Computational Ranking of Yerba Mate Small Molecules Based on Their Predicted Contribution to Antibacterial Activity against Methicillin-Resistant Staphylococcus aureus

Abstract

The aqueous extract of yerba mate, a South American tea beverage made from Ilex paraguariensis leaves, has demonstrated bactericidal and inhibitory activity against bacterial pathogens, including methicillin-resistant Staphylococcus aureus (MRSA). The gas chromatography-mass spectrometry (GC-MS) analysis of two unique fractions of yerba mate aqueous extract revealed 8 identifiable small molecules in those fractions with antimicrobial activity. For a more comprehensive analysis, a data analysis pipeline was assembled to prioritize compounds for antimicrobial testing against both MRSA and methicillin-sensitive S. aureus using forty-two unique fractions of the tea extract that were generated in duplicate, assayed for activity, and analyzed with GC-MS. As validation of our automated analysis, we checked our predicted active compounds for activity in literature references and used authentic standards to test for antimicrobial activity. 3,4-dihydroxybenzaldehyde showed the most antibacterial activity against MRSA at low concentrations in our bioassays. In addition, quinic acid and quercetin were identified using random forests analysis and 5-hydroxy pipecolic acid was identified using linear discriminant analysis. We also generated a ranked list of unidentified compounds that may contribute to the antimicrobial activity of yerba mate against MRSA. Here we utilized GC-MS data to implement an automated analysis that resulted in a ranked list of compounds that likely contribute to the antimicrobial activity of aqueous yerba mate extract against MRSA.

Fig 1. Overlay of initial yerba mate extract fraction chromatograms.

A) The black chromatogram corresponds to a yerba mate extract fraction that demonstrated antibacterial activity against methicillin-resistant Staphylococcus aureus (MRSA); the red chromatogram corresponds to a yerba mate fraction that had no antibacterial activity against MRSA. B) Retention times of identified compounds and quantification in sorbitol equivalents were reported.

Fig 2. Growth of methicillin-sensitive (SA) and methicillin-resistant Staphylococcus aureus (MRSA) in the presence of single or multiple pure compounds at proportions approximated from GC-MS quantification.

Growth with compounds alone or together was compared to the positive growth control (no chemical added) to determine inhibitory activity. Statistically significant differences greater (*) or less (**) than control are marked by asterisks. Concentrations follow the GC-MS quantification values in Fig 1, but in μg/ml. Growth of A. SA 27708, B. MRSA 35591, and C. MRSA 35593 are reported at 24 h.

Fig 3. Growth of methicillin-sensitive (SA) and methicillin-resistant Staphylococcus aureus MRSA) in the presence of pure compounds.

At concentrations of 10 μg/ml (chemical_1), 20 μg/ml (chemical_2) and 100 μg/ml (chemical_3), growth with compounds was compared to the positive growth control (no chemical added) to determine inhibitory activity. Statistically significant differences greater (*) or less (**) than control are marked by asterisks. Growth of A. SA 113, B. SA 27708, C. MRSA 35591, and D. MRSA 35593 are reported at 48 h. SA113 had a significant block by treatment interaction, so no conclusions can be drawn from it.