Abstract
Herbal products are widely used in healthcare due to their therapeutic potential. However, concerns persist among medicinal chemists and regulatory toxicologists regarding their safety and potential toxicity. Botanical products used in herbal formulations requires rigorous risk assessments. In this study, we applied in silico toxicity testing using natural flavouring complexes (NFCs) approach and Toxtree® v3.1.0 software to assess toxicological risk associated with Curtisia dentata, a widely used medicinal plant for treating diseases. We first conducted experimental analysis of the hexane crude extract of C. dentata using gas chromatography-mass spectrometry (GC-MS) to identify its chemical constituents. A total of 124 compounds were identified from GC-MS analysis and verified using reputable databases, including PubChem, Human Metabolome Database (HMDB), National Institute of Standards and Technology (NIST) database, and scientific literature. The Cramer's toxicity class of the identified components was determined based on their congeneric groups. The identified constituents were then subjected to in silico toxicity evaluation using Toxtree® v3.1.0. Certain constituents, including aniline, 2,6-dimethoxybenzoquinone, and 1-iodododecane triggered structural alerts for genotoxic carcinogenicity and mutagenicity, while phthalic acid, 4-chloro-3-methylphenyl undecyl ester exhibited alerts for non-genotoxic carcinogenicity. While C. dentata has undoubted proven pharmacological effects, further studies should focus on confirming the in silico predictions through targeted in vitro and in vivo studies, particularly for constituents that fired multiple structural alerts.