Previous studies have established the significant roles of DSF-family signals in microbial ecology as well as in intraspecies signaling regulation [14, 17, 22–24, 37]. It was reported that DSF signals could modulate various biological functions including virulence, biofilm formation, antibiotic resistance and persistence through interspecies communication [23, 24, 37]. Additionally, DSF-family signals were also found to play a role in inter-kingdom communication by inhibiting morphological transition of C. albicans[14, 17, 22]. The results from this study present a new role of DSF and its structurally related molecules, i.e., increasing the antibiotic susceptibility of some bacterial species (Figure 1, Table 2). Given that DSF at a final concentration of 5 μM, which appears to be a physiological relevant concentration [14, 22], could substantially increase bacterial sensitivity to antibiotics (Figure 2A), it appears plausible that DSF-family signals may have a role in shaping local microbial ecology as they could reduce the competitive advantage of some community residents by down regulation of their antibiotic or toxin tolerance. Furthermore, our results also suggest that DSF and its structurally related molecules may be used as a new kind of antibiotic adjuvant for the treatment of infectious diseases caused by bacterial pathogens, subjecting to further evaluation of their toxicological and pharmacological properties.
DSF-family signals share a fatty acid carbon chain with variations in chain length, double-bond configuration, and side-chain . Evidence is emerging that these structural features may contribute to their biological activity in intraspecies signalling and interspecies communication [14, 17, 37]. Our study showed that the synergistic activity of DSF and its structurally related molecules with antibiotics is influenced by their structural features. Each of these molecules has a distinct synergistic activity among which the disparity could be up to 128-fold (Figure 1A). As a general rule, our results showed that the unsaturated long chain DSF related molecules have better synergistic activity with antibiotics, especially the aminoglycoside antibiotics, than the short chain and saturated molecules. Meanwhile, the synergistic activity of DSF and related molecules may also seem to be affected by the mode of action of antibiotics as the synergistic activities of DSF and related molecules with aminoglycoside antibiotics such as gentamicin and kanamycin were much better than with other types of antibiotics (Figure 1, Table 2).
It was reported that BDSF signalling system positively regulates the antibiotic resistance of B. cenocepacia. The same research group also found that addition of DSF signal to P. aeruginosa could increase the bacterial antibiotic tolerance to polymyxins . Intriguingly, our results suggest an opposite effect of DSF and its structurally related molecules by increasing the bacterial antibiotic susceptibility (Figure 1, Table 2). The contradictory results may be due to the differences in the bacterial species or strains and the antibiotics used in studies, which is evident from our results (Table 2). It should also be noted that DSF-family signals were shown to play dual roles in regulation of biofilm formation as they positively control the biofilm development in some bacterial species, and they could also disperse the biofilms of other bacterial species [15, 19, 21, 37].
Our results suggest that DSF and related molecules may influence the bacterial antibiotic susceptibility by multiple ways, including modulation of the biofilm formation, antibiotic resistant activity and bacterial persistence (Figure 4; Additional file 1: Table S1). In addition, we also examined the possibility of DSF and related molecules acting as biosurfactants to influence bacterial susceptibility to antibiotics by using rhamnolipid, which is a well characterized biosurfactants, as a control in MIC and growth analysis. We found that rhamnolipid could also increase the antibiotic susceptibility of B. cereus at the final concentration of 50 μM (data not shown), but it also inhibits bacterial growth at this concentration and its toxicity on B. cereus cells was at least 5-fold higher than DSF (Additional file 1: Figure S3), which complicates the comparison. With all considered, at this stage we could not rule out the possibility that DSF and related molecules may have biosurfactant property and this property may contribute to their synergistic effects with antibiotics. Furthermore, several lines of evidence from this study and previous reports seem to suggest that the signalling activity of DSF and its structurally related molecules may contribute to their ability in changing bacterial antibiotic susceptibility. Firstly, it was reported that BDSF signalling system positively controls the antibiotic resistance in B. cenocepacia, and addition of 50 μM DSF signal increased the antibiotic resistance of P. aeruginosa to polymyxins [21, 23], indicating that DSF-family signals are possibly widely involved in regulation of bacterial antibiotic resistance. Secondly, different from rhamnolipid which has a strong hydrophilic head group glycosyl, DSF and related molecules only have a very weak hydrophilic activity, suggesting that they could not be good surfactants. This notion appears to be supported by the different inhibitory activity of DSF and rhamnolipid on the growth of B. cereus (Additional file 1: Figure S3). Thirdly, our findings showed that addition of 50 μM DSF signal showed no cytotoxicity to HeLa cells, didn’t affect the B. cereus virulence (Figure 3), but could significantly change the expression patterns of many genes in B. cereus, some of which are known to be associate with antibiotics resistance or tolerance (Additional file 1: Table S1). Fourthly, the synergistic activity of DSF is antibiotic specific. While DSF and its structurally related molecules have strong synergistic effect with gentamicin and kanamycin against B. cereus, they showed a moderate effect with rifampicin, or even no synergistic effect with other antibiotics such as ampicillin, tetracycline (Data not shown), which may not be solely explainable with biosurfactant properties. Fifthly, the synergistic effect of DSF with antibiotics is also bacterial species specific. We showed that DSF signal had a strong synergistic effect with gentamicin against B. cereus, B. thuringiensis and S. aureus, while it had only a moderate effect with gentamicin against M. smegmatis, N. subflava and P. aeruginosa (Figure 1, Table 2). In particular, DSF signal did not show any synergistic activity with any of the tested antibiotics, including gentamicin, kanamycin, rifampicin, ampicillin, tetracycline, chloramphenicol, and trimethoprim, against Escherichia coli (Data not shown). Finally, DSF and its structurally related molecules share a very similarly chemical structure, hydrophobic and hydrophilic properties, suggesting that they should have similar chemical properties. However, their synergistic activities were significantly different with disparity up to 128 folds (Figure 1A).
Taken together, the results from this study have established the role of DSF and its structurally related molecules in modulation of antibiotic susceptibility in some but not all bacterial pathogens. It is also clear that the synergistic activity with antibiotics is related to the structural features of DSF-related molecules and likely the chemical property or the mode of action of antibiotics. At least stage, it is not clear how DSF and its structurally related molecules could influence bacterial antibiotic sensitivity. Much work remains to be done to determine whether their functionality in modulating bacterial antibiotic sensitivity is related to their pure chemical properties such as biosurfactant or hydrophobic activities, or associated with their potential roles in interference of bacterial signalling and regulatory networks, or both. In this regard, DSF and its analogues may be served as a useful tool to probe the potential mechanisms governing bacterial sensitivity to antibiotics.