Synergistic Effects and Mechanisms of Budesonide in Combination with Fluconazole against Resistant Candida albicans

Candida albicans is an important opportunistic pathogen, causing both superficial mucosal infections and life-threatening systemic diseases in the clinic. The emergence of drug resistance in Candida albicans has become a noteworthy phenomenon due to the extensive use of antifungal agents and the development of biofilms. This study showed that budesonide potentiates the antifungal effect of fluconazole against fluconazole-resistant Candida albicans strains both in vitro and in vivo. In addition, our results demonstrated, for the first time, that the combination of fluconazole and budesonide can reverse the resistance of Candida albicans by inhibiting the function of drug transporters, reducing the formation of biofilms, promoting apoptosis and inhibiting the activity of extracellular phospholipases. This is the first study implicating the effects and mechanisms of budesonide against Candida albicans alone or in combination with fluconazole, which may ultimately lead to the identification of new potential antifungal targets.


Introduction
Globally, Candida albicans (C. albicans) is the leading human fungal pathogen, and the incidence of invasive fungal infections caused by C. albicans is rising, along with an increasing mortality rate [1][2][3]. Fluconazole (FLC), one of the triazole antifungal agents, is the most widely used antifungal agents to treat candidiasis infections due to its high bioavailability and low toxicity [4][5][6]. However, the excessive and indiscriminate use of FLC in the clinic has led to the emergence of drug resistance in C. albicans [7][8][9]. Biofilms formed on medical devices not only provide protection from environmental stress for C. albicans but also enhance the resistance of C. albicans to antifungal agents by up to 1000-fold greater than that needed for the treatment of its planktonic counterparts [10][11][12]. Therefore, it is necessary to develop new antifungal strategies that can eliminate the phenomenon of drug resistance in C. albicans. The discovery of new antifungal agents is a complicated and long-term process [13,14]. To develop new approaches for the treatment of invasive fungal infections, combination therapy may be a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 an ideal approach. It could significantly improve the antifungal efficacy of a drug compared to each drug used alone and reduce adverse side effects from the drugs used. In recent years, efforts have been made to overcome the resistance of C. albicans by using drug combinations [15,16].
Glucocorticoids usually are not used in general clinical anti-infection therapy in consideration of their immunosuppressive actions, and they are only used in severe infections and septic shock for their strong anti-inflammatory effects. We found, however, that budesonide (BUD), an inhaled corticosteroid, can enhance the antifungal effect of FLC against resistant C. albicans in vitro.
In the present report, we first investigated the in vitro effects of BUD combined with FLC against resistant C. albicans and susceptible C. albicans by a microdilution checkerboard method. Galleria mellonella (G. mellonella), a new fungal infection model, was used to study the effects of the combined treatment on C. albicans in vivo. The results in vitro and in vivo strongly revealed that the combination of FLC and BUD has synergistic effects against resistant C. albicans. Then, we further investigated the molecular mechanisms of BUD for enhancing the susceptibility of resistant C. albicans to FLC by assaying its impact on the function of drug transporters, biofilm formation, the induction of apoptosis and the activity of extracellular phospholipases, as well as assaying the expression levels of the genes related to these physiological processes in C. albicans. Our results demonstrated that the combination of FLC and BUD can reverse the resistance of C. albicans by inhibiting the function of drug transporters, reducing the formation of biofilms, promoting apoptosis and inhibiting the activity of extracellular phospholipases. We believe that researchers in the field of C. albicans may benefit from the conclusions of our studies.

Strains and growth medium
Two resistant C. albicans strains: C. albicans 10 (CA10) and C. albicans 16 (CA16) and two susceptible C. albicans strains: C. albicans 4 (CA4) and C. albicans 8 (CA8) were used, and C. albicans ATCC10231 was used as a quality control. CA4, CA8, CA10 and CA16 were provided by clinical laboratory in Qianfoshan Hospital, Jinan, China. ATCC10231 strain was generously provided by the Department of Pharmacology, School of Pharmaceutical Sciences, Shandong University, Jinan, Shandong Province, Jinan, China. All strains were refreshed from frozen storage at -80˚C and subcultured on yeast-peptone-dextrose (YPD) solid medium (1% yeast extract, 2% peptone, 2% glucose and 2% agar) overnight at 35˚C at least twice. Before each experiment, all strains were inoculated in YPD liquid medium (1% yeast extract, 2% peptone and 2% glucose) for 24 h at 35˚C.

