains of antibodies represent the smallest antigen binding fragments but are less stable than when associated with their cognate variable domains. Here we have attempted to improve the thermodynamic stability of a model heavy chain variable domain (V_H) by “proteolytic selection” a method whereby the protease-resistance of the displayed protein is coupled to the infectivity of a filamentous bacteriophage. The gene encoding the heavy chain variable domain was taken from the anti-lysozyme antibody HyHEL-10, mutated at random by error-prone PCR, and displayed on filamentous bacteriophage by fusion between the domains of the phage p3 protein. As the entire p3 protein is required for phage infectivity, treatment of the phage library with trypsin at an elevated temperature (which leads to cleavage of p3 fusions with unfolded variable domains) selects for infectious phages bearing the more stable variable domains. After several rounds of selection, a mutant (S65G/T70S/D99N) was obtained with improved stability (T_m=58.5 °C and ΔG_25°C=6.3 kcal/mol compared to 51.6 °C and 4.2 kcal/mol for the parent domain). These mutations are conservative and the mutant domain retains the ability to pair with its cognate light chain variable domain in an Fv fragment and to mediate binding to lysozyme. Our results show that the thermodynamic stability of antibody single domains can be improved by “proteolytic selection” and this may represent a step towards making useful antibody single domains for biotechnological application.     

Fibronectin's cell-adhesive domain and an amino-terminal matrix assembly domain participate in its assembly into fibroblast pericellular matrix

Fibroblasts organize the modular cell-adhesive glycoprotein fibronectin into a highly structured pericellular matrix by poorly understood mechanisms. Previous studies implicated an amino-terminal domain in matrix assembly and suggested that fibronectin's cell-adhesive domain and the corresponding fibroblast receptor were not involved in this process. To further elucidate the fibronectin region(s) involved in matrix assembly, we mapped a library of proteolytic fragments and antibodies to various fibronectin domains. The fragments and antibodies were used to probe the role of fibronectin's amino-terminal and cell-adhesive domains in a fibroblast matrix assembly assay. We found that fibronectin fragments including the first 25-kDa sequence of fibronectin and antibodies to amino-terminal domains inhibited pericellular matrix assembly. Polyclonal antibodies to the 40-kDa collagen binding domain following the 25-kDa amino-terminal domain also inhibited matrix assembly. However, collagen binding is not required for matrix assembly as neither monoclonals blocking collagen binding nor purified collagen binding domains themselves inhibited matrix assembly. Therefore, the amino-terminal region of fibronectin contains a site important in matrix assembly, and most activity is present in the first 25-kDa of fibronectin. Fibronectin's cell-adhesive domain and the fibroblast receptor binding to this domain also play an important role in fibronectin matrix assembly. Apart from a monoclonal antibody to the amino-terminal domain, only monoclonal antibodies binding to fibronectin's cell-adhesive domain and inhibiting cell adhesion also inhibited matrix assembly. In addition a 105-kDa fragment containing the cell-adhesive domain inhibited matrix assembly. We conclude that at least two discrete and widely separated sites in fibronectin with different binding properties--the carboxyl-terminal fibroblast cell-adhesive domain and an amino-terminal matrix assembly domain localized primarily within the first 25 kDa--are required for fibronectin pericellular matrix assembly by fibroblasts. Fibronectin's cell-adhesive domain and its cell surface-receptor complex appear to be involved in the matrix assembly process prior to a step involving the amino-terminal domain. We believe that this step is likely to be the initiation of cell-associated fibronectin fibril formation by the fibronectin-adhesive-receptor complex.     

Energy-Based Analysis and Prediction of the Orientation between Light- and Heavy-Chain Antibody Variable Domains

