Antifungal Activity of Oleylphosphocholine on In Vitro and In Vivo Candida albicans Biofilms

ABSTRACT

In this study, we investigated the potential antifungal activity of the alkylphospholipid oleylphosphocholine (OlPC), a structural analogue of miltefosine, on in vitro and in vivo Candida albicans biofilm formation. The effect of OlPC on in vitro and in vivo C. albicans biofilms inside triple-lumen polyurethane catheters was studied. In vivo biofilms were developed subcutaneously after catheter implantation on the lower back of Sprague-Dawley rats. Animals were treated orally with OlPC (20 mg/kg of body weight/day) for 7 days. The effect of OlPC on biofilms that developed on the mucosal surface was studied in an ex vivo model of oral candidiasis. The role of OlPC in C. albicans morphogenesis was investigated by using hypha-inducing media, namely, Lee, Spider, and RPMI 1640 media. OlPC displayed activity against both planktonic cells and in vitro C. albicans biofilms. To completely abolish preformed, 24-h-old biofilms, higher concentrations (8, 10, and 13 mg/liter) were needed. Moreover, OlPC was able to reduce C. albicans biofilms formed by caspofungin-resistant clinical isolates and acted synergistically when combined with caspofungin. The daily oral administration of OlPC significantly reduced in vivo C. albicans biofilms that developed subcutaneously. In addition, OlPC decreased biofilm formation on mucosal surfaces. Interestingly, the application of subinhibitory concentrations of OlPC already inhibited the yeast-to-hypha transition, a crucial virulence factor of C. albicans. We document, for the first time, the effects of OlPC on C. albicans cells and suggest the potential use of OlPC for the treatment of C. albicans biofilm-associated infections.

INTRODUCTION

In the past decade, we have been dealing with a growing population at risk of invasive fungal infections. The persistence of these infections is a problem, especially when fungal pathogens form biofilms on implantable medical devices. Since the employment of a device is often compulsory for the survival of patients with an impaired immune system, the ability of Candida albicans to colonize and develop biofilms on artificial substrates, as well as on mucosal surfaces, has a significant impact on the mortality and morbidity of this patient group. In comparison with their planktonic counterparts, Candida cells associated with biofilms withstand high concentrations of antifungal drugs (1). So far, only echinocandins and liposomal formulations of amphotericin B have displayed significant activity against fungal biofilms (2). The increased use of echinocandins as therapy against invasive candidiasis may lead to increasing rates of acquired resistance against these compounds, which is associated with amino acid changes in “hot spot” regions of the cellular targets echinocandins, the Fks subunits (3). Due to the limited number of drugs that are active against biofilms, it is crucial to search for novel strategies to prevent adhesion, which is the first step of biofilm development, or to obviate biofilm formation once the cells are attached to a substrate. Recently, miltefosine, an alkylphospholipid, was shown to have very good activity against C. albicans biofilms (4). It was also effective against oral candidiasis that developed on murine tongues (5). However, side effects and teratogenicity narrow the therapeutic window and limit its potential use (6, 7). A variant of miltefosine, oleylphosphocholine (OlPC), exhibited fewer side effects when tested in vivo and had good oral bioavailability. Moreover, it demonstrated very good antineoplastic and antiprotozoal activities, with a large safety margin (8). It was previously shown that OlPC was effective against infections caused by Leishmania infantum in a hamster model (8) and in naturally infected dogs (9) and furthermore was effective against cutaneous leishmaniosis caused by Leishmania major that developed in a mouse model (10). Interestingly, OlPC was also active against Aspergillus spp. (11) and Cryptosporidium parvum (12). Because of these findings, we assumed that OlPC could be a good candidate to treat device-associated C. albicans infections. Therefore, we investigated the potential effect of OlPC on C. albicans planktonic cells and on in vitro biofilms that developed on polyurethane catheters. Moreover, the activity of OlPC on in vivo C. albicans biofilms that developed on foreign bodies in a subcutaneous rat model is documented here for the first time. Additionally, we assessed the potency of OlPC against biofilm formation on mucosal tissue using an ex vivo tongue assay as a model. We were also able to demonstrate the activity of OlPC against C. albicans strains with increased tolerance to the echinocandin caspofungin and a possible mode of action for this molecule based on the inhibition of the yeast-to-hypha transition.

RESULTS

Planktonic cells of C. albicans display susceptibility to OlPC.

First, we evaluated the MICs of OlPC that inhibited the growth of planktonic C. albicans cells in vitro by 50% (MIC₅₀). These results are presented in Table 1. The MIC₅₀ indicated that a concentration range between 1 mg/liter and 4 mg/liter of OlPC inhibited the growth of C. albicans. Moreover, we found that concentrations starting from 2 mg/liter of OlPC had a fungicidal effect on planktonic growth, documented as the minimal fungicidal concentration (MFC).

