Colorectal cancer remains a major cause of cancer mortality worldwide, creating an ongoing need for novel therapeutic agents. This study explored phytochemicals from Artemisia annua for potential anti-colorectal cancer activity using an integrated computational pipeline. The authors focused on identifying compounds that target key molecular drivers implicated in colorectal cancer, with particular attention to the serine/threonine kinase AKT1.
The investigation used a multi-step in-silico approach combining gene expression analysis, ADMET/toxicity screening, network pharmacology, molecular docking, density functional theory (DFT), molecular dynamics (MD) simulation, and post-simulation trajectory analyses. These methods were applied to a library of phytochemicals from Artemisia annua to prioritize candidates based on predicted safety, binding affinity to AKT1, conformational stability, and interaction consistency across simulations.
Initial toxicity screening identified 13 phytochemicals predicted to be non-toxic. Subsequent ADME analyses indicated that the study’s active compounds exhibited favorable drug-likeness profiles according to the reported in-silico assessments. The abstract reports these results qualitatively; detailed ADME parameters and thresholds were not presented in the abstract and therefore are not reported here.
Molecular docking ranked several phytochemicals by predicted binding affinity to AKT1. Notable docking scores reported in the abstract include:
For context, two control compounds were used: 5-fluorouracil (5fu) with a reported docking score of −5 kcal/mol and the AKT inhibitor capivasertib with −7.6 kcal/mol. These comparisons indicate that several Artemisia-derived phytochemicals, including isobonducellin, had predicted docking affinities to AKT1 equal to or better than the control AKT inhibitor and substantially better than 5fu by docking score.
The authors conducted 200-nanosecond MD simulations and analyzed multiple trajectory metrics to assess complex stability. The isobonducellin–AKT1 complex displayed a stable conformational profile with the following reported mean values (± SD):
According to the authors, these parameters indicated that isobonducellin produced a more stable complex with AKT1 compared to the other ligands tested and to capivasertib during the simulated time window.
Additional in-silico analyses reported in the abstract included density functional theory (DFT) calculations, principal component analysis (PCA), dynamic cross-correlation matrix (DCCM) analysis, and MM-GBSA binding free energy estimations. These complementary methods reportedly supported isobonducellin (CID: 10423880) as a strong AKT1-targeting candidate. The abstract notes that MM-GBSA suggested slightly better binding for another ligand, but the authors prioritized isobonducellin because of its superior dynamic stability and consistent interaction profile across simulations.
The computational results propose that isobonducellin from Artemisia annua could act as an AKT1-targeting therapeutic candidate for colorectal cancer, potentially modulating multiple signaling pathways relevant to tumor biology. However, the study is strictly in-silico. The abstract explicitly states that further experimental validation is required, including cancer cell-line assays and animal-model testing, to corroborate predicted binding, biological activity, pharmacokinetics, and safety in biological systems.
Limitations inherent to the reported work include reliance on predicted docking scores and simulation metrics rather than empirical biochemical or cellular activity assays. Details beyond summary statistics—for example, exact ADMET parameter values, specific network pharmacology pathway maps, or full MM-GBSA energy components—were not reported in the abstract and thus are not reproduced here.
Using an integrated computational strategy, the authors identified isobonducellin as a promising phytochemical from Artemisia annua with predicted binding and stable interaction with AKT1, supported by molecular docking, MD simulation, DFT, PCA, DCCM, and MM-GBSA analyses. Docking scores for isobonducellin and other lead compounds compared favorably to control drugs in the computational models. The study concludes that isobonducellin merits further experimental investigation in cell-based and in vivo models to validate its anti-colorectal cancer potential.
Study citation and metadata: Md Maruf Khan et al., Comput Biol Med, 2026. PMID: 42431013. DOI: 10.1016/j.compbiomed.2026.111854. The authors declared no competing interests.