Being a local infection not associated with mortality, vulvovaginal candidiasis (VVC) is a clinical problem that is relatively often underestimated in comparison to other types of infections, including other candidosis. Meanwhile, it is a common disease affecting a vast majority of women in the world and having a negative impact on their quality of life and health 9,42. Because the pathomechanism of VVC is still not fully understood, the achievement of therapeutic success is often difficult. Additionally, as the condition has a high tendency to relapse, it is often a challenge for both the patient and the clinician 43. Previous studies on the development of vaginal mycosis indicate a multifactorial pathological mechanism, including the formation of biofilm by Candida strains and the growing resistance of yeast-like fungi to conventionally used antimycotics 3,5,9,44. Among the etiological factors of VVC, C. albicans remains the most common, although other species are also increasingly often isolated, such as C. glabrata, C. lusitaniae, C. parapsilosis, or C. kefyr 3,5,9. Therefore, the search for new antimicrobial agents, including antifungal agents, also focuses on anti-biofilm activity.
AMPs (including lipopeptides) are among antimicrobial compounds that are of great interest for researchers, especially in recent years. These are compounds with a broad spectrum of activity and a different mechanism of action compared to traditional antibiotics, which makes it much more difficult for microorganisms to acquire resistance to them 13–15,18. The compounds used in this study were ultrashort cationic lipopeptides (USCLs). In general, USCLs consist of at most 7 amino acid residues with a fatty acid attached. Usually, USCLs are rich in arginine or lysine, which makes them cationic (net positive charge). Owing to this structure, USCLs interact with the membrane bilayers of various microorganisms, permeabilize them and cause cell death 13–15,18. Based on our previous study, compounds with the most potent antifungal and antibiofilm activities were selected. Two with linear structure, one modified by replacing one lysine residue (K) with an arginine residue (R) – C16-KKKK-NH2 (L1) and C16-KRKK-NH2 (L2), as well as two cyclic analogs – C16-CKKKKC-NH2 (C1) and C16-CKRKKC-NH2 (C2) 13. In the present study, antifungal activity of these lipopeptides was tested against 62 clinical strains of various Candida species isolated from VVC, both in planktonic and biofilm form. Reference strains of C. albicans ATCC 90028 and C. glabrata ATCC 15126 were also included. Moreover, combinations of the USCLs with fluconazole were tested for their potential synergistic or additive effect. The literature indicates the possibility of a beneficial effect of combinations of fluconazole with various AMPs against e.g. yeast-like fungi, most likely due to the different mechanisms of action of these two groups of compounds 27,34−39. Meanwhile, to the best of our knowledge, no such experiments have been performed using ultrashort cationic lipopeptides against Candida isolated from VVC.
With one exception of a single C. lusitaniae isolate (MIC = 64 µg/mL), all tested strains were found to be susceptible to fluconazole—the most common MIC was 0.125 µg/mL or less (Fig. 1A). This is not an unusual situation; more surprising is the fact of so frequent clinical therapeutic failures with this mycostatic. Considering the multifactorial pathomechanism of vaginal invasion by yeast-like fungi, their ability to form highly resistant biofilm structure may be the major cause of the ineffectiveness of conventionally used azoles. Hence the focus of research on compounds with antibiofilm activity 10,45−47. All four analyzed ultrashort lipopeptides showed activity against Candida strains. In the case of cyclic analogs, the achieved concentrations inhibiting the growth of planktonic cells are 2–3 times lower than those of their linear counterparts. The most common MIC value obtained for both cyclic lipopeptides was 4 µg/mL; ≈48% of cases for C1 and ≈ 52% for C2. Analogous concentrations for linear lipopeptides were 32 µg/mL (≈ 42%) and 16 µg/mL (≈ 47%), respectively. On the other hand, when comparing the activity of USCLs consisting of only lysine residues with lipopeptides with one arginine residue, the differences still seem to depend on the cyclic/linear structure of the compared compounds. In the case of cyclic analogs, the differences between MICs for C1 and C2 are virtually unnoticeable, which can be easily observed by comparing the histograms presented in Figs. 1B and 1C, which is consistent with previous studies on reference Candida strains 13. Generally, the antimicrobial activity of linear L2 exceeds that of lipopeptide consisting exclusively of lysine. The most common MIC was 16 µg/mL and only ≈ 32% of the determined MICs exceeded this value, while in the case of L1, as much as ≈ 69% of MICs were higher than 16 µg/mL. This finding is consistent with those of our earlier study in which activity against different Candida reference strains was analyzed 13. Similarly to our previous reports, disulfide-cyclized lipopeptides were substantially more active against both biofilm and planktonic cultures than the corresponding parent molecules. It was mentioned by Neubauer et al. that cyclic lipopeptides seem to be much more active against various species of yeast-like fungi. There is an unconfirmed hypothesis that disulfide cyclized USCLs are transported inside the fungal cell, causing degradation of the cell membrane and its interior and leading to cell death 48. Also, the potential advantage of linear analogs with the arginine residue, observed both by Neubauer and in this study, may support the first reports of the accumulation of protamine (salmon) rich in cationic arginine as necessary for anti-Candida activity 48. The determined SIs (Table 1) of cyclic lipopeptides (5.50 and 8.83) are much higher than those of their linear counterparts (0.28 and 0.89). Those results are in agreement with the literature. It has been shown that linear short cationic lipopeptides with hexadecanoic acid chain exhibited no selectivity between pathogens and normal human cells 13,15,18,49. It is worth mentioning that for similar USCL, consisting of only two lysine residues (C16-KK-NH2), antifungal activity was already demonstrated, e.g. against Cryptococcus neoformans and dermatophytes 40,50.
