Page 114 - ITPS-7-2

INNOSC Theranostics and
Pharmacological Sciences PI3K-α inhibitors for cancer immunotherapy

through various scientific approaches. Furthermore, PI3K/AKT/mTOR. Drug discovery and development
5
various essential metabolic pathways and key factors serve represent a lengthy and costly process, relying heavily
as a pool of associated therapeutic targets for anticancer on in vitro assays, animal models, and clinical trials for
therapy development. 30 reliable testing of drug molecules. Hence, our objective is
The drug discovery process is known for its time- to contribute to the discovery of a novel PI3K-α inhibitory
28
consuming nature and high costs. However, advancements compound through a rational drug design approach with
potential immunomodulatory, immunobiological, and
in molecular biology, genomics, and computational clinical implications. We aim to explore a series of clinically
technologies have accelerated the understanding of cancer approved selective PI3K-α inhibitors with excellent
development, leading to the discovery of novel biomarkers, pharmacokinetic properties obtained from the public web-
targeted therapies, and immunotherapies. These accessible molecular recognition database, BindingDB
advancements have significantly improved cancer diagnosis (http://www.bindingdb.org). 34,35 Rational drug design holds
and treatment outcomes. Three-dimensional quantitative the potential to discover novel drugs or drug combinations
structure-activity relationship (3D-QSAR) in combination and repurpose existing drugs for new indications. This
with molecular dynamics simulations is a hybrid in silico work seeks to introduce innovative methods for discovering
approach for the design and synthesis of drugs and is potential drug molecules through molecular docking and
instrumental in identifying new compounds with superior 3D-QSAR methods, coupled with robust rational drug
activity. 31,32 The combination of 3D-QSAR and molecular design techniques focused on selected PI3K-α inhibitors.
docking approaches continues to demonstrate how in silico
pharmacological systems can work together to produce 2. Methods
novel therapeutics using structural data. The hybrid offers
a more comprehensive insight into the structure-activity 2.1. Dataset
relationship of a compound (ligand) and its interaction The dataset, comprised ligands, was retrieved from
33
with the target protein. This understanding enables drug BindingDB. These cogeneric ligands were represented in
designers to modify the compound’s structure to enhance its three-dimensional (3D) coordinate structures, all sharing
activity, selectivity, and binding affinity, ultimately leading to common orientation codes. Initially, this exploratory
a rational design of more potent and effective drugs. dataset contained 3994 rows of different inhibitory
molecules and 56 columns, providing information on the
In the present study, the drug discovery process using
PI3K-α inhibitors involved ligand preparation based on inhibitory molecules targeting the phosphatidylinositol-3
kinase regulatory subunit alpha in Homo sapiens (human).
predefined descriptors by assessing the drug-like properties Furthermore, the dataset contained some important
of the ligands and assessing their binding affinity to the column attributes, such as the half-maximal inhibition
target through molecular docking. In addition, the study concentration (IC ), BindingDB ligand name, UniProt
considered pharmacophore screening with the least energy (SwissProt) primary and secondary IDs of the target chain,
50
core to prioritize the binding of ligands to the active site molecule ID, and ROMol information. The UniProt serves
of PI3K-α, predicting the most energetically favorable as a focal point for collecting functional information
conformations and estimating the binding affinities. on proteins with precise, dependable, and extensive
Moreover, the study included the prediction of absorption, annotations. It integrates biologically significant data
33
distribution, metabolism, excretion, and toxicity (ADMET) obtained from selected resources and the manual curation
profiles. These approaches aided in the rational design and of protein features, such as functional domains and
optimization of PI3K-α inhibitors, contributing to the active sites, amino acid variations, ligand binding sites,
development of potential therapeutic agents for cancer and and post-translational modifications (PTMs). UniProt
other diseases.
records provide mechanistic insights into disease-drug
Current preclinical and clinical evidence suggests relationships. On the other hand, ROMol (the Read-Only
36
that inhibitors targeting the PI3K/AKT/mTOR pathway molecule) is a representation of a molecule or a chemical
are utilized in combination with other anticancer structure that is strictly read-only. It is an object within
therapies to combat resistance in cancer cells. Multiple RDKit, an open-source cheminformatics library written
ongoing clinical studies are investigating this approach. in C++ with Python bindings, enabling operations with
However, most targeted anticancer therapies, as well as chemical structures and data.
cytotoxic and radiation therapies, are complicated by
secondary resistance in cancer cells. Resistance is an 2.2. Data preprocessing
intricate phenomenon involving numerous mechanisms, Data preprocessing is a crucial step in preparing and
comprising the activation of signaling pathways such as transforming datasets to improve raw data quality,

Volume 7 Issue 2 (2024) 4 doi: 10.36922/itps.2340