TABLE 1.

MICs and MFCs of OlPC against planktonic Candida albicans cells

Strain or clinical isolate	MIC50 (mg/liter)	MFC (mg/liter)
C. albicans SC5314	2	2
C. albicans U-0503-18	2	2
C. albicans U-0503-12	1	2
C. albicans U-0203-29	1	2
C. albicans U-0203-15	1	2
C. albicans U-04030-255	1	2
C. albicans HC2601-30	2	2
C. albicans HC0202-25	2	2
C. albicans HC2102-19	2	2
C. albicans HC1602-5	2	2
C. albicans HC0702-14	2	2
C. albicans M89	4	4
C. albicans M177	2	2
C. albicans M205	4	4

OlPC alters in vitro C. albicans biofilm development and preformed biofilms, whereas adhesion remains unaffected.

Prior to biofilm susceptibility testing, we evaluated the abilities of C. albicans SC5314 and all clinical isolates to adhere to and form biofilms on polyurethane devices (see Fig. S1 in the supplemental material). The Candida wild-type (WT) strain and all clinical isolates exhibited comparable abilities to attach to and form biofilms on catheters.

Next, we assessed whether OlPC would influence the process of biofilm formation. Therefore, we investigated the influence of OlPC on three different developmental stages in biofilm formation, including adhesion, biofilm development, and on preformed biofilms. To know the effect of OlPC on C. albicans SC5314 adhesion, catheters preincubated with serum were submerged in a mixture containing C. albicans cells and different concentrations of OlPC for a period of 90 min at 37°C. The results indicated that OlPC had no effect on C. albicans adhesion (Fig. 1A). Next, we investigated the activity of OlPC on biofilm formation by treatment of the biofilms after adhesion for 24 h. The results show that biofilm development of the Candida WT strain (Fig. 1B) and clinical isolates (Fig. 2A and B) was significantly reduced at 2 mg/liter and 4 mg/liter of OlPC, respectively (P < 0.05). Strikingly, higher concentrations (8, 10, and 13 mg/liter) completely abolished biofilm development. The biofilm architecture of nontreated WT cells showed the formation of a thick layer of hyphal cells embedded in material resembling an extracellular matrix (ECM) (Fig. 1D). Candida cells treated with 1 and 2 mg/liter of OlPC displayed a rudimentary biofilm composed of hyphal cells attached to a substrate. Strikingly, only yeast cells were observed on the surface of catheters when treated with 4 mg/liter of OlPC. Scanning electron microscopy (SEM) images were in agreement with the CFU counts, displaying that no biofilms were observed on substrates treated with 8, 10, and 13 mg/liter of OlPC (data not shown).

FIG 1.

Antifungal activity of OlPC against different stages of Candida albicans SC5314 biofilm formation. (A to C) Different concentrations of OlPC (0, 0.5, 1, 2, 4, 8, 10, and 13 mg/liter) were introduced to C. albicans cells during the period of adhesion (90 min) (A) and after adherence for the next 24 h of biofilm development (B) and on preformed 24-h-old biofilms (C). Experiments with adhesion and biofilm formation were performed on serum-coated polyurethane catheter pieces. After the desired period, catheters were washed, sonicated, and plated onto YPD plates to enumerate the amounts of adhered and biofilm-forming cells. Each column represents the mean log₁₀ CFU ± standard errors of the means retrieved from catheters from 3 independent experiments. The asterisk indicates a statistically significant difference between the nontreated group and the OlPC-treated groups (*, P ≤ 0.05). (D) Scanning electron microscopy of catheters treated with OlPC during biofilm development for 24 h. Samples were coated with Au-Pd prior to microscopy. Arrows indicate the presence of scattered hyphal and yeast cells on the catheter surface.

FIG 2.

Efficacy of OlPC against Candida albicans biofilms formed by clinical isolates isolated from urinary catheters and blood culture. (A and B) Effect of different concentrations of OlPC (0, 0.5, 1, 2, 4, 8, 10, and 13 mg/liter) on biofilm development of C. albicans clinical isolates obtained from urinary catheters (A) and blood culture (B). OlPC was administered to Candida cells after the period of adhesion for the next 24 h. (C and D) Activity of OlPC on preformed C. albicans biofilms formed by clinical isolates from urinary catheters (C) and blood culture (D). In this assay, biofilms were allowed to develop in drug-free medium. OlPC was added subsequently for the next 24 h. Data are presented as the mean log₁₀ CFU ± standard errors of the means for catheters from 2 independent experiments.