Similar conclusions are provided by the analysis of the obtained concentrations of biofilm eradication. While fluconazole failed to deal with Candida biofilm (Fig. 2A), all lipopeptides tested were capable of eradicating this structure (Figs. 2B and 2C). Again, cyclic analogs exhibited enhanced antibiofilm activity than linear parent molecules. While the most common MBEC value of both cyclic USCLs was 64 µg/mL (against ≈ 66% of strains for C1 and ≈ 76% for C2), the MBEC distribution for both linear compounds was more variable, with a dominant value of 256 µg/mL (against ≈ 40% and ≈ 48% of strains, respectively). However, there are no substantial differences between the values of eradicating concentrations obtained for analogs composed only of lysine residues versus compounds enriched with an arginine residue; the histograms from the obtained distribution of values for both cyclic and linear lipopeptides are almost identical (Fig. 1B and 1C). The calculated GM_MBECs support this thesis. The results of research on biofilm are also consistent with our previous reports 13. Higher minimum concentrations of compounds obtained for the biofilm structure compared to planktonic cells are not surprising. Biofilms, both bacterial and fungal, are characterized by a much higher resistance to antimicrobial agents than planktonic cells. Moreover, biofilm is associated with therapeutic failure. A single strain can be characterized as susceptible or resistant to fluconazole if examined against planktonic cells or biofilm, respectively 8,47.
The present results showed that USCLs in combination with fluconazole can give various effects. The FIC indices obtained in the checkerboard method indicate that additive antifungal effect was obtained more frequently for a combination of fluconazole and linear lipopeptides than for cyclic ones. Preliminary experiments on 24 random Candida strains and combinations of all four USCLs with the tested azole showed that indifferent antifungal effect against vaginally isolated fungi was predominantly observed with cyclic USCLs (≈ 71% for C1 and ≈ 58% for C2). In the case of the L2 lipopeptide, the additive effect with fluconazole occurred more often than for its cyclic analog (≈ 46% vs ≈ 17%). Moreover, additive effect was the most frequent with L1 (≈ 63%). Analyzing the above data with regard to the Candida species (15 isolates of C. albicans vs 9 NCAC), the obtained results for cyclic lipopeptides are very similar - both C. albicans and other species represented an indifferent effect, while for strains from the NCAC group no additive effect was observed. A comparison of the effect of linear USCLs against C. albicans indicates the advantage of the analog consisting of four lysine residues, for which a vast majority has an additive effect (73%). Studies on the linear compound enriched with arginine and combined with fluconazole revealed that it had an additive and indifferent effect on a similar percentage of strains (53% and 47%, respectively). Due to the small number of isolates from the NCAC group, a detailed analysis of the distribution of the obtained results seems unreliable.
Lipopeptide L1 was selected for further studies with fluconazole owing to promising results in the preliminary results discussed above. The cyclic analog was included in this study to learn how different structures of USCLs in combination with fluconazole can affect antifungal activity against strains derived from VVC. Although in the case of both tested cyclic compounds the indifference was definitely the dominant effect, in the case of C1, a negative (antagonistic) effect was observed less frequently than for C2 (8% and 25%, respectively). Hence, testing of the remaining pool of 40 strains was performed on a pair of USCLs composed only of lysine residues. The overall FIC index analysis for the entire pool of 62 isolates and 2 reference strains confirmed the predominant additive effect of the fluconazole-L1 combination: 84% in total, 91% including C. albicans strains and 55% of NCAC, although in this case, the relatively small number of isolates (11) should still be kept in mind. Interestingly, similar results were obtained for the cyclic lipopeptide, for which an additive effect was observed in 53% of cases, of which 60% against C. albicans. In the NCAC group, indifference remained the most frequent result (64%). The literature has described the possibility of a favorable antifungal effect due to combinations of fluconazole with various AMPs 34–39. The use of compounds with different mechanisms of action is generally recommended. Combined antifungal therapy has many potential benefits, such as enhancement of the fungicidal effect and broadening the spectrum of activity, which enable to fight polymicrobial infections, reduce the dose of the compounds, and thus also reduce dose-dependent toxicity as well as overcome the resistance of microorganisms 51,52. It is noted that antimicrobial peptides interacting with the membrane bilayers could, in a way, sensitize Candida cells to fluconazole by increasing azole penetration into the cell, where its molecular target — 14α-lanosterol demethylase (Erg enzyme)—is located, involved in ergosterol synthesis. In effect, the composition of the cell membrane changes; it liquefies and increases the permeability for K+ and ATP causing a fungistatic effect 53. On the other hand, the interaction of fluconazole with the cell membrane may enhance its permeabilization by various AMPs, including lipopeptides, and enhance their fungicidal activity 51,54,55. Other mechanisms that may be responsible for the synergistic effect of combining triazoles with compounds with a different mechanism of action include sequential inhibition of different stages in the mutual biochemical pathway or simultaneous interaction with the Candida cell wall and/or membrane 51. However, this hypothesis remains unconfirmed as of today.
