Work

Antibacterial activity, phytochemical profiling, and toxicity evaluation of green extracts from Garcinia Mangostana pericarp for food antibacterial applications

Year: 2026

Type: article

Abstract: This study investigated the antibacterial activity, phytochemical composition, and safety of Garcinia mangostana pericarp (GMP) extracts obtained using aqueous or ethanolic solvents via maceration–filtration or maceration–centrifugation–filtration. Extracts were tested against a panel of 13 bacterial strains, foodborne pathogens, spoilage bacteria, and food-contamination-associated microorganisms and analysed for phenolic and flavonoid content, gas chromatography–mass spectrometry (GC/MS) profiles, and toxicity to Artemia salina . Ethanolic GMP extract from maceration–filtration (GMPEF) inhibited all bacteria except K. pneumoniae and S. sonnei , showing the lowest MIC values (0.70–2.81 mg/mL), the highest extraction yield of 0.129 ± 0.002 g/100 g DW (w/w), and total phenolic content (238.44 ± 0.79 mg GAE/g DW). The upper fraction of ethanolic extract from maceration–centrifugation–filtration (GMPECA) had the highest flavonoid content (4.98 ± 0.001 mg QE/g DW). The GC/MS identified 33 putative compounds (≥ 90% match to the NIST 14 library), including β-sugars, phenolics, and stigmasterol, linked to antibacterial potency. All extracts were non-toxic to A. salina (LC 50 > 1 mg/mL); aqueous extracts generally showed higher LC 50 values, suggesting a greater safety margin. Principal Component Analysis (PCA) revealed correlations between specific phytochemicals and antibacterial performance, providing a possible chemical basis for selecting extracts. The GMPEF is well-suited in contexts where higher antibacterial activity is desired, and solvent removal is manageable, whereas GMPWCB offers a safer alternative when prioritising a conservative safety profile. While these findings indicate the potential of GMP extracts as sustainable, low-cost antibacterial agents from fruit by-products, further validation in mammalian systems and real food matrices is needed before considering practical applications.

Source: Journal of Food Measurement & Characterization

DOI: 10.1007/s11694-026-04184-2