These data demonstrated the activity of OlPC against biofilm development; however, preformed biofilms represent a major clinical problem in hospitalized patients. Therefore, we examined the activity of this compound on preformed (24-h-old) biofilms of C. albicans SC5314 (Fig. 1C) and clinical isolates (Fig. 2C and D). Concentrations starting from 2 mg/liter and 8 mg/liter of OlPC significantly reduced the biofilms of clinical isolates and the WT, respectively (P < 0.05). No cells were recovered from WT biofilms treated with 10 and 13 mg/liter of OlPC. Only a few Candida cells were retrieved from biofilms formed by clinical isolates U-0203-15 and U-0403-255 after treatment with 10 and 13 mg/liter of OlPC.

Oral administration of OlPC significantly decreases C. albicans biofilm development in a rat subcutaneous model.

We investigated the activity of OlPC on in vivo C. albicans biofilm development in a rat subcutaneous biofilm model. Figures 3A and B demonstrate that the daily oral administration of OlPC (20 mg/kg of body weight/day for 7 days) to Sprague-Dawley rats resulted in a significant inhibition of C. albicans biofilms (P < 0.05) (CFU determined on day 9). Confocal laser scanning microscopy (CLSM) images of biofilms retrieved from the control group of animals displayed a thick network (∼135 μm) of hyphal cells characteristic of Candida biofilm architecture (Fig. 3C; see also Movie S1 in the supplemental material). Strikingly, the examination of biofilms from OlPC-treated animals showed decreased fungal loads with scattered yeast and hyphal cells alongside the lumen. The thickness of such biofilms was ∼67 μm (Fig. 3C). Taken together, these results demonstrate the potency of OlPC as a candidate to treat device-associated infections.

FIG 3.

OlPC is active against in vivo Candida albicans SC5314 biofilms. (A) Effect of oral administration of OlPC on C. albicans SC5314 biofilms developed in the subcutaneous rat model. Animals were treated with OlPC (20 mg/kg of body weight/day) daily for 7 days. Data are presented as log₁₀ numbers of CFU retrieved from each catheter piece. (B) Mean log₁₀ CFU ± standard errors of the means from independent catheters/rats. The horizontal lines indicate the mean log₁₀ CFU obtained per device or per rat. The asterisk indicates a statistically significant difference between the control group (nontreated) and the OlPC-treated group (*, P ≤ 0.05). (C) Confocal laser scanning microscopy of biofilm structures that developed inside polyurethane catheters of control and OlPC-treated animals. After catheter removal, devices were cut longitudinally and stained with concanavalin A (50 mg/liter) (green fluorescence) at 37°C for 30 min. Arrows demonstrate rudimentary biofilms retrieved from catheters from OlPC-treated animals.

OlPC decreases Candida albicans biofilm formation on the mucosal surface in an ex vivo model of candidiasis.

The emergence of biofilm-related infections on mucosal surfaces represents a major medical problem. Oral thrush, found in the oral cavity of immunocompromised patients, is one of the most common forms of biofilms (13). Therefore, we employed an ex vivo model for biofilm formation on mucosal surfaces (tongues extracted from sacrificed mice) to assess the effect of OlPC on C. albicans biofilm development on soft tissues. Candida cells were allowed to attach to the tissue during the adhesion phase and were subsequently grown for 24 h (see Fig. S2 in the supplemental material) in the presence of OlPC. Histopathology images of nontreated tongues and tongues treated with 1 and 2 mg/liter of OlPC exhibited intense yeast-to-hypha switching and massive dissemination of Candida hyphae inside the tissues (Fig. 4). However, scattered amounts of hyphae and a large majority of yeast cells were displayed on tissues treated with 4 mg/liter of OlPC. Strikingly, Candida cells treated with 8 and 10 mg/liter of OlPC remained solely in the yeast form, with no signs of dissemination. Importantly, no alterations in the structures of the tissues between noninfected tongues and samples treated with the highest concentration of OlPC (13 mg/liter) were observed (Fig. S3). These data suggest that OlPC has the potential to treat not only catheter-associated infections but also biofilm development on mucosal surfaces.

FIG 4.

Effect of OlPC on Candida albicans SC5314 biofilm formation on mucosal surfaces in an ex vivo model of oral infection. Shown are data from histopathology analyses of mouse tongues infected with C. albicans in the presence and absence of OlPC. Representative microscopic images from periodic acid-Schiff-stained tissues display significantly decreased adherence and invasion of tongue tissue treated with 4, 8, and 10 mg/liter of OlPC. Candida cells treated with 8 and 10 mg/liter of OlPC did not undergo a yeast-to-hypha transition and were found solely as yeasts. Magnification, ×40.

