Exploration of Drug Science

Special Issue Topic

Bioactive Molecules from Natural Sources

Submission Deadline: August 31, 2025

Guest Editor

Prof. Michio Kurosu E-Mail

Professor, College of Pharmacy, University of Tennessee Health Science Center, USA

About the Special Issue

Natural products constitute an important source for drug discovery to improve human health. At the present time, novel structures of natural products are difficult to isolate from specimens grown under normal conditions. Taking advantage of the genomic, metabolomic, and proteomic databases, the new natural product sciences emerged in the late 20th century. These efforts have resulted in finding novel bioactive natural products. In parallel, robust assay methods to screen natural product libraries and structure-based drug design has proven to be progressable tactics in modern drug discovery programs. The aim of this special issue seeks original research articles and reviews on bioactive natural products, natural product-based medicinal chemistry, chemical syntheses of bioactive natural products, new assay methods to screen natural products, computational chemistry to improve efficacy of natural products, pharmacokinetic and dynamic aspects of drugs of natural product origin, and new strategy to use approved natural product derivatives.

Keywords: Bioactive natural products, assay-guided isolations, genome mining, synthesis, medicinal chemistry, biosynthesis, proteomics, drug targets

Call for Papers

Published Articles

Open Access

Original Article

Revolutionizing diabetes treatment: computational insights into 4-hydroxy isoleucine derivatives and advanced molecular screening for antidiabetic compounds
Aim: This study aimed to computationally identify and optimize 4-hydroxy isoleucine (4HILe) derivatives from fenugreek as multitarget antidiabetic agents against α-glucosidase, α-amylase, and a [...] Read more.

Aim:

This study aimed to computationally identify and optimize 4-hydroxy isoleucine (4HILe) derivatives from fenugreek as multitarget antidiabetic agents against α-glucosidase, α-amylase, and aldose reductase [PDB (protein data bank) IDs: 5NN8, 4GQR, 4QX4].

Methods:

A multi-step computational workflow was employed to identify and optimize 4HILe derivatives as antidiabetic agents. Molecular docking using the Schrödinger Suite screened 23 ligands against three enzyme targets to evaluate binding affinities and interactions. Molecular dynamic (MD) simulations conducted with GROMACS (Groningen machine chemical simulations) over 100 ns assessed conformational stability through RMSD (root mean square deviation) and RMSF (root mean square fluctuation) analysis. Binding free energy calculations [MM-GBSA (molecular mechanics-generalized Born surface area)] and free energy landscape (FEL) studies are performed to validate the thermodynamics of protein-ligand interactions. Additionally, generative AI modeling using LigDream generated 100 novel compounds derived from 4HILe, subsequently validated through docking studies to identify promising inhibitors.

Results:

The study identified 4HILe-4, 2R-3S-4R-4HILe, and 4HILe-Amide-2 as potent derivatives with superior binding affinities [ΔG (Gibbs free energy): −49.3 to −42.3 kcal/mol] compared to co-crystal ligands (−45.3 kcal/mol), as determined by docking and MM-GBSA calculations. MD revealed stable protein-ligand complexes, evidenced by low RMSD values (0.2–0.4 nm) and minimal residue fluctuations (RMSF), confirming their structural integrity. The generative AI approach using LigDream also generated 100 novel 4HILe derivatives, with top candidates exhibiting strong docking scores and key molecular interactions against α-glucosidase, α-amylase, and aldose reductase. Notably, compound 10 (−9.424 kcal/mol), compound 4 (−8.167 kcal/mol), and compound 28 (−13.760 kcal/mol) emerged as promising inhibitors for further investigation.

Conclusions:

The study highlights 4HILe derivatives as promising inhibitors for diabetes-associated enzymes, demonstrating robust binding and dynamic stability. Integrating molecular dynamics, free energy calculations, and AI-driven generative modeling provides a strong framework for accelerating antidiabetic drug discovery. These findings pave the way for experimental validation and the development of next-generation therapeutics targeting insulin resistance and hyperglycemia.

Lakshmi Mounika Kelam ... M. Elizabeth Sobhia

Published: April 10, 2025 Explor Drug Sci. 2025;3:1008104
DOI: https://doi.org/10.37349/eds.2025.1008104

Full Text

PDF

View:190

Download:12

Times Cited: 0

Aim:

This study aimed to computationally identify and optimize 4-hydroxy isoleucine (4HILe) derivatives from fenugreek as multitarget antidiabetic agents against α-glucosidase, α-amylase, and aldose reductase [PDB (protein data bank) IDs: 5NN8, 4GQR, 4QX4].

Methods:

A multi-step computational workflow was employed to identify and optimize 4HILe derivatives as antidiabetic agents. Molecular docking using the Schrödinger Suite screened 23 ligands against three enzyme targets to evaluate binding affinities and interactions. Molecular dynamic (MD) simulations conducted with GROMACS (Groningen machine chemical simulations) over 100 ns assessed conformational stability through RMSD (root mean square deviation) and RMSF (root mean square fluctuation) analysis. Binding free energy calculations [MM-GBSA (molecular mechanics-generalized Born surface area)] and free energy landscape (FEL) studies are performed to validate the thermodynamics of protein-ligand interactions. Additionally, generative AI modeling using LigDream generated 100 novel compounds derived from 4HILe, subsequently validated through docking studies to identify promising inhibitors.

Results:

The study identified 4HILe-4, 2R-3S-4R-4HILe, and 4HILe-Amide-2 as potent derivatives with superior binding affinities [ΔG (Gibbs free energy): −49.3 to −42.3 kcal/mol] compared to co-crystal ligands (−45.3 kcal/mol), as determined by docking and MM-GBSA calculations. MD revealed stable protein-ligand complexes, evidenced by low RMSD values (0.2–0.4 nm) and minimal residue fluctuations (RMSF), confirming their structural integrity. The generative AI approach using LigDream also generated 100 novel 4HILe derivatives, with top candidates exhibiting strong docking scores and key molecular interactions against α-glucosidase, α-amylase, and aldose reductase. Notably, compound 10 (−9.424 kcal/mol), compound 4 (−8.167 kcal/mol), and compound 28 (−13.760 kcal/mol) emerged as promising inhibitors for further investigation.

Conclusions:

The study highlights 4HILe derivatives as promising inhibitors for diabetes-associated enzymes, demonstrating robust binding and dynamic stability. Integrating molecular dynamics, free energy calculations, and AI-driven generative modeling provides a strong framework for accelerating antidiabetic drug discovery. These findings pave the way for experimental validation and the development of next-generation therapeutics targeting insulin resistance and hyperglycemia.

Guest Editor

About the Special Issue

Call for Papers

Published Articles

Aim:

Methods:

Results:

Conclusions:

Aim:

Methods:

Results:

Conclusions: