In Vitro Antifungal Activity of Sertraline and Synergistic Effects in Combination with Antifungal Drugs against Planktonic Forms and Biofilms of Clinical Trichosporon asahii Isolates

Trichosporon asahii (T. asahii) is the major pathogen of invasive trichosporonosis which occurred mostly in immunocompromised patients. The biofilms formation ability of T. asahii may account for resistance to antifungal drugs and results a high mortality rate. Sertraline, a commonly prescribed antidepressant, has been demonstrated to show in vitro and in vivo antifungal activities against many kinds of pathogenic fungi, especially Cryptococcus species. In the present study, the in vitro activities of sertraline alone or combined with fluconazole, voriconazole, itraconazole, caspofungin and amphotericin B against planktonic forms and biofilms of 21 clinical T. asahii isolates were evaluated using broth microdilution checkerboard method and XTT reduction assay, respectively. The fractional inhibitory concentration index (FICI) was used to interpret drug interactions. Sertraline alone exhibited antifungal activities against both T. asahii planktonic cells (MICs, 4–8 μg/ml) and T. asahii biofilms (SMICs, 16–32 μg/ml). Furthermore, SRT exhibited synergistic effects against T. asahii planktonic cells in combination with amphotericin B, caspofungin or fluconazole (FICI≤0.5) and exhibited synergistic effects against T. asahii biofilms in combination with amphotericin B (FICI≤0.5). SRT exhibited mostly indifferent interactions against T. asahii biofilms in combination with three azoles in this study. Sertraline-amphotericin B combination showed the highest percentage of synergistic effects against both T. asahii planktonic cells (90.5%) and T. asahii biofilms (81.0%). No antagonistic interaction was observed. Our study suggests the therapeutic potential of sertraline against invasive T. asahii infection, especially catheter-related T. asahii infection. Further in vivo studies are needed to validate our findings.


Introduction
Trichosporon asahii (T. asahii) is an opportunistic pathogen which belongs to the member of basidiomycete yeast-like fungi and can cause invasive trichosporonosis in immunocompromised patients [1].
The incidence of invasive trichosporonosis has been increased over the past 4 decades with the increased immunocompromised population, mainly those with hematological malignant diseases, AIDS patients and organ transplant recipients [2]. Additional risk factors include the use of corticosteroid, chemotherapy, as well as the use of medical implanted devices [2].
Various antifungal drugs have been used in the treatment of invasive trichosporonosis, including polyenes (such as amphotericin B), echinocandins (such as caspofungin) and the azoles (such as fluconazole, itraconazole and voriconazole). However, T. asahii often cause breakthrough infections in patients treated with AMB or echinocandins [3][4][5].
In a clinical guideline for the diagnosis and management of rare invasive yeast infection (including Trichosporon species), amphotericin B monotherapy is not recommended for invasive trichosporonosis, because of its limited in vitro activity against T. asahii (MICs!2 mg/L) and poor response rates (between 16% and 24%) to trichosporonosis [6]. Echinocandins are also not recommended for treating invasive trichosporonosis since Trichosporon spp. is intrinsic resistant to this antifungal drug class [1,6].
The newer triazoles (such as voriconazole) are now considered to be the most effective drugs class for invasive trichosporonosis treatment because they exhibit good in vitro and in vivo activity against Trichosporon spp. and result good clinical outcome [1,6]. However, the high cost of new triazoles impedes their widespread use in China. Furthermore, since azoles are all fungistatic, the sustained use of azoles antifungal drugs may result in drug-resistance, especially when used as low-dose prophylactic/empirical therapy. Actually, decreased susceptibility of T. asahii to azoles has been reported and multidrug-resistant Trichosporon strains have already been isolated [7,8].