OlPC is active against C. albicans biofilm development by clinical isolates resistant to caspofungin; moreover, OlPC synergizes with caspofungin to induce cell death of caspofungin-resistant isolates.

The echinocandins form a unique class of drugs that possess activity against in vitro and in vivo biofilms (14, 15). However, by its excessive use, there is an increasing incidence of clinical isolates of C. albicans that show resistance against them. This raised the question of whether OlPC would be an effective alternative against C. albicans isolates M177, M205, and M89 (Table 2), with elevated MICs against caspofungin (16). OlPC at 2 mg/liter and 4 mg/liter inhibited fungal growth by 50% (MIC₅₀). Additionally, 2 mg/liter of OlPC caused a fungicidal effect on C. albicans M177 planktonic cells, while a dose of 4 mg/liter was sufficient to abolish the growth of C. albicans M89 and M205 (Table 1).

TABLE 2.

Strains and clinical isolates used in this study

Strain or clinical isolate (reference)	General characteristic	Origin
C. albicans SC5314 (24)	Wild type	Disseminated candidiasis
C. albicans U-0503-18	Clinical isolate	Urinary catheter
C. albicans U-0503-12	Clinical isolate	Urinary catheter
C. albicans U-0203-29	Clinical isolate	Urinary catheter
C. albicans U-0203-15	Clinical isolate	Urinary catheter
C. albicans U-04030-255	Clinical isolate	Urinary catheter
C. albicans HC2601-30	Clinical isolate	Blood culture
C. albicans HC0202-25	Clinical isolate	Blood culture
C. albicans HC2102-19	Clinical isolate	Blood culture
C. albicans HC1602-6	Clinical isolate	Blood culture
C. albicans HC0702-14	Clinical isolate	Blood culture
C. albicans M89 (16)	Clinical isolate
C. albicans M177 (16)	Clinical isolate
C. albicans M205 (16)	Clinical isolate

Furthermore, we investigated the activity of OlPC against biofilms formed by WT strain SC5314 (Fig. 5A) and caspofungin-resistant isolates (Fig. 5B to D). OlPC (4 mg/liter) significantly diminished preformed C. albicans M177 and M205 biofilms, whereas 8 mg/liter significantly decreased C. albicans M89 biofilms (P < 0.05). As expected, caspofungin alone was shown to be ineffective. Strikingly, no fungal cells were recovered from C. albicans M89 and M177 biofilms treated with 8, 10, and 13 mg/liter of OlPC (P < 0.05). A similar phenomenon was observed for C. albicans M205 biofilms treated with 10 and 13 mg/liter of OlPC (P < 0.05). These results strongly emphasize that OlPC may be an important component in the treatment of planktonic cells and biofilms formed by clinical isolates that are resistant to caspofungin.

FIG 5.

OlPC displays antifungal activity against Candida albicans biofilms formed by caspofungin-resistant clinical isolates (C. albicans M89, M177, and M205). C. albicans SC5314 biofilms were susceptible to caspofungin (A), whereas fungal cells within biofilms formed by three caspofungin-resistant isolates were tolerant to the highest concentration of this drug (B to D). OlPC significantly decreased the development of preformed biofilms by C. albicans M89, M177, and M205 (*, P < 0.05). Data are presented as the mean log₁₀ CFU ± standard errors of the means from catheters from 3 independent experiments. Black columns represent caspofungin-treated samples, whereas white bars show samples treated with OlPC.

In a next step, we explored a novel therapeutic approach, based on the combination of caspofungin and OlPC. This may result in the inhibition of C. albicans clinical isolates M177, M205, and M89, which are resistant to caspofungin. Overall, the combination of both drugs led to an improvement of MIC₅₀ values against all strains tested (Table 3). The combination of both drugs against the growth of C. albicans SC5314 resulted in a fractional inhibitory concentration index (FICI) value of 0.73, indicating no interaction. Strikingly, when this strategy was used on caspofungin-resistant isolates, a synergistic effect was observed, suggesting that this combination may represent an alternative way to treat infections caused by isolates that are resistant to caspofungin (Table 3).

TABLE 3.

Potential combination of caspofungin and OlPC against planktonic cells of caspofungin-susceptible (Candida albicans SC5314) and caspofungin-resistant (Candida albicans M177, M205, and M89) clinical isolates^a

Strain	MICCAS[alone] (mg/liter)	MICCAS[combined] (mg/liter)	MICOlPC[alone] (mg/liter)	MICOlPC[combined] (mg/liter)	FICI
SC5314	0.125	0.06	2	0.5	0.73
M177	8	1	2	0.5	0.38
M205	16	4	4	1	0.50
M89	8	0.5	4	1	0.31

^aInterpretation of drug combination interactions was based on the FICI (26) {FICI = (MICA_[combined]/MICA_[alone]) + (MICB_[combined]/MICB_[alone])}, where an FICI of ≤0.5 indicates a synergistic interaction, an FICI of 0.5 to 4 indicates no interaction, and an FICI of >4.0 indicates an antagonistic interaction.