Drugs and G. mellonella larvae
All drugs (BUD, FLC and ampicillin) were purchased from Dalian Meilun Biotech Co., Ltd (Dalian, China). The stock solution of BUD (12800 μg/mL) was prepared in ethyl alcohol absolute. Stock solutions of FLC (2560 μg/mL) and ampicillin (2560 μg/mL) were prepared in distilled water. All stock solutions were stored at -20˚C until use. G. mellonella larvae were purchased from Tianjin Huiyude Biotech Co., Ltd (Tianjin, China). Randomly chosen G. mellonella larvae (approximately 0.25 g) in the final instar larval stage were used in our experiments. G. mellonella larvae were cleaned with 70% ethanol and incubated in the dark at 37˚C overnight prior to each experiment.

Determination of MICs of planktonic cells in vitro
Two resistant C. albicans strains (CA10 and CA16) and two susceptible C. albicans strains (CA4 and CA8) were used in this experiment. The antifungal effects of BUD in combination with FLC on C. albicans planktonic cells were evaluated using a checkerboard broth microdilution method in 96-well plates according to the protocol of the Clinical and Laboratory Standards Institute (CLSI, 2008 M27-A3). Drugs were added to the wells and serially diluted to final concentrations ranging from 0.125 μg/mL to 64 μg/mL for FLC and from 2 μg/mL to 128 μg/mL for BUD. 100μL C. albicans suspension at a final concentration of 2.5×10 3 colony forming units (CFU)/mL was subsequently added to each well. A well containing 100 μL of C. albicans suspension and 100 μL of RPMI 1640 medium was included to serve as a drug-free control. The 12th column of wells on the plate was filled with only 200 μL of RPMI 1640 medium to act as negative background controls during subsequent analysis and quantification. Ninety-six-well plates were then incubated at 35˚C for 24 h. A 2,3-bis-(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxanilide (XTT) reduction assay was performed to determine the minimum inhibitory concentrations (MICs). The MICs are defined as the lowest antifungal concentrations at which there is an 80% (MIC 80 ) reduction in XTT-colorimetric readings in comparison to the drug-free control. Colorimetric changes were measured at 492 nm with a microtiter plate reader (SpectraMax190, USA). All experiments were performed three times on different days.

4. Determining the antifungal effects in vivo
G. mellonella survival assay. The G. mellonella infection model was performed as we described previously [16] with minor modifications, and the FLC-resistant isolate CA10 was used in this experiment. Briefly, 16 randomly chosen G. mellonella larvae were included in each group, and four groups were used in this experiment. C. albicans inoculum was suspended in ampicillin-PBS containing 2 μg/mL ampicillin to avoid bacterial interference. C. albicans inoculum at a concentration of 5×10 8 CFU/mL was selected to monitor virulence. A 50-μL micro syringe (Shanghai Gaoge Biotech Co., Ltd. China) was used to inject 10 μL C. albicans inoculum (5×10 6 CFU/larva) into the G. mellonella larval hemocoel via the last right proleg. After 2 h, G. mellonella larvae in four groups were injected with 10 μL PBS, FLC (160 μg/ mL), BUD (160 μg/mL) or with a combination of FLC and BUD at the same concentrations. G. mellonella larvae injected with PBS were used as controls to control for the impact of physical trauma. After injection, G. mellonella larvae were incubated in sterile petri dishes at 37˚C for 4 days, and larval death was monitored by visual inspection every day. A G. mellonella larva is considered dead when it displays no movement or proleg movement in response to touch [17]. Survival curves were analyzed by the Kaplan-Meier method using the Statistical Product and Service Solutions 19 software.
Fungal burden determination. The CFU counting method was used to determine the fungal burden of G. mellonella larvae infected with C. albicans CA10 [16]. G. mellonella larvae in four groups were treated with C. albicans inoculum and drugs as mentioned above. Three infected G. mellonella larvae were selected randomly from each group daily, washed with 70% ethanol and cut into small pieces with a scalpel. Pieces of each G. mellonella larva were put into a sterile tube containing 5 mL PBS-ampicillin and were mixed using a VORTEX GENIE 2 vibrator (Scientific Industries, USA). Then, different dilutions were made for each sample, and a 10-μL suspension from each dilution was placed on YPD solid medium dishes. The dishes were then incubated at 35˚C for 24 h, and the CFU was counted. Graph Pad Prisma 5 software was used to analyze the CFU. All experiments were performed three times on different days.
Histology. Histology was used to characterize the development of the infection in G. mellonella larvae and to determine the effect of drugs on the G. mellonella larvae infected by C. albicans CA10 [18]. G. mellonella larvaein four groups were treated with C. albicans inoculum and drugs as mentioned above. One infected G. mellonella larva was randomly selected from each group after 2 days of infection, washed with 70% ethanol and cut into tissue sections (7 μm) using a freezing microtome (LEICA CM 1950, Germany). The tissue sections were then stained with Periodic Acid Schiff reagent (PAS) and observed under an OLYMPUS BX51 microscope using a 10×eyepiece and a 10×objective.

