Mutation of G234 amino acid residue in Candida albicans drug-resistance-related protein Rta2p is associated with fluconazole resistance and dihydrosphingosine transport

Abstract

Widespread and repeated use of azoles has led to the rapid development of drug resistance in Candida albicans. Our previous study found Rta2p, a membrane protein with 7 transmembrane domains, was involved in calcineurin-mediated azole resistance and sphingoid long-chain base release in C. albicans. Conserved amino acids in the transmembrane domain of Rta2p were subjected to site-directed mutagenesis. The sensitivity of C. albicans to fluconazole in vitro was examined by minimum inhibitory concentration and killing assay, and the therapeutic efficacy of fluconazole in vivo was performed by systemic mice candidiasis model. Furthermore, dihydrosphingosine transport activity was detected by NBD labeled D-erythro-dihydrosphingosine uptake and release assay, and the sensitivity to sphingolipid biosynthesis inhibitors. We successfully constructed 14 mutant strains of Rta2p, screened them by minimum inhibitory concentration and found Ca²⁺ did not completely induce fluconazole resistance with G158E and G234S mutations. Furthermore, we confirmed that G234S mutant enhanced the therapeutic efficacy of fluconazole against systemic candidiasis and significantly increased the accumulation of dihydrosphingosine by decreasing its release. However, G158E mutant didn't affect drug therapeutic efficacy in vivo and dihydrosphingosine transport in C. albicans. G234 of Rta2p in C. albicans is crucial in calcineurin-mediated fluconazole resistance and dihydrosphingosine transport.

Introduction

Candida albicans is a commensal fungus colonizing the skin, mucosal surface and gastrointestinal tract in most healthy individuals. While in immunocompromised patients, C. albicans can cause oropharyngeal, esophageal or disseminated candidiasis.^1,2 However, widespread and incorrect use of azoles in clinical have resulted in the fast development of multidrug resistance, which has been a tricky problem in the treatment of fungal infections.^3,4 In recent years, a diverse range of drug resistance mechanisms have been identified, including multi-drug resistance efflux gene (MDR1) as well as CDR-type drug efflux pumps (CDR1 and CDR2),^5-7 mutation of lanosterol 14-α demethylase Erg11p,⁸ and activation of calcineurin pathway.^9,10 Previous studies indicated that some efflux pumps were found to be over-expressed in the fluconazole (FLC) resistance clinical C. albicans isolates from 12 hospitals in California.¹¹

The protein information of C. albicans genome database indicates that RTA2 encodes a 50.9 kDa membrane protein, which is predicted to have 7 transmembrane domains. There are 4 Rta2p homologs (Rsb1p, Rta1p, Yer185w, and Ylr046c) in the S. cerevisiae genome database, with more than 40% similarity with RTA2.¹² It was found in our previous study that disruption of RTA2 could increase the sensitivity to azoles in C. albicans, and the azole resistance was mediated in calcineurin by Rta2p and the function of Rta2p was to release sphingoid long chain base (LCB).¹³ Elevated RTA2 expression levels were also found in clinical isolates of azole-resistant C. albicans.¹² As Rta2p is considered a potential target to combat antifungal resistance in clinical, the relationship between structure and function is important for designing antifungal drug targeting Rta2p. In this work, we developed and selected conserved amino acids in the transmembrane domains of Rta2p for site-directed mutagenesis, finding that G234 of RTA2 played a critical role in FLC resistance and the transport of dihydrosphingosine (DHS) in C. albicans.

Materials and Methods

Drugs

FLC was purchased from Changhai Hospital (Shanghai, China). Bovine serum albumin (BSA), fumonisin B1 and myriocin were from Sigma-Aldrich (St. Louis, MO, USA). C₆ NBD labeled D-erythro-dihydrosphingosine (NBD-DHS) was obtained from Larodan Fine Chemical AB (Sweden).

C. albicans strains growth conditions

C. albicans strains used in this study were maintained on SDA agar plates (1% peptone, 4% dextrose, and 1.8% agar) and grown in YPD broth (1% yeast extract, 2% peptone, and 2% dextrose) at 30°C or SC medium lacking uridine.