Invasive T. asahii infections are usually associated with the use of medical implanted devices (such as central venous catheters, vesical catheters, and peritoneal catheter-related devices) [1]. The ability of T. asahii to form biofilms on medical implanted devices may account for the clinical resistance to antifungal drugs and results a high mortality rate. Although the newer triazoles have been demonstrated to show excellent in vitro activity against T. asahii planktonic cells, they have been reported failing to eradicate T. asahii biofilms and may result in treatment failure [9,10]. Thus, in the views of drug-resistance and pharmacoeconomics, it is necessary to develop new therapeutic approach against T. asahii infection. To our knowledge, combination of traditional antifungal drugs with non-antifungal agents has been proposed to be a promising strategy to cope with resistant fungal infections [11,12], this antifungal strategy may also be beneficial to cope with resistant T. asahii infections.
Sertraline (SRT) is a commonly prescribed antidepressant that belongs to the group of selective serotonin reuptake inhibitors [13]. It has been demonstrated that SRT exhibit antifungal activities against Candida spp., Aspergillus spp. and Cryptococcus species [14][15][16][17][18]. SRT has also been demonstrated to show in vitro synergistic effects in combination with antifungal drugs against Aspergillus spp. and Cryptococcus neoformans (C. neoformans) [19][20][21]. Furthermore, SRT was demonstrated to exhibit adjunctive antifungal effect against HIV-associated cryptococcal meningitis clinically [22]. Considering that Trichosporon spp. is phylogenetically closed to Cryptococcus species, we wonder if SRT has similar antifungal activity and synergistic effect against T. asahii. To our knowledge, no studies have been conducted on the antifungal activity of SRT against Trichosporon species.
In the present study, the in vitro antifungal activities of SRT alone or in combination with clinical commonly used antifungal drugs against planktonic forms of 21 clinical T. asahii isolates were examined by a broth microdilution checkerboard method based on M27-A3 reference method documented by Clinical and Laboratory Standards Institute (CLSI) [23]. In vitro anti-biofilms activities of SRT alone or in combination with antifungal drugs were examined by a XTT reduction assay. The results of our in vitro antifungal susceptibility testing against T. asahii may be helpful to evaluate the possible application of SRT in treating T. asahii infections.
All strains were removed from -80˚C freezer, and then subcultured twice on Sabouraud dextrose agar (SDA, Merck KGaA, Darmstadt, Germany) at 35˚C for 24 to 48 h to ensure purity and viability. The subcultures were further cultured overnight in yeast peptone dextrose (YPD, Oxoid Limited, England) liquid medium at 37˚C in a rotating incubator at 130 rpm. Following growth, the cells were harvested by centrifugation and washed twice with sterile phosphate-buffered saline (PBS). The cells were resuspended in RPMI 1640 medium, which has been adjusted to pH 7.0 with 0.165 M morpholinepropanesulfonic acid (MOPS) (Sigma-Aldrich, St Louis, MO, USA) to the densities of 10 3 CFU/ml for the in vitro susceptibility testing against planktonic cells and 10 6 CFU/ml for the in vitro anti-biofilms susceptibility testing. Candida parapsilosis ATCC 22019 was included as quality control strain for our in vitro susceptibility testing.