Diversity in antibody structure is crucial to the ability of the adaptive immune system to recognize the tremendously diverse set of potential antigens. The diversity in structure is most apparent in the six hypervariable loops of the complementarity-determining regions. However, given that these loops occur at the interface of the heavy- and light-chain variable domains and form the antigen-binding site, the relative orientation of the heavy- and light-chain variable domains can create another source of structural diversity leading to changes in antigen binding. Here, we first reexamine the diversity of V_L:V_H orientations in existing antibody crystal structures using 153 nonredundant sequences, demonstrating that the variation in V_L:V_H orientation is greater than that expected from effects of crystal packing, antigen binding, or the presence of antibody constant regions and increases, on average, as sequence similarity decreases for residues in the interface between the domains. We developed a tool for predicting the relative orientations of the heavy- and light-chain variable domains using side-chain rotamer sampling in the interface and molecular-mechanics-based energy calculations. When using variable domain backbones from the crystal structures, the predicted orientation is very close (< 1 Å RMSD) to the crystallographically observed orientation in most cases, confirming that the V_L:V_H orientation is determined by the antibody sequence and suggesting an approach to predicting the relative orientation of the variable domains when building homology models of antibodies. When applied to antibody homology models generated from templates with 55-75% sequence identity, we predict the V_L:V_H orientation of 20 antibodies with an average/median RMSD of 2.1/1.6 Å to the crystal structures.     

Combining phage display with de novo protein sequencing for reverse engineering of monoclonal antibodies

The enormous diversity created by gene recombination and somatic hypermutation makes de novo protein sequencing of monoclonal antibodies a uniquely challenging problem. Modern mass spectrometry-based sequencing will rarely, if ever, provide a single unambiguous sequence for the variable domains. A more likely outcome is computation of an ensemble of highly similar sequences that can satisfy the experimental data. This outcome can result in the need for empirical testing of many candidate sequences, sometimes iteratively, to identity one which can replicate the activity of the parental antibody. Here we describe an improved approach to antibody protein sequencing by using phage display technology to generate a combinatorial library of sequences that satisfy the mass spectrometry data, and selecting for functional candidates that bind antigen. This approach was used to reverse engineer 2 commercially-obtained monoclonal antibodies against murine CD137. Proteomic data enabled us to assign the majority of the variable domain sequences, with the exception of 3–5% of the sequence located within or adjacent to complementarity-determining regions. To efficiently resolve the sequence in these regions, small phage-displayed libraries were generated and subjected to antigen binding selection. Following enrichment of antigen-binding clones, 2 clones were selected for each antibody and recombinantly expressed as antigen-binding fragments (Fabs). In both cases, the reverse-engineered Fabs exhibited identical antigen binding affinity, within error, as Fabs produced from the commercial IgGs. This combination of proteomic and protein engineering techniques provides a useful approach to simplifying the technically challenging process of reverse engineering monoclonal antibodies from protein material.     

Inhibition of microtubule assembly in vitro by anti-tubulin monoclonal antibodies

Five monoclonal antibodies against N-terminal domains of alpha- or beta-tubulin were tested for their ability to interfere with the in vitro formation of microtubules. Although all the antibodies exhibited similar association constants for immobilized tubulin, they differed in their inhibitory effect on microtubule assembly. For the most potent antibody, TU-13, the antibody/tubulin molar ratio of about 1:320 was sufficient for a 50% inhibition. The data indicate that the surface regions of N-terminal domains of tubulin are involved in the formation of microtubules.     

Contributions of a highly conserved VH/VL hydrogen bonding interaction to scFv folding stability and refolding efficiency.

The assembly of single-chain Fv (scFv) antibody fragments, consisting of an interconnected variable heavy chain (VH) and variable light chain (VL), is a cooperative process that requires coupled folding and domain association. We report here an initial investigation of VH/VL domain-domain assembly with a site-directed mutagenesis study that probes a highly conserved VH/VL hydrogen bonding interaction. Gln168 of the S5 scFv (Kabat VH 39) is absolutely conserved in 95% of all VH, and Gln44 (Kabat VL 38) is found in 94% of all kappa VL (Glx in 95% of all lambda VL). These side chains form two hydrogen bonds in head-to-tail alignment across the VH/VL interface. Double mutant cycles at Gln168 and Gln44 were constructed to first investigate their contribution to thermodynamic folding stability, second to investigate whether stability can be improved, and third to determine whether refolding efficiencies are affected by mutations at these positions. The results demonstrate that the Gln168-Gln44 interaction is not a key determinant of S5 scFv folding stability, as sequential modification to alanine has no significant effect on the free energy of folding. Several mutations that alter the glutamines to methionine or charged amino acids significantly increase the thermodynamic stability by increasing the m(g) associated with the unfolding isotherm. These effects are hypothesized to arise largely from an increase in the VH/VL association free energy that leads to tighter coupling between domain-domain association and folding. All of the mutants also display a reduced antigen binding affinity. Single and double methionine mutants also displayed significant increases in refolding efficiency of 2.4- to 3-fold over the native scFv, whereas the double alanine/methionine mutants displayed moderate 1.9- to 2.4-fold enhancement. The results suggest that reengineering the VH/VL interface could be useful in improving the stability of single-chain antibodies, as Ala/Met mutations at these conserved positions increase the free energy of folding by 46% while minimally perturbing binding affinity. They also could be useful in improving scFv recovery from inclusion bodies as the mutations increase the refolding efficiency by more than twofold.     