5. Assaying the uptake and efflux of Rh6G
Rhodamine 6G (Rh6G) uptake and efflux in C. albicans CA10 cells were determined as previously described [15,19,20]. The cells were diluted to a final concentration of 1×10 5 CFU/ mL with YPD liquid medium and incubated at 35˚C. After 18-19 h, the cells were centrifuged, washed with glucose-free PBS (pH = 7.0) three times, and resuspended in glucose-free PBS to a concentration of 1×10 7 CFU/mL. The cells were de-energized for 1-2 h in glucose-free PBS and then resuspended in glucose-free PBS for further use. For the Rh6G uptake experiment, Rh6G (Sigma, USA) and BUD were added to the above-mentioned cells at a final concentration of 10 μM and 128 μg/mL, respectively. A drug-free sample with Rh6G served as a control. After incubation at 35˚C for 50 min, a 10 min of ice-water bath was used to stop cellular uptake of Rh6G, and then the extracellular Rh6G was removed using glucose-free PBS. The mean fluorescent intensity (MFI) of the remaining intracellular Rh6G was detected using a BD FAC-SAria II flow cytometer (Becton Dickinson, USA) with excitation at 488 nm and emission at 530 nm. For the Rh6G efflux experiment, Rh6G was first added to the above-mentioned cells at a final concentration of 10 μM. After the incubation at 35˚C for 50 min, a 10 min ice-water bath was used to stop Rh6G uptake into cells. Cells were then harvested, washed three times with glucose-free PBS to remove the extracellular Rh6G, and resuspended in PBS containing 5% glucose to initiate ATP-driven efflux. Meanwhile, BUD was added to the suspension at a concentration of 128 μg/mL. A drug-free sample with only Rh6G served as the control group. The MFI of the intracellular remaining Rh6G was detected every 30 min as mentioned above. Graph Pad Prisma 5 software was used to analyze the results. All experiments were performed three times on different days.

6. Assaying the sessile MICs of C. albicans biofilms
The effects of BUD in combination with FLC against the biofilms of C. albicans CA10 were tested in 96-well plates as Ramage et al. described previously [21]. Briefly, each well in the plate was filled with 200 μL of C. albicans cells at a final concentration of 2.5×10 3 CFU/mL, and the plates were incubated at 35˚C for 4, 8, 12 or 24 h to obtain biofilms at different stages of maturation. The 12th column of wells on the plate was filled with 200 μL RPMI 1640 medium as a negative background control for subsequent analysis and quantification. After incubation, planktonic and nonadherent cells were removed, and the biofilms were washed with sterile PBS three times. Drugs were serially diluted to final concentrations ranging from 2 μg/mL to 1024 μg/mL for FLC or from 8 μg/mL to 512 μg/mL for BUD and added to the biofilm-coated wells. A drug-free well was included to serve as a positive control. The plates were incubated for an additional 24 h at 35˚C. An XTT reduction assay was performed to determine the sessile minimum inhibitory concentrations (SMICs). The SMICs are defined as the lowest antifungal concentrations at which an 80% (SMIC 80 ) reduction in XTT-colorimetric readings is found in comparison to the drug-free control. Colorimetric changes were measured at 492 nm with a microtiter plate reader (SpectraMax190, USA). All experiments were performedthree times on different days.

