Improvements in the Paradigm of Modern Cancer Treatments through Integration of Different Computational Approaches

Abstract

Pancreatic cancer is a highly lethal malignancy with a mortality rate of over

89% within the first year of diagnosis. The integration of modern computational ap- proaches can improve its treatment by identifying specific therapeutic targets and

developing advanced therapies with minimal toxicity profiles. Therefore, aim of this study is to identify cancer-specific biomarkers, therapeutic targets, and associated pathways involved in pancreatic cancer progression, as well as designing therapeutics with optimal efficacy. The study is broken down into several steps to achieve these

objectives. First, the transcriptomic data analysis was performed for the identifica- tion of pancreatic cancer-specific biomarkers, and therapeutic targets. RNA-seq and

microarray datasets of pancreatic cancer tissues and adjacent healthy tissues were ob- tained from public repositories such as European Bioinformatics Institute (EBI) and

Gene Expression Omnibus (GEO) databases. Differential expression (DE) analysis

was performed to identify significant differentially expressed genes (DEGs) in pancre- atic cancer cells compared to the normal cells. Gene co-expression network analysis

was executed to identify the co-expressed hub genes, which are strongly associated

with pancreatic cancer. The key underlying pathways were obtained from the enrich- ment analysis of hub genes. The significant pathways, hub genes, and their expression

profile were validated against the Cancer Genome Atlas (TCGA) data. Important hub genes identified included: integrins, MET, LAMB1, VEGFA, PTK, PAK, Rac, and TGFβ 1, etc. Enrichment analysis characterizes the involvement of hub genes in multiple pathways. The interaction of overexpressed surface proteins of these

pathways with extracellular molecules initiates multiple signaling cascades. Includ- ing stress fiber and lamellipodia formation, PI3K-Akt, MAPK, JAK/STAT, and Wnt

signaling pathways. The overexpressed surface receptors (SLC2A1, MET, IL1RAP, NPR3, GABRP, SLC6A6, and TMPRSS4) on pancreatic cancer cells surface were considered for further analysis.

Abstract

xxvi Oncolytic virus therapy is a type of immunotherapy that is currently under deliberation by the researchers for multiple cancer types in various clinical trials. The oncolytic virus not only kills cancer cells but also activates anti-cancer immune

response. Therefore, it is preferred over other therapies to deal with aggressive pancre- atic cancer. The efficacy of therapy primarily depends on how effectively the oncolytic

virus enters and infects the cancer cell. Cell surface receptors and their interactions with virus coat proteins is a crucial step for oncolytic viruses entry and a pivotal determinant. The L5 proteins of the viral coat are the initial points of contact with host cell surface receptors. Therefore, the objective of this study is to analyze the interaction profile of L5 protein of oncolytic adenovirus with overexpressed surface receptors of pancreatic cancer. The L5 proteins of three adenovirus serotypes HAdV2, HAdV5, and HAdV3 were utilized in this study. Overexpressed pancreatic cancer receptors include SLC2A1, MET, IL1RAP, NPR3, GABRP, SLC6A6, and TMPRSS4.

Protein structures of viral and cancer cell protein were interacted using the HAD- DOCK server. Binding affinity and interaction profile of viral proteins against all the

receptors were analyzed. Results suggest that HADV3 L5 protein shows better inter- action compared to HAdV2 and HAdV5 by elucidating high binding affinity with 4

receptors (NPR3, GABRP, SLC6A6, and TMPRSS4). The current study proposed that HAdV5 or HAdV2 virus pseudotyped with the L5 protein of HAdV3 can effectively interact with receptors of pancreatic cancer cells. Furthermore, affinity maturation

was performed to enhance the interaction profile of HAdV3 L5 protein with the re- maining three pancreatic cancer receptors. Specifically, mutations were introduced

into the important interacting residues of HAdV3 fiber knob. The mCSM-PPI2 server was utilized to examine the induced mutation on residue Asn186, Tyr mutant HAdV3 L5 complexes exhibit an increase in binding affinity for all receptors. Prodigy was utilized to validate the increase in affinities for Tyr mutated HAdV3 L5-receptors complexes. Presently, affinity maturation analysis was conducted for only Asn186, which, maybe showing successful results in strengthening the interaction with the

Abstract

xxvii targeted receptors, while, simultaneously maintaining or improving binding affinity to other receptors. All the receptors show promising interactions with mutant HAdV3 L5 accept MET. In addition to analyzing the interaction profile of oncolytic adenovirus, this research also focuses on developing inhibitors against overexpressed c-MET receptor. The MET gene encodes for c-MET, overexpression of MET in pancreatic cancer was reported in this study and multiple previous studies. The c-MET overexpression in

cell is associated with the activation of multiple cancer related pathways like PI3K- Akt and K-Ras signaling pathway, focal adhesion, and central carbon metabolism in

cancer pathways. Therefore, c-MET is considered as an important therapeutic target for pancreatic cancer treatment. The structure of c-MET receptor with high resolution was downloaded from Protein Data Bank (PDB). Binding pocket was predicted on the bases of co-crystallized ligand in pdb structure. Fragment based drug designing was performed to generate novel hybrid compounds from existing c-MET receptor drug molecules. The drug molecules and generated hybrid compounds were docked to compare their interactions in the c-MET binding pocket. Hybrid compounds show better docking score and interaction profile compared to existing drug compounds. Toxicity profile (ADMET analysis) of hybrid compounds was determined, and 3 lead compounds with better ADMET values were selected. To validate the interactions

stability of lead compounds with c-MET binding pocket residues, 200ns MD sim- ulation of ligand-protein complexes was performed. MD simulation results exhibit

stability of interactions, which demonstrates the significance of these compounds as potential therapeutics. Conclusively in this study, computational approaches were utilized for devising therapeutic options with least side effects, targeting pancreatic cancer specific biomarkers.