Himani Tandon

So-called “cross-docking” is the prediction of the bound configuration of small-molecule ligands that differ from the cognate ligand of a protein co-crystal structure. This is a much more challenging problem than re-docking the cognate ligand, particularly when the new ligand is structurally dissimilar from prior known ones. We have updated the previously introduced PINC (“PINC Is Not Cognate”) benchmark which introduced the idea of temporal segregation to measure cross-docking performance. The temporal set encompasses 846 future ligands for ten targets based on information from the earliest 25% of X-ray co-crystal structures known for each target. Here, we extend the benchmark to include thirteen targets where the bound poses of 128 macrocyclic ligands are to be predicted based on knowledge from structures of bound non-macrocyclic ligands. Performance was roughly equivalent for both the temporally-split non-macrocyclic ligand set and the macrocycle prediction set. Using standard and fully automatic protocols for the Surflex-Dock and ForceGen methods, across the combined 974 non-macrocyclic and macrocyclic ligands, the top-scoring pose family was correct 68% of the time, with the top-two pose families achieving a 79% success rate. Correct poses among all those predicted were identified 92% of the time. These success rates far exceeded those observed for the alternative methods AutoDock Vina and Gnina on both sets.

Phosphorylation, a fundamental cellular mechanism, intricately regulates protein function and signaling pathways. Our study employs extensive computational analyses on a curated dataset of phosphorylated and unphosphorylated protein structures to explore the multifaceted impact of phosphorylation on protein conformation. Our findings reveal that phosphorylation induces not only local changes at the phosphorylation site but also extensive alterations in distant regions, showcasing its far-reaching influence on protein structure-dynamics. Using Normal Mode Analysis (NMA), we investigate changes in protein flexibility post-phosphorylation, highlighting an enhanced level of structural dynamism. Through in-depth case studies on Polyubiquitin-B and Glycogen Synthase Kinase-3 Beta, we elucidate how phosphorylation at distinct sites leads to variable structural and dynamic modifications, potentially dictating functional outcomes. While phosphorylation largely preserves residue motion correlation, it significantly disrupts low-frequency global modes, presenting a dualistic impact on protein dynamics. We also explore alterations in the total accessible surface area (ASA), emphasizing region-specific changes around phosphorylation sites. This study sheds light on phosphorylation-induced conformational changes, dynamic modulation, and surface accessibility alterations, contributing to a comprehensive understanding of cellular regulation and suggesting promising avenues for therapeutic interventions.

Mycobacterium tuberculosis genome encodes over 80 toxin–antitoxin (TA) systems. While each toxin interacts with its cognate antitoxin, the abundance of TA systems presents an opportunity for potential non-cognate interactions. TA systems mediate manifold interactions to manage pathogenicity and stress response network of the cell and non-cognate interactions may play vital roles as well. To address if non-cognate and heterologous interactions are feasible and to understand the structural basis of their interactions, we have performed comprehensive computational analyses on the available 3D structures and generated structural models of paralogous M. tuberculosis VapBC and MazEF TA systems. For a majority of the TA systems, we show that non-cognate toxin–antitoxin interactions are structurally incompatible except for complexes like VapBC15 and VapBC11, which show similar interfaces and potential for cross-reactivity. For TA systems which have been experimentally shown earlier to disfavor non-cognate interactions, we demonstrate that they are structurally and stereo-chemically incompatible. For selected TA systems, our detailed structural analysis identifies specificity conferring residues. Thus, our work improves the current understanding of TA interfaces and generates a hypothesis based on congenial binding site, geometric complementarity, and chemical nature of interfaces. Overall, our work offers a structure-based explanation for non-cognate toxin-antitoxin interactions in M. tuberculosis.