7. Assaying the changes in reactive oxygen species
The endogenous reactive oxygen species (ROS) in C. albicans CA10 cells were measured using the dihydrorhodamine 123 (DHR 123, Sigma, USA) staining method [22]. Briefly, cells (1×10 6 CFU/mL) were incubated in YPD liquid medium overnight at 35˚C with drugs alone or in combination at the following final concentrations: FLC at 1 μg/mL; BUD at 128 μg/mL. A sample without the drug treatment was included as a control. After incubation, cells were centrifuged, washed with sterile PBS (pH = 7.0) three times, resuspended in sterile PBS to a concentration of 1×10 7 CFU/mL and stained with DHR 123 (10 μM) in the dark at 35˚C for 50 min. Finally, the MFI was detected using the BD FACSAria II flow cytometer (Becton Dickinson, USA) with excitation at 488 nm and emission at 530 nm. Graph Pad Prisma 5 software was used to analyze the results. All experiments were performed three times on different days.

8 Assaying phospholipase activity
The determination of extracellular phospholipase activity for C. albicans CA10 cells was performed according to the egg yolk agar method as previously described [23,24] with a few modifications. Cells (1×10 6 CFU/mL) were incubated in YPD liquid medium overnight at 35˚C with treatment using drugs alone or in combination at the following final concentrations: FLC at 1 μg/mL; BUD at 128 μg/mL. Samples without drug treatment were included as controls. After incubation, a 10 μL suspension of each sample was deposited on plates containing the egg yolk agar medium (1% peptone, 3% glucose, 5% NaCl, 0.0006% CaCl2, 2% agar and 10% egg yolk). Plates were then incubated at 35˚C for 48 h. The precipitation zone (Pz) value was calculated as follows [25]: Pz = Diameter of colony Diameter of the colony plus precipitation zone . The Pz value was then used to evaluate the phospholipase activity and it was defined as follows [16]: Pz = 1, negative phospholipase activity; Pz = 0.90-0.99, very low phospholipase activity; Pz = 0.80-0.89, low phospholipase activity; Pz = 0.70-0.79, high phospholipase activity; Pz 0.69, very high phospholipase activity. All experiments were performed three times on different days.

Real-time quantitative PCR
CA10 cells at a final concentration of 2.5× 10 5 CFU/mL were incubated in RPMI 1640 medium overnight at 35˚C with the treatment using drugs alone or in combination at the following final concentrations: FLC at 1 μg/mL; BUD at 128 μg/mL. A sample without drug treatment was included as a control. After incubation, the cells were harvested for RNA extraction. The total RNA was extracted using an RNApure Yeast Kit (CWBiotech, Beijing, China). The first strand of cDNA was synthesized by reverse transcription using a First-Stand cDNA Synthesis-SuperMix Kit (CWBiotech, Beijing, China). The thermal cycling condition was 95˚C for 10 min as an initial denaturation step, followed by 40 cycles of 95˚C for 15 s, 60˚C for 1 min and 72˚C for 32 s. The ACT1 gene was used as an endogenous reference control for normalization of the relative expression. Drug transporter encoding genes (CDR1, CDR2, MDR1 and FLU1), biofilm-related genes (EFG1, HWP1 and ALS1), apoptosis-related genes (MCA1, HSP90, CRZ1, CNA1), and virulence factor-related genes (PLB1, PLB2, PLB3, PLB4, PLB5, PLC1) were determined by the realtime-PCR (RT-PCR) assay as mentioned above. Sequences of the primers are listed in Table 1. The result was calculated using the 2 -(ΔΔCt) method [26]. Graph Pad Prisma 5 software was used to analyze the results. All experiments were performed three times on different days.