Delivery of antibody-captured proteins into living cells using PTD-fused protein A

Protein transduction domain (PTD)-mediated protein delivery into animal cells is a useful technique for regulating cellular functions. Proteins captured by antibodies were delivered into living cells using an antibody/PTD-fused protein A complex. As a model protein, fluorescent-modified antibodies, captured by their respective primary antibody, were analyzed by fluorescence-activated cell sorting (FACS) which showed that the fluorescent-modified antibodies were directly delivered into cells. Peroxidase, captured by its specific antibody, was also delivered into cells and retained its activity.     

Role of native-state topology in the stabilization of intracellular antibodies

The role played by the geometric position of each amino acid in the folding process of the immunoglobulin (Ig) variable domain is identified and measured through molecular dynamics simulations of models based on the topology of its native state. This measure allows identifying the parts of the protein that, for geometrical reasons, when mutated, would result in relevant protein stability changes. Simulations were performed without considering the covalent disulfide bond present in most of the Ig domains. The results are in good agreement with site-directed mutagenesis experiments on the folding of intracellular antibodies in which the disulfide bond does not form. We also found agreement with data on amino acid conservation in the Ig variable domain sequences. This indicates a new way for a rational approach to the design of intracellular antibodies more resistant to the suppression of the disulfide bond that occurs in the cytoplasm.     

Assembly and misassembly of cystic fibrosis transmembrane conductance regulator: folding defects caused by deletion of F508 occur before and after the calnexin-dependent association of membrane spanning domain (MSD) 1 and MSD2

Cystic fibrosis transmembrane conductance regulator (CFTR) is a polytopic membrane protein that functions as a Cl⁻ channel and consists of two membrane spanning domains (MSDs), two cytosolic nucleotide binding domains (NBDs), and a cytosolic regulatory domain. Cytosolic 70-kDa heat shock protein (Hsp70), and endoplasmic reticulum-localized calnexin are chaperones that facilitate CFTR biogenesis. Hsp70 functions in both the cotranslational folding and posttranslational degradation of CFTR. Yet, the mechanism for calnexin action in folding and quality control of CFTR is not clear. Investigation of this question revealed that calnexin is not essential for CFTR or CFTRΔF508 degradation. We identified a dependence on calnexin for proper assembly of CFTR's membrane spanning domains. Interestingly, efficient folding of NBD2 was also found to be dependent upon calnexin binding to CFTR. Furthermore, we identified folding defects caused by deletion of F508 that occurred before and after the calnexin-dependent association of MSD1 and MSD2. Early folding defects are evident upon translation of the NBD1 and R-domain and are sensed by the RMA-1 ubiquitin ligase complex.     

The Folding Pathway of the Antibody VL Domain

Antibodies are modular proteins consisting of domains that exhibit a β-sandwich structure, the so-called immunoglobulin fold. Despite structural similarity, differences in folding and stability exist between different domains. In particular, the variable domain of the light chain V_L is unusual as it is associated with misfolding diseases, including the pathologic assembly of the protein into fibrillar structures. Here, we have analysed the folding pathway of a V_L domain with a view to determine features that may influence the relationship between productive folding and fibril formation. The V_L domain from MAK33 (murine monoclonal antibody of the subtype κ/IgG1) has not previously been associated with fibrillisation but is shown here to be capable of forming fibrils. The folding pathway of this V_L domain is complex, involving two intermediates in different pathways. An obligatory early molten globule-like intermediate with secondary structure but only loose tertiary interactions is inferred. The native state can then be formed directly from this intermediate in a phase that can be accelerated by the addition of prolyl isomerases. However, an alternative pathway involving a second, more native-like intermediate is also significantly populated. Thus, the protein can reach the native state via two distinct folding pathways. Comparisons to the folding pathways of other antibody domains reveal similarities in the folding pathways; however, in detail, the folding of the V_L domain is striking, with two intermediates populated on different branches of the folding pathway, one of which could provide an entry point for molecules diverted into the amyloid pathway.     