Toxin-antitoxin (TA) systems are ubiquitously existing addiction modules with essential roles in bacterial persistence and virulence. The genome of Mycobacterium tuberculosis encodes approximately 79 TA systems. Through computational and experimental investigations, we report for the first time that Rv0366c-Rv0367c is a non-canonical PezAT-like toxin-antitoxin system in M. tuberculosis. Homology searches with known PezT homologues revealed that residues implicated in nucleotide, antitoxin-binding and catalysis are conserved in Rv0366c. Unlike canonical PezA antitoxins, the N-terminal of Rv0367c is predicted to adopt the ribbon-helix-helix (RHH) motif for deoxyribonucleic acid (DNA) recognition. Further, the modelled complex predicts that the interactions between PezT and PezA involve conserved residues. We performed a large-scale search in sequences encoded in 101 mycobacterial and 4500 prokaryotic genomes and show that such an atypical PezAT organization is conserved in 20 other mycobacterial organisms and in families of class Actinobacteria. We also demonstrate that overexpression of Rv0366c induces bacteriostasis and this growth defect could be restored upon co-expression of cognate antitoxin, Rv0367c. Further, we also observed that inducible expression of Rv0366c in Mycobacterium smegmatis results in decreased cell-length and enhanced tolerance against a front-line tuberculosis (TB) drug, ethambutol. Taken together, we have identified and functionally characterized a novel non-canonical TA system from M. tuberculosis.

Hepatitis C virus (HCV) is a leading cause of chronic viral hepatitis, but an effective vaccine is still not available to prevent infection. Use of neutralizing antibodies could be a potential therapeutic option. In this study, the presence of anti-HCV antibodies in HCV-infected patients was assessed from 50 patients and the presence of neutralizing antibodies was examined using ‘hepatitis C virus-like particles’. Antibodies from two samples exhibited significant inhibitory activity, suggesting that these may neutralize viral infection. Antigenic determinants generating the neutralizing antibodies from these two samples were delineated by epitope mapping using the core, E1 and E2 regions and a stretch of 45 amino acid peptide (E2C45) derived from the C-terminal region of HCV-E2 protein (aa 634–679) was designed. Results suggest that this hitherto uncharacterized region has the potential to generate neutralizing antibodies against HCV and thus be effective in preventing virus entry into liver cells. Computational analysis of the structure of the modelled peptide (E2C45) suggested high conformational entropy for this region. Furthermore, E2C45 peptide-generated antibodies could block virus entry and monoclonal antibodies generated against this peptide could also significantly reduce virus replication in a cell culture system. It is possible that the inhibition could be partly due to a conformational alteration of the CD81-binding region, preventing virus attachment to liver cells. In conclusion, this work focused on the discovery of a novel epitope at the C terminus of E2 that induces potent neutralizing antibodies in HCV-infected patients.

Peptide‐based antiviral therapeutics has gradually paved their way into mainstream drug discovery research. Experimental determination of peptides' antiviral activity as expressed by their IC ₅₀ values involves a lot of effort. Therefore, we have developed “AVP‐IC ₅₀ Pred,” a regression‐based algorithm to predict the antiviral activity in terms of IC ₅₀ values (μM). A total of 759 non‐redundant peptides from AVPdb and HIPdb were divided into a training/test set having 683 peptides (T ⁶⁸³ ) and a validation set with 76 independent peptides (V ⁷⁶ ) for evaluation. We utilized important peptide sequence features like amino‐acid compositions, binary profile of N8‐C8 residues, physicochemical properties and their hybrids. Four different machine learning techniques (MLTs) namely Support vector machine, Random Forest, Instance‐based classifier, and K‐Star were employed. During 10‐fold cross validation, we achieved maximum Pearson correlation coefficients (PCCs) of 0.66, 0.64, 0.56, 0.55, respectively, for the above MLTs using the best combination of feature sets. All the predictive models also performed well on the independent validation dataset and achieved maximum PCCs of 0.74, 0.68, 0.59, 0.57, respectively, on the best combination of feature sets. The AVP‐IC ₅₀ Pred web server is anticipated to assist the researchers working on antiviral therapeutics by enabling them to computationally screen many compounds and focus experimental validation on the most promising set of peptides, thus reducing cost and time efforts. The server is available at http://crdd.osdd.net/servers/ic50avp . © 2015 Wiley Periodicals, Inc. Biopolymers (Pept Sci) 104: 753–763, 2015.