BUD increased the susceptibility of resistant C. albicans to FLC
To assess the nature and intensity of the in vitro interactions between FLC and BUD against C. albicans planktonic cells, the fractional inhibitory concentration index (FICI) model and the percentage of growth difference (4E) model were used in analyzing the data obtained from the checkerboard broth microdilution assays. The FIC index was calculated according to the following formula [27]: FICI = MIC A comb /MIC A alone +MIC B comb /MIC B alone. The interpretation of the FICI was as follows: FICI 0.5, synergistic effect; FICI > 0.5 but 4, no interaction; FICI > 4, antagonistic effect. As for the 4E model, interactions with<100% were considered weak, interactions with 100%-200% were considered moderate, and interactions with >200% were considered strong, as described previously [27]. Our results showed strong synergistic antifungal interactions between FLC and BUD against resistant C. albicans strains in vitro ( Table 2, S1 Table). No synergism was obtained with susceptible C. albicans strains (S1 Table). As shown in Table 2 and S1 3.2 BUD combined with FLC had a synergistic effect against resistant C. albicans in vivo Survival assay. In the present study, we compared the G. mellonella larvae survival rates of four groups to primarily evaluate the effects of the drug on fungal infected G. mellonella larvae. The survival rate for G. mellonella larvae treated with FLC or BUD alone was only a little higher than that of G. mellonella larvae treated with PBS (Fig 1, S2 Table). Notably, the survival rate of G. mellonella larvae was significantly higher when treated with the drugs in combination than that of G. mellonella larvae treated with PBS or the drug alone (P < 0.05), indicating that FLC and BUD have a synergistic antifungal effect on resistant C. albicans in vivo (Fig 1, S2 Table).
Fungal burden determination. Fungal burden determination was performed to evaluate the drug's effects on the tissue burden of C. albicans in G. mellonella larvae. Data showed that treatment with either FLC or BUD alone had almost no inhibitory effect, as there was a rapid proliferation of C. albicans within the G. mellonella larvae in these two groups over 4 days (Fig  2, S3 Table) and the recorded numbers of C. albicans in these two groups were similar to the control G. mellonella larvae. In contrast, reduction of the fungal burden was very significant in G. mellonella larvae treated with the combination of FLC and BUD (Fig 2, S3 Table).
Histology of larval tissue. To characterize the development of the fungal infection in G. mellonella larvae with the treatment of drugs, a histological study was performed by observing tissue sections after 2 days of infection. Drug combinations had a significant impact on the histology of infected G. mellonella larvae. The number of melanized nodules containing yeast cells and filaments in the larvae treated with FLC or BUD alone was almost the same as in the control G. mellonella larvae (Fig 3). Compared to untreated G. mellonella larvae or single drug treated G. mellonella larvae, a significant decrease in the number of melanized nodules was observed in G. mellonella larvae treated with a combination of FLC and BUD (Fig 3).

BUD increased the intracellular concentration of Rh6G by inhibiting drug transporter genes
Both FLC and Rh6G are the substrates of drug transporters in C. albicans. To investigate the effect of BUD on FLC uptake and efflux, we assayed the uptake and efflux of Rh6G in C. albicans   Table), indicating that BUD can promote the uptake of extracellular FLC. For Rh6G efflux, the MFI both in the BUD group and the control group gradually decreased with the increase in experimental time, but the MFI in the BUD group was significantly lower than that of the control group (Fig 4B, S5  Table), indicating that BUD can inhibit the efflux of intracellular FLC. Interestingly, synergistic inhibitory effects on the expression of CDR1, CDR2, MDR1 and FLU1 were observed in cells treated with a combination of FLC and BUD (Fig 5, S9 Table). Of note, both FLC and BUD alone failed to inhibit the expression levels of these four genes (Fig 5, S9 Table). Above all, the BUDinduced synergistic effect on enhancing the susceptibility of resistant C. albicans to FLC may be due to BUD-induced uptake of FLC into cells and inhibition of the efflux of intracellular FLC.