Domain unfolding of monoclonal antibody fragments revealed by non-reducing SDS-PAGE

Monoclonal antibodies and derived fragments are used extensively both experimentally and therapeutically. Thorough characterization of such antibodies is necessary and includes assessment of their thermal and storage stabilities. Thus, assessment of the underlying conformational stabilities of the antibodies is also important. We recently documented that non-reducing SDS-PAGE can be used to assess both monoclonal and polyclonal IgG domain thermal unfolding in SDS. Utilizing this same h2E2 anti-cocaine mAb, in this study we generated and analyzed various mAb antibody fragments to delineate the structural domains of the antibody responsible for the observed discrete bands following various heating protocols and analysis by non-reducing SDS-PAGE. Previously, these domain unfolding transitions and gel bands were hypothesized to stem from known mAb structural domains based on the relative thermal stability of those CH2, CH3, and Fab domains in the absence of SDS, as measured by differential scanning calorimetry. In this study, we generated and analyzed F(ab’)₂, Fab, and Fc fragments, as well as a mAb consisting of only heavy chains, and examined the thermally induced domain unfolding in each of these fragments by non-reducing SDS-PAGE. The results were interpreted and integrated to generate an improved model of thermal unfolding for the mAb IgG in SDS. These results and the model presented should be generally applicable to many monoclonal and polyclonal antibodies and allow novel comparisons of conformational stabilities between chemically or genetically modified versions of a given antibody. Such modified antibodies and antibody drug conjugates are commonly utilized and important for experimental and therapeutic applications.     

Structural information about organized cholesterol domains from specific antibody recognition

Cholesterol-rich domains have been observed to exist in cell membranes under physiological and pathological conditions. Their compositions and the microenvironment of their formation vary over a wide range. Very little information is however available on the molecular structure and organization of these domains. The techniques available to provide such structural information are reviewed here first. The possibility of using tailor-made antibodies as reporters of molecular organization in membranes is then considered. The concept of antibodies recognizing molecular organization rather than single molecular epitopes is established, reviewing the existing works on antibody and protein recognition of crystalline molecular arrays. The information that such antibodies could provide in cells is finally examined together with a proof of application.     

Assembly and function of integrin receptors is dependent on opposing alpha and beta cytoplasmic domains.

The membrane proximal regions of integrin alpha and beta subunits are highly conserved in evolution. In particular, all integrin alpha subunits share the KXGFFKR sequence at the beginning of their cytoplasmic domains. Previous work has shown that this domain is important in integrin receptor assembly. Using chimeric integrin alpha and beta subunits, we show that the native cytoplasmic domains of both subunits must be present for efficient assembly. Most strikingly, chimeric alpha 1 and beta 1 subunits with reciprocally swapped intracellular domains dimerize selectively into collagen IV receptors expressed at high levels on the surface. However, these receptors, which bind ligand efficiently, are deficient in a variety of post-ligand binding events, including cytoskeletal association and induction of tyrosine phosphorylation. Furthermore, deletion of the distal alpha cytoplasmic domain in the swapped heterodimers leads to ligand-independent focal contact localization, which also occurs in wild-type subunits when the distal cytoplasmic domain is deleted. These results show that proper integrin assembly requires opposed alpha and beta cytoplasmic domains, and this opposition prevents ligand-independent focal contact localization. Our working hypothesis is that these two domains may associate during receptor assembly and provide the mechanism for integrin receptor latency.     