OlPC strongly inhibits the yeast-to-hypha transition, a crucial virulence factor for C. albicans pathogenesis.

In this study, we noticed an increased amount of yeast cells and reduced hypha formation in OlPC-treated biofilms (Fig. 1D) compared to the control. Therefore, we examined the morphological changes of planktonic cells in hypha-inducing media, namely, RPMI 1640, Spider, and Lee media, upon treatment with OlPC (0.5, 1, and 2 mg/liter) at 37°C. The pictures shown in Fig. 6 were taken after 4 h and 24 h. Cells grown in RPMI 1640 and Spider media for 4 h and treated with 0.5 mg/liter of OlPC were able to undergo morphogenesis. Cells treated with higher concentrations (1 and 2 mg/liter) and grown in RPMI 1640 and Spider media for 4 h remained in the yeast form. Cells cultured in Lee medium and treated with even the lowest concentration of OlPC (0.5 mg/liter) were found solely as yeasts. After 24 h of growth, nontreated cells and Candida cells treated with 0.5 mg/liter of OlPC created typical clusters of hyphae and pseudohyphae. OlPC-treated (1 and 2 mg/ml) Candida cells appeared individually as yeasts or in groups composed of aberrant hyphae. These findings demonstrate a strong effect of OlPC on C. albicans morphogenesis, which can be considered a potential mode of action for this molecule.

FIG 6.

OlPC inhibits Candida albicans SC5314 yeast-to-hypha transition. Shown are light microscopy images of the C. albicans yeast-to-hypha transition in hypha-inducing media, such as RPMI 1640, Spider, and Lee media, at 37°C. Samples were taken 4 h and 24 h after the administration of different concentrations of OlPC (0.5, 1, and 2 mg/liter) to fungal cells. Control (no drug) and OlPC-treated (0.5 mg/liter) cells were able to undergo the switch after 4 h of growth in RPMI 1640 and Spider media. Cells treated with 0.5 mg/liter OlPC in Lee medium remained in the yeast form. At higher concentrations tested (1 and 2 mg/liter), Candida cells did not switch from yeast to hyphae. The formation of hyphae and pseudohyphae was noticed after 24 h of growth under control conditions (no drug) and with 0.5 mg/liter OlPC. At 1 and 2 mg/liter OlPC, cells were present as pseudohyphae, and the majority of cells remained in the yeast form. MOPS, morpholinepropanesulfonic acid. Bar, 10 μm. Magnification, ×60.

DISCUSSION

In the present work, we proved that OlPC has antifungal activity against planktonic and biofilm-forming C. albicans cells. The present study is the first to deal with the antifungal effects of OlPC on C. albicans. Used against planktonically grown cells, OlPC showed high activity against all clinical isolates tested, with MIC values ranging between 1 and 4 mg/liter. Moreover, fungicidal activity of OlPC against C. albicans cells was reported for concentrations ranging between 2 and 4 mg/liter, which are identical to the MICs of almost all isolates tested. Interestingly, similar concentrations of OlPC were able to reduce or to completely abolish biofilm development, reminiscent of the results that were previously obtained with miltefosine (5, 11). Additionally, we show that highly dense preformed biofilms can also be inactivated with OlPC at concentrations only 2 to 4 times higher than the MIC for planktonic cells. Existing data show that, in vitro, all azoles, including the “gold standard” fluconazole and the newer molecule voriconazole, require concentrations of >1,000 and 256 mg/liter, respectively, to inactivate cells present in a biofilm. Moreover, Candida biofilms displayed decreased sensitivity to almost all available antifungals, including amphotericin B, flucytosine, nystatin, and terbinafine, as reviewed by Taff et al. (17). Next, we employed a subcutaneous C. albicans biofilm rat model to determine whether the in vitro activity of OlPC against Candida biofilms could be translated in vivo. In this particular model, C. albicans biofilms develop inside the lumen of a catheter piece, and cells dispersing from biofilms spread only into the tissue surrounding the device, without affecting vital organs (18). This model has already successfully been used to study the effects of fluconazole and echinocandins on preformed C. albicans biofilms (14, 15). Fluconazole (125 mg/kg of body weight/day) failed to be active against C. albicans biofilms, whereas caspofungin (5 mg/kg of body weight/day), anidulafungin (10 mg/kg of body weight/day), and micafungin (20 mg/kg of body weight/day) remained effective (14, 15). Despite the effectiveness of echinocandins against C. albicans biofilms, the intravenous route of administration of these drugs is a major disadvantage. OlPC is an orally bioavailable drug, which represents a great practical advantage for a patient. In our study, rats were given OlPC (20 mg/kg of body weight/day) orally, daily for 7 days. This treatment resulted in a significant reduction but not a complete eradication of biofilm-forming cells retrieved from catheters, accompanied by the rudimentary biofilm architecture. In comparison with in vitro data, where the action of drugs on C. albicans catheter-associated cells is more direct, we could not achieve a complete inhibition of biofilms in vivo. In fact, pharmacodynamics (PD), pharmacokinetics (PK), bioavailability, and drug interactions represent major factors that may influence the effect of drugs on pathogens in vivo. Hence, it is crucial to understand the limitations of the C. albicans in vitro model to mimic the situation in vivo. As mentioned above, we tested only one regimen of OlPC (20 mg/kg of body weight/day for 7 days) on C. albicans biofilms in vivo and looked only at one time point after treatment (day 8), which can be considered limitations of this study. Although PK/PD analyses of OlPC were not performed in this study, pharmacokinetic analyses have been established by using mice (10), hamsters (8), and dogs (9), documenting its good oral bioavailability and extensive tissue distribution. Moreover, based on the long half-life of OlPC (∼50 h in rodents), 7-day treatment is most likely to lead to an accumulation of the drug, which will prolong the therapeutic effect beyond the end of the treatment period (8). In the future, we suggest that other total doses of OlPC should be administered and that the evolution of the treatment effect should be monitored for a longer period of time, for example, by bioluminescence imaging.