BUD combined with FLC had synergistic effects against biofilms at different stages
The antifungal effects of drugs on C. albicans biofilms were evaluated using the FICI model as mentioned above. As shown in Table 3 and S6 Table, BUD combined with FLC showed synergistic antifungal effects against 4, 8, 12 and 24 h biofilms with FICI < 0.5. In addition, the synergism for the 12 and 24 h biofilms (FICI = 0.252, 0.258) was weaker than that of the 4 and 8 h biofilms (FICI = 0.127). EFG1, HWP1, and ALS1 were closely involved in biofilm formation and development. Our data revealed that the expression levels of the biofilm-related genes EFG1, HWP1, and ALS1 in C. albicans were significantly down-regulated with the combination treatment of FLC and BUD (P < 0.05), as shown in Fig 6 and S9 Table. 3

.5 BUD combined with FLC induced apoptosis
Intracellular ROS in C. albicans was promoted, which was revealed by the significant fluorescence increases of DHR123 in the drug combination treatment group (Fig 7, S7 Table). The intensity of DHR123 fluorescence in the drug combination group increased by 9-, 2-and 6-fold compared with the control group, the FLC group and the BUD group, respectively (Fig  7, S7 Table). The up-regulation of MAC1 induced by the combination treatment revealed that BUD combined with FLC can promote apoptosis in C. albicans (Fig 8, S9 Table). Other apoptosis-related genes, including HSP90, CRZ1 and CNA1, were also reported to be involved in resistance, and their down-regulation demonstrated that BUD combined with FLC revises the resistance of C. albicans partly via inducing cellular apoptosis.

BUD combined with FLC inhibited the activity of extracellular phospholipase
Pz values were obtained by averaging the triplicates. There was almost no difference in the Pz values between drug alone groups (0.64 ± 0.02 for FLC and 0.62 ± 0.02 for BUD) and the control group (0.62 ± 0.02), as shown in Table 4. Notably, no precipitation zone was observed in the drug combination group (Table 4, S8 Table), indicating that treatment with BUD combined with FLC could completely inhibit the extracellular phospholipase activity of resistant C. albicans. As shown in Fig 9 and S9 Table, phospholipase-related genes PLB1-5 and PLC1 in C. albicans were significantly down-regulated by the combination treatment of BUD and FLC (P < 0.05). These results told us that BUD combined with FLC resulted in synergistic antifungal effects involved in inhibiting the activity of extracellular phospholipase.