Method for generation of human hyperdiversified antibody fragment library

The selection of antibody fragments from libraries using in vitro screening technologies has proven to be a very good alternative to the classical hybridoma technology, and has overcome the laborious process of antibody humanization. However, the complexity of the library is critical in the probability of being able to directly isolate a high affinity antibody specific to a target. We report a method to make hyperdiversified antibody fragment libraries, based on human immunoglobulin variable genes mimicking the somatic hypermutation process. This mutagenesis technology, MutaGen, was used for the first time on the entire variable domain (frameworks and CDRs) of large repertoires of human variable antibody domains. Our MutaGen process uses low-fidelity human polymerases, known as mutases, suggested to be involved in the somatic hypermutation process of immunoglobulin genes. Depending on the mutases used, we generated complementary mutation patterns with randomly distributed mutations. The libraries were generated with an average of 1.8 mutations per 100 amino acids. The hyperdiversified antibody fragment libraries constructed with our process should enable the selection of antibody fragments specific to virtually any target.     

Design and characterization of a plasmid vector system capable of rapid generation of antibodies of multiple isotypes and specificities

We present a detailed method for constructing a vector system composed of plasmids encoding immunoglobulin genes that have been constructed in such a fashion so as to allow for the generation and secretion of antibodies of multiple specificities and isotypes via a rapid and easy cloning-and-ligation scheme. Restriction sites within each plasmid allow for the removal of variable domains, constant domains, leader sequences, or the entire immunoglobulin gene. Degenerate primers are used to clone variable regions from hybridoma cDNAs, allowing for the creation of antibodies with varying binding specificities. Sequence-specific primers are used to clone antibody constant domains, allowing for the creation of antibodies of multiple isotypes from a variety of lineages. A high-efficiency leader sequence has been inserted into the start of each gene to improve secretion. Antibodies constructed through this system are efficiently secreted by transfected cells, and are fully functional in antigen-specific binding assays.     

The crystal structure of the fab fragment of the monoclonal antibody MAK33. Implications for folding and interaction with the chaperone bip

The Fab fragment of the murine monoclonal antibody, MAK33, directed against human creatine kinase of the muscle-type, was crystallized and the three-dimensional structure was determined to 2.9 A. The antigen-binding surface of MAK33 shows a convex overall shape typical for immunoglobulins binding large antigens. The structure allows us to analyze the environment of cis-prolyl-peptide bonds whose isomerization is of key importance in the folding process. These residues seem to be involved with not only domain stability but also seem to play a role in the association of heavy and light chains, reinforcing the importance of beta-strand recognition in antibody assembly. The structure also allows the localization of segments of primary sequence postulated to represent binding sites for the ER-specific chaperone BiP within the context of the entire Fab fragment. These sequences are found primarily in beta-strands that are necessary for interactions between the individual domains.     

Domains b' and a' of protein disulfide isomerase fulfill the minimum requirement for function as a subunit of prolyl 4-hydroxylase. The N-terminal domains a and b enhances this function and can be substituted in part by those of ERp57

Protein disulfide isomerase (PDI) is a modular polypeptide consisting of four domains, a, b, b', and a', plus an acidic C-terminal extension, c. PDI carries out multiple functions, acting as the beta subunit in the animal prolyl 4-hydroxylases and in the microsomal triglyceride transfer protein and independently acting as a protein folding catalyst. We report here that the minimum sequence requirement for the assembly of an active prolyl 4-hydroxylase alpha(2)beta(2) tetramer in insect cell coexpression experiments is fulfilled by the PDI domain construct b'a' but that the sequential addition of the b and a domains greatly increases the level of enzyme activity obtained. In the assembly of active prolyl 4-hydroxylase tetramers, the a and b domains of PDI, but not b' and a', can in part be substituted by the corresponding domains of ERp57, a PDI isoform that functions naturally in association with the lectins calnexin and calreticulin. The a' domain of PDI could not be substituted by the PDI a domain, suggesting that both b' and a' domains contain regions critical for prolyl 4-hydroxylase assembly. All PDI domain constructs and PDI/ERp57 hybrids that contain the b' domain can bind the 14-amino acid peptide Delta-somatostatin, as measured by cross-linking; however, binding of the misfolded protein "scrambled" RNase required the addition of domains ab or a' of PDI. The human prolyl 4-hydroxylase alpha subunit has at least two isoforms, alpha(I) and alpha(II), which form with the PDI polypeptide the (alpha(I))(2)beta(2) and (alpha(II))(2)beta(2) tetramers. We report here that all the PDI domain constructs and PDI/ERp57 hybrid polypeptides tested were more effectively associated with the alpha(II) subunit than the alpha(I) subunit.