Implications in substrate recognition, catalysis and drug-design Structural basis for neutralization of ribosome-inactivating proteins by their monoclonal antibodies

Abstract

Cryptic elements of proteins are not visible in unbound structures but become apparent in ligand/substrate-bound structures. These binary states of cryptic elements in protein structures can impact important aspects of proteins such as substrate recognition, catalytic mechanism and are also utilized for drug-design against difficult protein targets. The thesis explores distinct facets of cryptic elements in different proteins through a combination of experimental and computational approaches. More specifically, while addressing a problem specific to a given protein, the work presented in the thesis shows how cryptic elements of that protein assist in finding a solution. Targeting cryptic sites for drug-design provides an attractive option for proteins not tractable by classical binding sites. However, owing to their hidden nature, it is difficult to identify cryptic sites. Use of small glycols as probe molecules to identify cryptic sites in proteins is demonstrated through crystallography experiments, molecular dynamics simulations, protein dataset construction and analysis. The study suggests the use of small glycols in both experimental as well as computational methods to identify cryptic sites in proteins, thus facilitating drug-design for undruggable and/or difficult protein targets and expanding the druggable proteome. Abrin is an extremely cytotoxic ribosome inactivating protein. Abrin A-chain (ABA) inactivates eukaryotic ribosomes leading to cell death. The crystal structures of apo-ABA and ABA complexed with substrate analogs were determined in order to study ABA’s substrate binding and catalytic mechanism. The analyses of ABA-substrate analog crystal structures show the presence of “two binding pockets" feature not apparent in unbound-ABA structure, making it a cryptic feature of ABA and sheds light on aspects of substrate recognition by ABA. Catalytic water was clearly located in ABA-Adenine crystal structure but was not observed in unbound-ABA crystal structure, making it another cryptic feature of ABA structure and was used to explain the catalytic mechanism of ABA. A monoclonal antibody (mAb) D6F10 is known to neutralize the toxicity of abrin. The structures of ABA-Adenine complex and ribosome were used to construct ABA-Ribosome complex. A 3D homology model of variable region (Fv) of mAb D6F10 was generated and was docked with apo-ABA structure to obtain computational model of ABA-D6F10 Fv complex. Structural superposition of ABA common to ABA-D6F10 Fv and ABA-Ribosome complexes reveals steric hindrance as the primary mechanism by which mAb D6F10 neutralizes abrin. Furthermore, crystal structure of fragment crystallizable (Fc) region of mAb D6F10 at 1.95 Å resolution was determined. Crystal structure based glycan analysis of mAb D6F10 Fc confirmed the presence of speculated terminal galactose on glycans of mAb D6F10 Fc fragment thereby pointing to the proposed model of abrin neutralization at 1:100 molar ratio of abrin:D6F10. Rv0731c, a protein from human pathogen Mycobacterium tuberculosis, is annotated as putative S-adenosyl-L-methionine(SAM)-dependent methyltransferase. A high resolution crystal structure of Rv0731c at 1.63 Å was determined and bioinformatics was used to predict its biological function. The structural analysis of Rv0731c revealed a conserved carboxy-terminal SAM binding domain and a methyl-accepting substrate binding domain similar to that of the leucine carboxyl methyltransferases from human and yeast. In apo-Rv0731c, the substrate binding site is occluded by a flexible loop-helix flap segment and substrate binding may displace the flap revealing the site thereby making it a cryptic feature of Rv0731c and sheds light on the features of substrate recognition by Rv0731c