Discussion
Our study showed that using such a combination of FLC and BUD can provide synergism against resistant C. albicans in vitro. No synergism has been obtained with susceptible C. albicans in vitro. As shown in Table 2 and S1 Table, CA10 and CA16 became susceptible when the MICs of FLC were redetermined in the presence of BUD using a checkerboard broth microdilution method. Both FICI and 4E models revealed the synergistic effects of BUD and FLC against resistant C. albicans in vitro. These findings may also have implications for the prevention of drug resistance. Insect infection models have significant ethical and economic advantages over mammalian infection models and can provide a rapid evaluation of the efficacy and toxicity of agents in vivo. G. mellonella, an insect infection model, has recently been used as an alternative to vertebrates for studying a number of important human pathogens [28]. The doses of FLC and BUD were chosen to mimic those used to treat human infections. Notably, exposure of infected G. mellonella larvae to a combination of FLC and BUD resulted in significantly enhanced survival (P<0.05) compared to monotherapy with either FLC or BUD (Fig 1, S2 Table). The fungal burden observed in the G. mellonella larvae correlated with the survival experiment. The measurement of G. mellonella larval burden in C. albicans showed that treatment with a combination of FLC and BUD had a synergistic inhibitory effect, as there was slower proliferation of C. albicans within the G. mellonella larvae over 4 days (Fig 2, S3 Table) and the numbers of C. albicans were decreased compared to those from infected G. mellonella larvae treated with PBS or drug alone (P < 0.05). To complement the studies in vivo, we further performed histopathology of infected G. mellonella larvae. Treatment with the drug alone hardly has any effect on the histopathology of the G. mellonella larvae, although the treatment with FLC and BUD simultaneously resulted in much fewer melanized nodules (Fig 3). Therefore, our data in vivo were in agreement with the antifungal effect determination in vitro. The mechanism involved in protecting G. mellonella larvae from C. albicans infection is unknown but might result from the immunomodulatory activities of BUD. It has been previously demonstrated that the highly frequent FLC resistance in C. albicans was in part attributed to the high expression of multidrug transporters, which can result in decreased drug concentrations in the fungal cells [29,30]. Two groups of drug transporters are found in C. albicans: the ATP-binding cassette   (ABC) transporters and the major facilitator superfamily (MFS) transporters [31]. The products of CDR1 and CDR2 are ABC transporters, and the products of MDR1 and FLU1 are MFStype transporters [32]. In our study, we found that there were significant inhibitory effects on the uptake and efflux of FLC in C. albicans treated with BUD (Fig 4, S4 and S5 Tables). In addition, we found that there was a significant down-regulation of drug transporter encoding genes CDR1, CDR2, MDR1 and FLU1 in the presence of BUD (Fig 5, S9 Table), suggesting that the BUD-mediated increased susceptibility of resistant C. albicans to FLC is at least in part through a mechanism of suppression of the expression of these drug transporters. Accumulating evidence reveals that the formation of C. albicans biofilms could enhance the resistance of this pathogen to most of the commonly used antifungal agents [33][34][35]. High drug resistance conferred by C. albicans biofilms is recognized as a significant and growing clinical problem. Here, we found that BUD combined with FLC has significant synergistic effects against the resistant C. albicans biofilms in different stages (Table 3, S6 Table). Over the past years, the genetic network controlling the biofilm-associated antifungal drug resistance has been investigated and partially elucidated [36]. Among several groups of genes involved in biofilm formation, it was found that EFG1, HWP1 and ALS1 are remarkable [36][37][38]. EFG1, one of the transcription factors, is required for biofilm formation and has a certain role in the resistance of C. albicans biofilms to antifungals in vivo [39,40]. HWP1, encoding a fungal cell wall protein, is required for hyphal development and yeast adhesion to epithelial cells as the  initial step of biofilm colonization [38]. ALS1, encoding cell surface glycoproteins, exhibits high expression in C. albicans biofilm cells and also plays a key role in biofilm formation [36]. Significant decreases were found in the expressions of EFG1, HWP1 and ALS1 (Fig 6, S9 Table), which was in accordance with the anti-biofilm effects of FLC and BUD.
Recent studies have revealed that apoptosis may be a novel antifungal mechanism in C. albicans [41,42]. Several studies have reported that ROS is both necessary and sufficient for inducing apoptosis in C. albicans [41,43]. Increases in ROS are regarded as one of the typical hallmarks of early apoptosis [44][45][46]. BUD used with FLC in the present study resulted in a significantly increased level of ROS (P < 0.05), as shown in Fig 7 and S7 Table. Apoptosis is a complex process involving multiple factors, including MCA1, HSP90, CRZ1 and CNA1 [47][48][49]. MCA1 is a central regulator of apoptosis and is involved in releasing ROS in fungi [43]. Hsp90 is a conserved molecular chaperone involved in the rapid development of drug resistance in C. albicans. It has been proven that Hsp90 potentiates the evolution of azole resistance in C. albicans via a calcineurin-dependent way [50]. CRZ1 and CNA1 are important elements of the calcineurin pathway, and both are reported to modulate azole resistance in C. albicans [48,49]. To gain further insight into the molecular mechanism underlying induced apoptosis, we evaluated the expression levels of genes closely related to fungal apoptosis: MCA1, HSP90, CRZ1 and CNA1 [47]. MAC1 was up-regulated, but HSP90, CRZ1 and CNA1 were down-regulated after treatment with BUD combined with FLC (Fig 8, S9 Table), indicating that apoptosis and its involved multiple factors play an important role in the proposed synergistic combination. We speculate that MCA1 may also be relevant to the clinical acquisition of fluconazole resistance in C. albicans.
Extracellular phospholipase is pivotal to the virulence of C. albicans, making it a possible target for the development of novel antifungals [51]. The pathogenicity of C. albicans is regulated by many virulence factors and by their interactions with the immune response to the host. These virulence factors are primary sources of drug resistance development and invasive infections because they are difficult or impossible to eradicate using conventional anticandidal agents [52]. Of the virulence factors studied in C. albicans, only extracellular phospholipase activity was predictive of mortality [53]. Thus, phospholipase is considered a major virulence factor in C. albicans. In the present report, we demonstrated that BUD had a potential to increase the susceptibility of resistant C. albicans to FLC by suppressing the activity of extracellular phospholipase (Table 4, S8 Table) and inhibiting the expression of phospholipase encoding genes (Fig 9, S9 Table). These findings indicate that the inhibition of phospholipase activity could be a key mechanism of action by which BUD sensitizes the resistant C. albicans to FLC.
There have been few reports of iatrogenic Cushing syndrome caused by interactions between BUD and strong inhibitors of cytochrome CYP 3A4, mainly itraconazole [54,55]. Inhaled BUD is a potent glucocorticoid that undergoes 99% first-pass metabolism through hepatic CYP 3A4. Although the oral adverse effects associated with inhaled corticosteroid use should not be ignored, the adverse effects of BUD are fairly low, which may due to its low systemic availability [56]. Despite the possibility that the strong CYP 3A4 inhibitor itraconazole in combination with BUD can cause complications, the inhibition of FLC on CYP 3A4 is much weaker than itraconazole and may subsequently cause fewer effects when combined with BUD in humans. In addition, the mechanisms and targets described in our study may help researchers design new FLC sensitizers with fewer side effects in the future, with the help of modern computer aids and structure-based selection techniques.
There have been few reports of iatrogenic Cushing syndrome caused by interaction between BUD and strong inhibitors of cytochrome CYP 3A4 mainly itraconazole [54,55]. Inhaled BUD is a potent glucocorticoid that undergoes 99% first-pass metabolism through hepatic CYP 3A4. Although the oral adverse effects associated with inhaled corticosteroid use should not be ignored, the adverse effects of BUD are fairly low, which may due to its low systemic availability [56]. Despite the strong CYP 3A4 inhibitor itraconazole in combination with BUD can cause complications, the inhibition of FLC on CYP 3A4 is much weaker than itraconazole and may cause few subsequent effects of BUD in human. In addition, the mechanisms and targets described in our study may help researchers designing new FLC sensitizers with few side effects in the future, with the help of modern computer aid and structure-based selecting techniques.