Last but not least, there is one more interesting aspect of research using the checkerboard method. Knowing the general nature of the interaction of the combination of fluconazole with the lipopeptide against Candida, the next step is to select the most favorable and effective concentrations of both compounds to combat fungi. For this purpose, the individual FIC (as the FIC index is the arithmetic mean of 8 different FIC values) obtained for a given pair of compounds should be interpreted and the corresponding concentrations selected. In this way, the most favorable (the lowest) FIC for each of the isolates was analyzed in both fluconazole-L1 and fluconazole-C1 combinations. In 72% of strains (75% of C. albicans, and 55% of NCAC) there was such a combination of the concentrations of linear lipopeptide and FLC for which FIC indicated a synergistic effect (FIC ≤ 0.5). The most beneficial FICs for the cyclic lipopeptide still showed a predominant additive effect (61%), although a synergistic effect was also observed (23%). A detailed analysis revealed that the concentration of L1 in combination with fluconazole that result in synergy is up to four-fold lower (2 vs. 16 µg/mL) than when lipopeptide is used separately (MIC value, Table 2). The literature contains an increasing body of reports about the results of similar studies of interactions of different compounds, not always having any activity against Candida alone, indicating a synergistic effect of their combinations with a number of antimycotics, including triazoles. There is a great interest in research on amphiphilic Lactoferrin (LF and its derivatives — cationic peptides), the use of which together with fluconazole (and not only) results in a significant increase in fungistatic activity and a decrease in MIC values 56–59. Although the mechanisms responsible for this phenomenon remain unexplained, attention is drawn to the beneficial effects of cationic compounds, such as LF, which may enhance the hydrophobicity of the surface of microbial cells and potentiate the antifungal activity of other compounds 57. Another example is the advantageous synergistic effect of combinations with fluconazole of such positively charged compounds as microbicidal cationic oligomers, styrylpyridinium compounds, novel antimicrobial peptides such as KABT-AMP derivatives or ToAP2 51,55,60,61. Recently, the synergism of fluconazole with surfactin (SU) against C. albicans has been described in more detail. Suchodolski et al. showed that SU binds to chitin and β-glucan on the surface of fungal cells, exposing it to the components of the host's immune system. However, to achieve the necessary effect, there seems to be required a reduction or complete lack of ergosterol, resulting in the corresponding changes in cell membrane and this is ensured by the presence of fluconazole 27. Derivatives of quaternary ammonium compounds (QAC) are other compounds whose activity is similar to that of cationic surfactants and which at the same time have a structure similar to lipopeptides (positive charge, presence of a lipid chain). One representative of this group, compound K21, has recently been tested for antifungal activity for the first time. It seems to be an effective alternative to fluconazole against Candida strains resistant to this mycostatic. K21 also shows synergism with triazoles towards NCAC, including C. dubliniensis and C. tropicalis, but no such effect was observed for C. albicans 62. Meanwhile, a combination of another quaternary ammonium compound, domiphen bromide, with miconazole (imidazole) showed a synergistic effect against not only Candida planktonic cells, but also a biofilm — although this effect did not occur in the case of triazoles, including fluconazole 63. On the other hand, the mechanism of action of benzimidazolium-based QAC gemini surfactants was identified as influence on ergosterol synthesis in a manner similar to that of triazoles. Nevertheless, benzimidazolium-based QACs were more effective and their combination with fluconazole results in synergistic effect against various Candida species 64. There are also reports in the literature about the synergistic effect of USCL with a structure similar to L1 and L2 with fluconazole and other triazoles, as well as with amphotericin B or terbinafine against Cryptococcus neoformans fungi and various representatives of dermatophytes 40,50. However, the mechanism of interaction of the above-mentioned compounds alone and in combination against fungi remains unclear. The literature and our results together clearly demonstrate the enormous potential of ultrashort cationic lipopeptides as compounds enhancing the activity of the existing antimycotics.
Therefore, the results achieved in this work constitute another important premise in the search for antifungal compounds and their combinations with conventional mycobiotics. Moreover, our findings can contribute to the broadening of knowledge in the search for mechanisms involved in the interactions of various cationic compounds with target cells and other antimicrobial compounds. Importantly, the most serious problems to be solved before the actual use of USCLs in the treatment of fungal infections are their relatively high toxicity and unsatisfactory selectivity between microorganisms and human cells 15,18. The use of combination therapy has a potential to significantly reduce the concentration of lipopeptides effective against Candida and to reduce toxicity towards human cells.