2.7. Checkerboard assay

The two-dimensional checkerboard microdilution assay was used to investigate the interactions between morusin/kuwanon G and antibiotics according to previously described protocols (Aelenei et al., 2019a, Hendry et al., 2009) with minor changes. In brief, in sterile 96-well microtiter plates containing Mueller-Hinton broth, the antibiotic (20 μL) was serially diluted horizontally while the prenylated compound (10 μL) was serially diluted vertically. Then, each well was inoculated with the bacterial suspension (1.5 × 10⁸ CFU/mL, 30 μL), the final volume in each well being 200 µL. The concentrations of antibacterial agents (antibiotic or prenylated compound) in the wells varied from 4 × MIC to 1/512 × MIC. Positive (bacterial inoculum and Mueller-Hinton broth) and negative (prenylated compound or antibiotic and Mueller-Hinton broth) control wells were also prepared. The data analysis was performed using the fractional inhibitory concentration index (FICI) (Mulyaningsih et al., 2010, Zuo et al., 2018), isobolograms (van Vuuren and Viljoen, 2011) and response surface approach using the ΔE model (Segatore et al., 2012). The FICI model is based on Loewe additivity theory whereas the ΔE model bases on Bliss independence theory (Meletiadis et al., 2005, Segatore et al., 2012). FICI was calculated for the first wells without bacterial growth neighboring the wells with bacterial growth as follows: FICI = FIC_antibiotic + FIC_{prenylated compound} = MIC_{antibiotic in combination} / MIC_{antibiotic alone} + MIC_{prenylated compound in combination} / MIC_{prenylated compound alone} (Otto et al., 2019, Zuo et al., 2018). The type of interaction between morusin/kuwanon G and antibiotics was defined on the basis of FICI value: synergy (the final effect greater than the sum of the individual effects) when FICI ≤ 0.5, addition (the final effect equal to the sum of the individual effects) when 0.5 < FICI ≤ 1.0, indifference (no interaction) when 1.0 < FICI ≤ 4, antagonism (the final effect less than the sum of the individual effects) when FICI > 4 (Bassolé and Juliani, 2012, Mulyaningsih et al., 2010, van Vuuren and Viljoen, 2011). In isobolograms (graphs representing the combination effects), the ratio points falling below the 0.5 : 0.5, 1 : 1 and 4 : 4 lines indicate synergy, addition and indifference, respectively whereas those falling above the 4 : 4 line express antagonism (Caesar and Cech, 2019, van Vuuren and Viljoen, 2011). In the response surface approach, the type of interaction is defined by the difference (ΔE) between the predicted (E_predicted) and measured (E_measured) percentages of bacterial growth, with E_predicted being the product of the experimental percentages of bacterial growth for each component of the combination when acting alone (E_predicted = E_{component 1 alone} × E_{component 2 alone}). E_predicted was calculated for all the combinations tested experimentally. The concentrations of the two antibacterial agents and ΔE values were further plotted three-dimensionally, with synergy and antagonism being expressed by the points of the response surface (ΔE) above and below zero, respectively. To characterize the whole interaction surface, the sum of all statistically significant synergistic (ΣSYN) and antagonistic (ΣANT) interactions were calculated; values lower than 100%, between 100% and 200% and higher than 200% indicate weak, moderate and strong interactions, respectively (Meletiadis et al., 2005, Segatore et al., 2012).