In our efforts to discover novel sources of compounds able to counteract the MDR linked to fungal ABC and MFS efflux pumps, a panel of 137 extracts from mostly native plants from Argentina, was screened.
The results obtained showed that 15 extracts rendered strains C. albicans 1114 and AD/CaMDR1 sensitive to FCZ by inhibiting the efflux linked to Mdr1 (Table 1). Among the effective extracts, those obtained from A. communis and S. atriplicifolium had the highest activity for chemosensitizing strains to FCZ, and together with the P. alopecuroides extract, also increased the antifungal effect of the azole in the Cdr1-overexpressing cells, AD/CaCDR1 and C. glabrata 109 (Table 1). According to these findings, A. communis, P. alopecuroides and S. atriplicifolium extracts behaved as dual inhibitors. The remaining extracts did not reverse the resistance to FCZ in the Cdr1-expressing yeasts, demonstrating a selective activity towards CaMdr1. These results would explain the lower chemosensitization activity observed for the clinical isolate C. albicans 1114 compared to S. cerevisiae AD/CaMDR1 for A. subfusiformis, B. salicifolia, F. campestris, F. oolepis, L. molleoides, M. dictyocarpa and S. palinacanthum extracts (Table 1), since although the MFS transporter Mdr1 may be targeted in, the ABC pump would be still active in C. albicans 1114, thus decreasing the intracellular concentration of FCZ and therefore evading partially its antifungal action.
The low MECs of 25 µg/mL that caused a reduction in the MICFCZ values of at least 4-fold observed with the clinical isolates of Candida, highlighted A. communis and S. atriplicifolium as promising sources of Mdr1 and Cdr1 inhibitors to augment FCZ activity in treatments for azole resistant candidiasis. None of the 15 active extracts increased the antifungal effect of FCZ in the sensitive C. albicans, C. glabrata and AD1-8u− strains, supporting the specific interference with the efflux function of the transporters.
L. cuneifolia is a hemiparasitic plant that grows in different hosts [49]. The activity of extracts obtained from this species from each of the host trees, L. molleoides, Vachellia sp. and Condalia buxifolia was similar (Table 1), suggesting that the parasitized plant did not influence the pump inhibitory effect.
Betulin, isolated from L. cuneifolia, has been found to strongly inhibit the human ABC transporter P-gp at micromolar concentrations by blocking the efflux of doxorubicin and consequently restoring the sensitivity of leukemia cells to its cytotoxic effect [29]. However, this triterpene did not reduce FCZ resistance in the target yeasts, even at the high concentration of 50 µM. This result suggests that betulin is a selective inhibitor of P-gp, at least with respect to Mdr1 and Cdr1. Although this result correlated with the lack of activity of L. cuneifolia extract against Cdr1, it is not in accordance with the effect of this plant against Mdr1, suggesting that another metabolite with Mdr1 blocking activity is present. The inability of betulin to inhibit the outward transport of FCZ mediated by yeast efflux pumps is in agreement with the observation that diterpenoid esters isolated from Euphorbia spp. were effective at inhibiting P-gp but were not active against C. albicans Mdr1 or Cdr1 [50].
Both A. communis and S. atriplicifolium extracts inhibited Nile Red transport at concentrations ≥ 12.5 µg/mL (Table 2 and Figs. 1 and 2) with FIR values ranging from 1.10 to 3.68 (Table 2). These results confirmed that the enhanced activity of FCZ was due to an increased intracellular accumulation. In these experiments, the Nile Red MFI values were lower for the untreated and treated Candida spp. than for the S. cerevisiae strains (Figs. 1 and 2). At the emission and excitation wavelengths used in this study (488 nm and a 574/26 nm, respectively), Nile Red labels neutral lipids within intracellular lipid droplets [51, 52]. The differences in the MFI values obtained, suggest that S. cerevisiae strains contain more neutral lipids than Candida strains.
Although the significantly (p < 0.05) lower MFI values for untreated AD/CaCDR1 (4,730) and AD/CaMDR1 (13,700) in comparison to AD1-8u− (MFI = 52,900), showed the effectiveness of the pumps at transporting Nile Red, the lower MFI value for AD/CaCDR1 than for AD/CaMDR1 (p < 0.05) indicated that Nile Red is more efficiently effluxed by CaCdr1 than by CaMdr1. Therefore, it is encouraging that A. communis, P. alopecuroides and S. atripicifolium extracts were able to inhibit this proficient ABC-type pump.
The same Nile Red-derived fluorescence in pump-deficient C. albicans, C. glabrata and AD1-8u− cells in the absence or presence of the extracts suggests that the increased FCZ susceptibility in MDR cells is likely achieved by inhibiting Mdr1 and Cdr1.
As far as we are aware, there have been few studies that have screened libraries of plant extracts for MDR reversal activity through interference with yeast transporters [32, 50]. In particular, no information was found in the literature with respect to this property in A. communis and S. atripicifolium extracts. It is important to note that assays were performed with crude plant extracts likely to contain a mixture of compounds. The next step in this research will be to purify the active component(s) from the extracts. These compounds are likely to have a higher pump-inhibitory specific activity.
The A. communis and S. atripicifolium extracts were devoid of toxic effects on mammalian cells. The IC50 values obtained towards PBMC were higher than 20 µg/mL, which is the threshold established by the US National Cancer Institute (NCI) to consider an extract as cytotoxic [26]. In addition, neither extract caused hemolysis of human erythrocytes at concentrations of 100 µg/mL.