Besides the activity of OlPC on biofilms formed on abiotic surfaces, we have also demonstrated that OlPC can be a good candidate to treat candidiasis that develops on mucosal surfaces. Quite recently, the topical administration of miltefosine (50 mg/kg twice a day) was successfully used to treat oral candidiasis in immunocompromised mice (5). Histopathology images (Fig. 4A) provide great evidence that tongues treated with 8 and 10 mg/liter of OlPC displayed lower fungal burdens. In order to support these promising findings, future experiments dedicated to OlPC treatment of oropharyngeal candidiasis in an animal model are essential.

Although echinocandins remain a favorable class of drugs for the treatment of invasive candidiasis, its widespread usage is one of the main reasons for drug resistance among clinical Candida isolates (19). Echinocandins proved to be some of the most active agents in the treatment of catheter-associated infections caused by wild-type C. albicans strain SC5314 in a subcutaneous rat model (15). In this study, we show that in vitro biofilms formed by C. albicans caspofungin-resistant clinical isolates did not respond to echinocandin therapy, whereas OlPC was highly active against device-associated infections by these strains. Besides biofilms, planktonic cells responded to OlPC therapy as well. The concentration for the fungicidal activity of OlPC against free-living cells of these clinical isolates ranged between 2 and 4 mg/liter. These findings demonstrate that OlPC can be a valuable option for the treatment of catheter-associated infections caused by C. albicans clinical isolates that are resistant to echinocandins. As an alternative approach, we performed a checkerboard assay to demonstrate potential synergism when both compounds are combined to fight infections caused by caspofungin-resistant isolates. Interestingly, all three caspofungin-resistant clinical isolates responded to the combinatorial therapy, and both drugs behaved synergistically (Table 3). In comparison with other compounds and existing drugs on the market, OlPC seems to be cleared rapidly, which may be beneficial with regard to the potential for the selection of resistance (10). Our findings demonstrate the potential of the combination of echinocandins with OlPC, which can reduce the growth of planktonic cells that are resistant to caspofungin.

In this study, we have revealed that C. albicans cells treated with 1 and 2 mg/liter of OlPC did not undergo a yeast-to-hypha transition, even in hypha-inducing media. After 24 h of treatment, only yeast cells and pseudohyphae were present in treated cultures (1 and 2 mg/liter of OlPC) compared to the control. Additionally, C. albicans cells found on the surface of catheters after treatment with 4 mg/liter of OlPC remained in the yeast form (Fig. 1). Although the exact mode of action of OlPC on C. albicans is not known, we observed that the application of a sublethal dose of OlPC (1 and 2 mg/liter of OlPC) already inhibited the C. albicans yeast-to-hypha transition, even in hypha-inducing media. After 24 h of treatment, only yeast cells and pseudohyphae were present in treated cultures compared to the control. Additionally, C. albicans cells found on the surface of catheters after treatment with 4 mg/liter of OlPC remained in the yeast form. Future analyses of the effect of OlPC on known transcriptional regulators involved in the control of the yeast-to-hypha transition should be able to identify the transcription factors and signaling pathway(s) involved in the OlPC response. The extent to which these pathways are also responsible for clearing mature biofilms and/or the fungicidal effect of OlPC will need to be investigated. Interestingly, miltefosine has been shown to impact mitochondrial functioning, thereby leading to enhanced oxidative stress and apoptotic-like cell death (20–22). As mitochondrial function was recently linked to the regulation of virulence pathways in C. albicans, including the morphogenetic switch to hyphae (23), mitochondria are an attractive potential target for OlPC, although this remains to be verified.