In vitro antifungal susceptibility testing against T. asahii planktonic cells
In vitro activities of FLC, ITC, VRC, CAS, AMB or SRT alone, and combinations of SRT with antifungal drugs against T. asahii planktonic cells were evaluated by using the broth microdilution checkerboard method based on the M27-A3 reference method (CLSI, USA) [23]. All tested drugs were distributed in 96-well microtitre plates. The final drug concentrations ranged from 0.062 to 64 μg/ml for FLC, from 0.008 to 8 μg/ml for ITC and AMB, from 0.001 to 1 μg/ml for VRC, from 0.031 to 32 μg/ml for CAS, and from 0.5 to 32 μg/ml for SRT. T. asahii cells suspensions were adjusted to a 0.5 McFarland standard transmittance at 530 nm wavelength. After that, the final inoculums of T. asahii were approximately 1.0×10 3 −3.0×10 3 CFU/ ml in each well after a serial dilution with RPMI 1640 broth medium. The plates were then incubated at 35˚C for 48 h. Thereafter, the minimum inhibitory concentrations (MICs) were recorded according to M27-A3 guideline [23]. The MICs of FLC, ITC VRC and CAS were defined as 50% reduction in turbidity compared to the growth control wells and the MIC for AMB was defined as complete inhibition of growth. To investigate the possible fungicidal activity of SRT, both MIC-2 (50% reduction in turbidity compared to the growth control well) and MIC-0 (complete inhibition of growth) endpoints were used for SRT in this study. The MIC-2 endpoint was also used for AMB to allow the antifungal combinations susceptibility testing to be comparable between all tested drugs. The MICs that inhibited 50 and 90% of the total isolates were defined as MIC 50 and MIC 90 , respectively. RPMI 1640 medium without T. asahii cells, as well as drug-free medium containing T. asahii cells, was used as negative and positive controls respectively. Experiments were repeated three times on different days.

Biofilms formation of T. asahii and in vitro anti-biofilms susceptibility testing
The biofilms formation of T. asahii was performed by a simple and reproducible 96-well plates-based method as previously described [24]. Briefly, 100 μl adjusted T. asahii suspension (10 6 CFU/ml) was added to 96-well plates. The wells containing RPMI 1640-MOPS medium without T. asahii cells were included as background controls. After 1 h of incubation (adhesion phase) at 37˚C, each well was washed twice gently with sterile PBS to remove non-adherent cells, 200 μl of fresh RPMI 1640-MOPS medium was then added to each well and the plates were further incubated at 37˚C for 24 h.
According to previously studies [10,23], the sessile MICs (SMICs) were defined as the lowest concentration capable of decreasing 50% in absorbance compared to the growth control wells measured by XTT reduction assay. Experiments were repeated three times on different days.

Drug interaction analysis
Drug combination interaction was evaluated on the basis of the fractional inhibitory concentration index (FICI) which is the sum of the fractional inhibitory concentration (FIC) of each drug [24]. The drug interaction was defined as the following: FICI 0.5, synergism; FICI > 0.5 to 4.0, indifference; FICI > 4.0, antagonism. The FICI values were calculated based on the MIC-2 endpoint for all drugs.

Results
To our knowledge, no standard interpretive breakpoints are available for in vitro antifungal susceptibility testing against Trichosporon species. However, the breakpoints for Candida species have been used in the in vitro antifungal susceptibility testing against Trichosporon isolates previously [25,26]. Thus, the reference breakpoints for Candida species were cautiously used to interpret results obtained in this study. To evaluate drug interactions, we compared the MICs of each antifungal drug alone to the MICs of antifungal combinations with SRT and calculated the FICIs of different antifungal combination.
The MIC range, geometric mean (GM) and the MICs for 50% or 90% of the isolates (MIC 50 /MIC 90 ) for T. asahii planktonic cells are present in Table 1. The in vitro antifungal susceptibility results by using both MIC-0 and MIC-2 endpoints on the anti-T. asahii activities of AMB and SRT are present in Table 2. The interactions of each antifungal combination against T. asahii planktonic cells are present in Table 3. The SMIC range, GM and the interactions of each antifungal combination against T. asahii biofilms are present in Table 4.
The MIC-2 and MIC-0 ranges for SRT were 4-8 μg/ml and 8-32 μg/ml, respectively. The MIC 50 /MIC 90 for SRT by using MIC-2 endpoint and MIC-0 endpoint were 4/8 μg/ml and   The results of our in vitro anti-biofilms susceptibility testing showed that the SMICs of most antifungal drugs against T. asahii biofilms increased up to 1000 times compared to the MICs against planktonic cells except for CAS, whose SMICs increased only 2-4 folds. The SMICs (16-32 μg/ml) of SRT against T. asahii biofilms increased 4 folds compared to the MICs (MIC-2, 4-8 μg/ml) against planktonic cells, which indicated the decreased susceptibility of SRT to T. asahii biofilms.