Conclusion
In conclusion, we first found that BUD can work synergistically with FLC against resistant C. albicans both in vitro and in vivo. These findings indicated that glucocorticoids enhance the anti-infection effect through not only anti-inflammatory effects but also a combined antifungal effect. Although the G. mellonella infection model has been used to test the efficacy of antifungals, this model still has limitations. The efficacy of FLC combined with BUD in vivo should be further evaluated using a mammalian model in future studies. Mechanism research demonstrated that this synergism is related to the inhibition of drug transporter function, the reduction of the formation of biofilms, the promotion of apoptosis and the inhibition of the activity of extracellular phospholipases. RT-PCR results revealed many potential targets for competing drug resistance in C. albicans, namely drug transporter genes (CDR1, CDR2, MDR1 and FLU1), biofilm-related genes (EFG1, HWP1 and ALS1), apoptosis-related genes (MAC1, HSP90, CRZ1 and CNA1) and phospholipases encoding genes (PLB1-5 and PLC1). This is the first study outlining the mechanisms of BUD against C. albicans alone or in combination with FLC. In addition, our study may lead to the identification of new potential antifungal targets and promote research on the in vitro effects of other glucocorticoids on C. albicans.
Supporting Information S1 Table. The data for in vitro drug effects against four C. albicans strains. The growth rates of C. albicans CA4, CA8, CA10 and CA16 after treatment with FLC alone or in combination with BUD are shown in S1a, 1b, 1c and 1d Table,