Taken together, data from this work confirmed that low concentrations of OlPC are efficacious in inhibiting C. albicans planktonic cells and biofilm formation in vitro and that preformed biofilms are also susceptible to this drug. Its activity against caspofungin-resistant clinical isolates highlights the importance of expanding studies on its antifungal properties. Finally, the demonstration of the oral activity of OlPC in vivo in a subcutaneous biofilm model of catheter-associated candidiasis supports the potential of this molecule as a promising antifungal candidate for the treatment of deep-seated Candida infections.

MATERIALS AND METHODS

Ethics.

All animal experimental procedures were performed in accordance with KU Leuven animal care guidelines and were approved by the Ethical Committee of KU Leuven (project number P090/2013).

Strains and clinical isolates.

C. albicans standard strain SC5314 (24) and clinical isolates obtained from urinary catheters and blood cultures (UZ Gasthuisberg, Belgium) were used. All clinical isolates used in this study are described in Table 2. All strains and clinical isolates were grown on YPD medium (1% yeast extract, 2% Bacto peptone, and 2% d-glucose, supplemented with 2% agar) at 37°C.

OlPC and caspofungin.

OlPC in a crystalline form was kindly provided by Dafra Pharma R&D (Turnhout, Belgium). The stock solution of OlPC was prepared in sterile water. For all experiments, different concentrations of OlPC (0.0625, 0.125, 0.25, 0.5, 1, 2, 4, 8, 10, and 13 mg/liter) were tested, unless stated otherwise. Because of the fact that OlPC is an alkylphospholipid with the tendency to foam, the highest concentration that did not cause this phenomenon was 13 mg/liter. Caspofungin (Cancidas; MSD) was prepared in dimethyl sulfoxide (DMSO) and further diluted in RPMI 1640 before use. Multiple concentrations of caspofungin, namely, 0.0625, 0.125, 0.25, 0.5, 1, 2, 4, 8, 10, and 13 mg/liter, were tested.

MIC and MFC determinations.

The antimicrobial activity of OlPC against planktonic growth was assessed according to the CLSI M27-A3 protocol (25). Data were determined as MIC₅₀ values. Next, 100 μl of a C. albicans suspension, treated with OlPC, was plated onto YPD plates, which were further incubated at 37°C overnight. The concentration of OlPC that completely abolished the growth of C. albicans was considered the MFC.

Checkerboard assay.

A checkerboard assay was performed according to methods described previously by Tobudic et al. (4). C. albicans planktonic cultures were inoculated into round-bottomed 96-well polystyrene plates and subsequently supplemented with either OlPC, caspofungin, or both. Control wells contained only RPMI 1640 medium. Plates were incubated for 48 h at 37°C. Here, plates were read visually, and the content of each well was plated onto YPD plates and incubated at 37°C overnight. Interpretation of the interactions of the drug combination against C. albicans planktonic cell growth was done based on the FICI (26). The FICI was calculated by the formula FICI = (MICA_[combined]/MICA_[alone]) + (MICB_[combined]/MICB_[alone]), in which MICA_[combined] and MICB_[combined] are the MICs of agents in combination and MICA_[alone] and MICB_[alone] are the MICs of drugs A and B alone, respectively. An FICI of ≤0.5 was defined as a synergistic interaction, an FICI of 0.5 to 4 was defined as no interaction, and an FICI of >4.0 was defined as an antagonistic interaction.

In vitro biofilm susceptibility testing.