Discussion
T. asahii is the major pathogen of invasive trichosporonosis which occurred mainly in immunocompromised patients. Compared to the low incidence rates of invasive trichosporonosis, invasive Trichosporon infections leads to high mortality rates up to 80% despite treated with antifungal drugs [1].
To date, treatment of invasive trichosporonosis remains a challenge. Most of the in vitro antifungal susceptibility tests demonstrated high MICs of AMB and echinocandins to T. asahii and indicated drug resistance. In contrast, azoles antifungal drugs, especially the newer triazoles, are the primary drug class for the treatment of invasive trichosporonosis based on available data. However, decreases of the susceptibility of T. asahii to azoles have been reported, including the newer triazoles [7,8].
As expected, VRC (MICs, 0.031-0.25 μg/ml) was the most effective drug against T. asahii planktonic cells in this study. No significant high MICs of three azoles were observed against most T. asahii isolates in this study. ITC (MICs, 0.25-1 μg/ml) and FLC (MICs, 1-16 μg/ml) were still sensitive to most T. asahii isolates. All tested T. asahii isolates were resistant to AMB (MICs ! 2 μg/ml) and CAS (MICs ! 8 μg/ml). Our results were in agreement with previous data in China of the in vitro antifungal susceptibility of VRC, CAS and AMB against clinical T. asahii isolates [27].
Previous antifungal susceptibility assay demonstrated a remarkable rise in the sessile MICs of azoles against T. asahii biofilms (SMIC>1024 μg/ml) compared to the MICs of planktonic cells and T. asahii biofilms were up to 16000 times more resistant to VRC than planktonic cells [28]. In agreement with previous reports, T. asahii biofilms were resistant to all three azoles tested in this study, as the SMICs were up to 1000 times higher than the MICs of T. asahii planktonic cells. T. asahii biofilms were more resistant to AMB than planktonic cells.
As is well known, T. asahii is intrinsic resistant to echinocandins [1,2]. However, CAS was observed to inhibit T. asahii biofilms at the final concentrations from 16 to 64 μg/ml in this study, and the SMICs of CAS increased only 2-4 folds compared to the MICs of planktonic cells. The inhibitory effect on the synthesis of β-(1,3)-glucan of the fungal cell wall is believed to be one of the mechanisms of CAS to exert anti-biofilms effects against C. albicans biofilms, since β-(1,3)-glucan is considered to be a major component of fungal biofilms [29,30]. Thus, the inhibitory effect on the synthesis of β-(1,3)-glucan may also account for the anti-biofilms effects of CAS against T. asahii in this study. The possible inhibitory effects of CAS against T. asahii biofilms needs further studies to be validated.
Based on a review of 185 reported cases from 1975 to 2014, Trichosporon fungemia, including catheter-related fungemia, represents the main type of invasive Trichosporon infection [2]. However, disseminated trichosporonosis can involve most human organs and results in pneumonia, endocarditis, brain abscess, meningitis, arthritis, esophagitis, lymphadenopathy, liver infection, splenic abscess, uterine infection and soft tissue infection [1]. Invasive trichosporonosis are usually associated with the use of medical implanted devices [1,2]. Peritoneal dialysis can cause fungal peritonitis due to Trichosporon species [1]. Endocarditis due to Trichosporon spp. in cardiac valve replacement patients has been increasingly reported [1]. Urinary tract infections and renal dysfunction caused by Trichosporon spp. have also been reported, especially in patients with vesical catheterization [1]. Removal of infected catheters may increase the efficacy of antifungal drugs and improve clinical outcomes. Unfortunately, most severe patients are catheter-dependent and it is a life-threaten matter to remove medical implanted catheters [31]. Thus, the development of new antifungal agents with antifungal activity against T. asahii biolfilms is necessary.
