Fatty-Acid-Dominant Phytochemistry, Divergent Bioactivity: A GC-MS Structure Activity Analysis of Four Medicinal Plant Extracts in Calcium Oxalate Inhibition and Cytotoxicity

Abstract

Background: Calcium oxalate urolithiasis remains one of the most common urological conditions worldwide, and plant extracts are routinely screened against its three crystallization stages  nucleation, aggregation, and growth  largely as isolated case studies rather than as a comparable set. Objective: This study asked whether four medicinal plant extracts that share a fatty-acid-dominant phytochemical matrix  Ammi visnaga, Carthamus tinctorius, Plantago ovata, and Eriobotrya japonica  also share a comparable pattern of calcium oxalate inhibition and toxicity, or whether their shared chemistry carries no predictive value for their biological behavior. Methods: Gas chromatography-mass spectrometry (GC-MS) compound-abundance profiles for the four extracts were reduced to six shared and marker compound classes and compared using Bray-Curtis dissimilarity. Bioactivity dissimilarity was computed from mean nucleation rate across four temperatures, zebrafish embryo viability at 900 µg/mL, and Vero cell half-maximal inhibitory concentration (IC₅₀), each min–max normalized. The two dissimilarity matrices were compared by Spearman rank correlation across the six unique extract pairs, and a thermal sensitivity index (percentage change in nucleation rate from room temperature to 40°C) was calculated for each extract. Results: Carthamus tinctorius was the most effective nucleation inhibitor at room temperature, 20°C, and 37°C (mean rate as low as 0.181 at 20°C) but lost the largest share of that activity by 40°C (+171.5%), while Ammi visnaga showed the smallest thermal loss (+4.8%) despite being the weakest nucleation inhibitor overall. Plantago ovata was the only cytotoxic extract of the four (Vero IC₅₀ = 1.68 µg/mL; zebrafish viability 12.07% at 900 µg/mL), while the remaining three extracts were non-toxic by both assays. Chemical dissimilarity did not track with bioactivity dissimilarity across the six extract pairs (Spearman ρ = −0.03, p = 0.96); the pair with the most similar bulk fatty-acid chemistry (Ammi visnaga–Eriobotrya japonica, Bray-Curtis = 0.415) was not the pair with the most similar bioactivity,