In vitro C. albicans biofilm drug susceptibility testing was performed on serum-coated polyurethane catheter fragments. The effect of OlPC on biofilms was tested in three setups differing in the developmental stages, namely, during the periods of adhesion and biofilm development and on preformed 24-h-old biofilms. For studies of the effect of OlPC on C. albicans adhesion (90 min at 37°C), the polyurethane devices were submerged in 1 ml Candida cells (5 × 10⁴ cells/ml) dissolved in medium (control sample) or in the presence of different concentrations of OlPC. After this, nonattached cells were removed by washing steps. The amount of attached cells per catheter piece was quantified by CFU counting as described previously by Řičicová et al. (18). For assessment of the efficacy of OlPC against biofilm development, the adhesion period was induced in drug-free medium. After 90 min, catheters were washed with phosphate-buffered saline (PBS) and submerged in fresh RPMI 1640 medium (control sample) or supplemented with OlPC. Biofilms were allowed to develop for 24 h at 37°C, after which the catheters were washed twice with PBS, and the number of biofilm-forming cells was quantified as CFU. For assessment of the effect of OlPC on preformed biofilms, Candida cells were allowed to form 24-h-old biofilms in RPMI 1640 medium only. After 24 h, biofilms were washed twice with PBS and submerged in fresh RPMI 1640 medium (control sample) or in medium containing different concentrations of OlPC. Catheters were further incubated for 24 h at 37°C. Afterwards, catheters were washed and transferred to microcentrifuge tubes containing 1 ml of PBS. The efficacy of OlPC against preformed biofilms was documented by CFU determination.

The effect of caspofungin on the formation of preformed C. albicans biofilms on polyurethane fragments was also studied. Twenty-four-hour-old biofilms were formed in drug-free RPMI 1640 medium as described above. After washing, catheters were supplemented with clean RPMI 1640 medium (control sample) or in medium containing caspofungin for the next 24 h at 37°C. The effect of caspofungin on Candida biofilms was also assessed by CFU determination.

In vivo biofilm drug susceptibility testing using a rat subcutaneous biofilm model.

In vivo C. albicans SC5314 biofilms were formed inside serum-coated polyurethane catheter pieces in immunosuppressed female Sprague-Dawley rats as described previously by Řičicová et al. (18). In vivo C. albicans biofilm treatment was initiated after 24 h of biofilm formation. OlPC was administered orally, once daily, at a concentration of 20 mg/kg of body weight/day for 7 days. Sterile water was administered orally to the control group of animals. On day 9, catheters were explanted from euthanized animals, and the number of cells per individual biofilm was determined as CFU (18).

Ex vivo model of infection.

Experiments using an ex vivo model of infection were performed according to methods described previously by Peters et al. (27), with small modifications. Tongues excised from sacrificed 8-week-old female BALB/c mice (Janvier, France) were placed into a 24-well tissue culture plate and infected with C. albicans SC5314 (1 × 10⁷ cells/ml) during 30 min of incubation at 37°C with gentle shaking (120 rpm). Afterwards, tongues were supplemented with fresh RPMI 1640 medium (control) or with medium containing OlPC, and they were further incubated for 24 h at 37°C with gentle shaking (120 rpm). Following washing with PBS, the tongues were transferred to tubes containing PBS. Half of the harvested tongues were fixed and processed for histopathology, whereas the other half were homogenized and sonicated for 10 min in a water bath sonicator. Series of dilutions were plated onto YPD agar plates, and the CFU were determined.

Yeast-to-hypha transition.

Cultures of Candida cells grown overnight were washed twice with water and adjusted to a final concentration of 1 × 10⁵ cells/ml in either RPMI 1640 medium (pH 7.0), Spider medium (28), or Lee medium (29) with 0.5, 1, or 2 mg/liter of OlPC. Samples supplemented with water were considered controls. Cultures were incubated for 4 h and 24 h at 37°C. The proportion of true hyphae versus budding yeasts was determined by light microscopy (Axiostar plus; Carl Zeiss, Germany) at a magnification of ×60.

Scanning electron microscopy.

A qualitative analysis of the biofilm architecture was performed by SEM (XL30-FEG; FEI). Prior to SEM, samples were coated with Au-Pd by using a sputtering device (Edwards S150). The scanning electron microscope was operated with standard high-vacuum settings using a 10-mm working distance and a 20-keV accelerating voltage.

Confocal laser scanning microscopy.

In vivo C. albicans biofilms were analyzed by CSLM (LSM510/ConfoCor2 system; Carl Zeiss, Jena, Germany). Catheters were stained with concanavalin A (50 mg/liter) (Molecular Probes, Eugene, OR, USA) at 37°C for 1 h (green fluorescence). Before biofilm thickness was assessed, three-dimensional images of biofilms were captured. Approximately 150 sections were made through the whole biofilm architecture. Next, biofilm thickness was estimated from the outer edges of the area where the fluorescent signal gain showed maximum intensity to the area where no fluorescent signal could be detected.

Statistical analyses and reproducibility of results.

Statistical analyses were performed by using Student's t test (GraphPad Prism software). Differences were considered significant at a P value of ≤0.05. All in vitro experiments were performed at least three times, always using three devices per concentration. The experiment with the ex vivo model of infection was repeated twice, including at least 3 tongues/concentration tested. In vivo experiments were performed three times (in total, 8 animals/group). Experiments related to the yeast-to-hypha transition were repeated three times.