SRT, a commonly prescribed psychotropic drug, was selected as a potential antifungal agent against T. asahii in this study based on its reported in vitro and in vivo fungicidal activity, low toxicity and lack of drug interactions [14][15][16][17][18][19][20][21][22]32]. SRT was firstly reported to show antifungal activity against Candida species in 2001 [14]. Lass-Flörl et al found that patients with recurrent vulvovaginal candidiasis were cured when treated with SRT for accompanying premenstrual dysphoric disorder [14]. From then on, the antifungal activities of SRT against Candida spp., Aspergillus spp. and Cryptococcus spp. have been extensively discussed [14][15][16][17][18][19][20][21]. The antifungal activity of SRT against C. neoformans has also been demonstrated in animal model studies [16,17]. SRT was demonstrated to reduce fungal burden in the brain, kidney and spleen in murine models of systemic cryptococcosis at clinically relevant concentrations [16,17]. More important, the antifungal effect of SRT against Cryptococcus infection has been demonstrated clinically [22]. Thus, SRT was speculated to have similar antifungal effect against T. asahii, which is phylogenetically closed to C. neoformans. Considering the important role of fungal biofilms in drug-resistance, the in vitro antifungal activities of SRT against both T. asahii planktonic cells and biofilms were evaluated in this study.
Our study demonstrated that SRT was fungicidal in high concentrations (MIC-0, 8-32 μg/ ml). The MIC 90 for SRT by using both MIC-2 and MIC-0 endpoints indicated that SRT could inhibit 90% of the total T. asahii isolates in concentration of 8 μg/ml and kill 90% of the total isolates in concentration of 32 μg/ml. The SMICs (16-32 μg/ml) of SRT demonstrated its inhibitory activity against T. asahii biofilms in concentrations higher than 16 μg/ml. Our results demonstrated that SRT exhibited fungicidal activity against T. asahii planktonic cells and inhibitory activity against T. asahii biofilms in relative high concentrations.
To evaluate the clinical therapeutic potential of SRT on invasive T. asahii infections, we compared the pharmacokinetic data of SRT with our in vitro antifungal susceptibility data. The MICs (MIC-2 ranges, 4-8 μg/ml; MIC-0 ranges, 8-32 μg/ml) against T. asahii are much higher than the reported blood concentrations of SRT (55-250 ng/ml) [13,33]. Pharmacokinetic studies of SRT also demonstrated that the concentrations of SRT in the brain were 20-50 times higher than blood concentrations [33]. Furthermore, the concentrations of SRT in the eyes, heart, lung, spleen, liver, kidney, stomach, small intestine, muscle and skin were also demonstrated to be much higher than the blood concentrations, although the organ/blood concentration ratios for these organs have not been assayed statistically [33,34]. Thus, the much higher concentrations in tissues and organs of SRT may be beneficial for treating disseminated trichosporonosis since T. asahii often disseminated to most organs of patients [1].
Although it is difficult to achieve high blood concentrations of SRT administrated orally, it is possible that the high concentrations may be attainable in catheters by way of intra-luminal lock therapy. Antifungal lock therapy is to use high local concentrations of antimicrobial agents within an infected catheter in attempt to sterilize the catheters [35]. The therapeutic potential of ethanol as lock therapy against T. asahii infection has been demonstrated by our research group previously [36]. Based on the anti-biofilm activity of SRT against T. asahii observed in this study, SRT may be used as a lock strategy with high local concentrations acting on infected catheters and may facilitate the clearance of T. asahii biofilms and improve clinical outcomes.
In addition to the antifungal activity of SRT alone, it has also been demonstrated that SRT exhibited in vitro synergistic effects in combination with antifungal drugs against Aspergillus spp. and C. neoformans. SRT was demonstrated to enhance the activity of AMB against Aspergillus spp. [19] and was also demonstrated to exhibit in vitro and in vivo synergistic effect in combination with FLC against C. neoformans [16,20]. SRT also exhibited in vitro synergistic effects combined with AMB against C. neoformans in another study [21]. Based on these previous studies, the possible synergistic effect of SRT in combination with antifungal drugs against both T. asahii planktonic cells and biofilms were further evaluated in this study.
Our results demonstrated that SRT indeed exhibited synergistic effects when combined with AMB, CAS or FLC against T. asahii planktonic cells. The combinations of SRT-AMB (90.5%), SRT-CAS (81.0%) and SRT-FLC (61.9%) yielded potent synergistic effects. In our anti-biolilms combinations study, the combination of SRT-AMB also showed the highest percentage of synergistic effects (81.0%). In contrast, SRT exhibited mostly indifferent interactions in combinations with three azoles. The SRT-CAS combination (47.6%) yielded relative lower synergistic effects against T. asahii biofilms compared to that of SRT-AMB.
Our antifungal combinations study highlights the therapeutic potential of SRT-AMB combination for T. asahii infection, since the SRT-AMB combination yielded highest percentage of synergistic effects against both T. asahii planktonic cells and biofilms. As is well known, AMB is a fungicidal drug with high toxic effect. Based on our results and previous data [14][15][16][17][18], SRT is also fungicidal. Thus, the SRT-AMB combination therapy may result a better therapeutic efficacy with reduced toxicity of AMB. Furthermore, The SRT-AMB combination may be beneficial for reducing the emergence of drug-resistance. Thus, the anti-biofilms effect of SRT alone and SRT-AMB combination on T. asahii highlights the potential utility of SRT or SRT-AMB combination on invasive T. asahii infections, especially suitable for the patients with medical implanted devices.
The antifungal mechanisms of SRT are not investigated in this study. However, some possible antifungal mechanisms of SRT have been discussed by different research group. A genetic study suggests that SRT may exert antifungal effect by perturbing translation and inhibiting protein synthesis of fungi [16]. Rainey et al. demonstrated that SRT may exhibit antifungal activity by targeting intracellular vesiculogenic phospholipid membranes in fungi [37]. Another study demonstrated that SRT can perturb membrane permeability and inhibit sphingolipid biosynthesis in fungi [38]. These reported antifungal mechanisms of SRT may also account for the antifungal activity against T. asahii in this study.
The antifungal synergistic mechanisms have also been discussed previously [22]. The different antifungal mechanisms of FLC (inhibiting ergosterol synthesis) and SRT (inhibiting mRNA translation into protein synthesis) may account for their synergistic antifungal effects [22]. As is well known, AMB exerts antifungal effect by binding with ergosterol, forming channels in fungal cell membranes that cause rapid leakage of cell contents and subsequent fungal cell death. The different antifungal mechanisms of AMB and SRT may also account for their synergistic anti-T. asahii effects observed in this study.
In summary, our study demonstrates the in vitro antifungal activities of SRT on both T. asahii planktonic cells and biofilms and highlights the therapeutic potential of SRT against invasive T. asahii infections, especially suitable for the patients with catheter-related fungal infections. The use of the SRT-AMB combination therapy may be advantageous in treating T. asahii infection based on their obviously synergistic effects. The anti-biofilms activity of SRT against T. asahii may be helpful to control biofilms-related fungal infection, not only for T. asahii, but also for other pathogenic fungi (such as C. neoformans). Considering the clinical T. asahii isolates used in this study were mainly from China, the in vitro fungicidal activity and anti-biofilms activity of SRT against T. asahii should be confirmed by researchers outside China. Lack of in vivo anti-T. asahii data of SRT is a limitation of this study. Further animal models and clinical trials are needed to validate the correlation of our findings. The precise antifungal mechanisms of SRT are also worthy to be investigated.