Site-directed mutation and C. albicans transformation

Site-directed mutation was performed using the QuikChange^® lightning site-directed mutagenesis kit from Stratagene (LaJolla, CA). The point mutation was introduced into the plasmid BES116 (ADE2-URA3-ADE2 AscI fragment in pBluescript II KS (+))¹⁴ according to the manufacturer's instructions (Table S1) and the desired nucleotide sequence alteration was confirmed by DNA sequencing. The original and mutant plasmids were linearized by NaeI and transformed to JXM101 using the Yeast Transformation System 2 from Clontech (Terra Bella, CA).

Determination of the minimum inhibitory concentration (MIC)

The drug MIC was determined by serial 2-fold dilutions in 96-well plates as described previously,¹⁵ based on the methods of the Clinical and Laboratory Standards Institute (M27-A2).¹⁶ The cells (100 μl; 5 × 10³-1 × 10⁴ cells/ml) were inoculated into a 96-well plate containing serial 2-fold dilutions of fluconazole (0.125–64 μg/ml) with or without 1 mM Ca²⁺ in RPMI 1640 medium. After 24 h incubation at 30°C, the optical density (OD) value was measured at 600 nm. The MIC₈₀ value was determined as the lowest concentration of FLC inhibited growth by 80%.

C. albicans killing assay

C. albicans strains grown at 30°C were harvested and re-suspended in RPMI 1640 medium and prepared at the starting inoculum of 10³ CFU/ml.¹⁷ The concentration of FLC was 0.125, 0.25, 0.5, 1, 2, 4, 8 and 16 μg/ml with or without 1 mM Ca²⁺. After 24 h, a 100 μl aliquot was serially diluted 10-fold in sterile water. A 100 μl aliquot from each dilution was spread on SDA plate. After 48 h incubation at 30°C, the number of colonies was counted and calculated. Three biological replicates were performed for the experiment.

Systemic mice candidiasis model. C

albicans strains grown at 30°C were harvested, washed and resuspended in sterile saline. ICR female mice (18–22 g) were from Shanghai SLAC Laboratory Animals and maintained in specific pathogen-free conditions at the Center for Experimental Medicine, Second Military Medical University. All mice were inoculated by injection of 0.2 ml C. albicans suspension into the lateral tail vein (1 × 10⁶ c.f.u/mouse) to establish the systemic candidiasis model.¹⁸ The mice were initially treated with FLC (1.0 mg/kg) or sterile saline 2 h after infection by C. albicans, and then once a day intraperitoneally for 7 days. Mice were monitored daily for survival until Day 30. Duplicate independent experiments were conducted (n = 10 in each group). Kaplan-Meier analyses were used to estimate survival probabilities.

NBD-DHS uptake and release Assay

For uptake analysis, exponentially growing C. albicans cells were washed and resuspended in PBS buffer (OD₆₀₀ = 0.45). The suspension was incubated with 5 mM NBD-DHS and 1 mg/ml BSA for 10, 30 and 60 min at 30°C. At each time cells were harvested and scanned by confocal microscopy (excitation wavelength, 488 nm; emission wavelength, 520–565 nm; Leica, Germany). In DHS release assay, cells were treated with 5 mM NBD-DHS for 1 h in PBS buffer and then incubated with 1 mg/ml BSA for the indicated time (0, 10 and 30 min) at 30°C. Cells were centrifugated and the concentrations of NBD-DHS in supernatant were measured by spectrophotometry using a standard curve for NBD-DHS in PBS buffer (excitation wavelength, 475 nm; emission wavelength, 525 nm; PolarStar, BMG).

Susceptibility testing

C. albicans cells (1 × 10⁴ cells/ml) were inoculated into successive wells of a 96-well microtiter plate containing serial 2-fold dilutions of fumonisin B1 and myriocin (initial concentration at 25 μg/ml, respectively). After incubation at 30°C for 16 h, absorbance at 600 nm was measured in a microplate reader.

Molecular docking

The template protein d2qpea of the original or mutant Rta2p was predicted through remote homology modeling by CPHmodels-3.2 Server (http://www.cbs.dtu.dk/services/CPHmodels/).¹⁹ With appropriate parameters, D-erythro-dihydrosphingosine could dock with the predicted structure and bind into a protein by Discovery Studio™ software.²⁰ The computer modeling results would indicate some core amino acids interacted directly between the nuclear protein structure and the chemical.

Ethics statement

The animal experiments were performed in compliance with the institutional guidelines and approved by the Animal Ethics Committee of Second Military Medical University, and all efforts were made to minimize suffering of the animals.

Statistical analysis

One-way analysis of variance (ANOVA) with Bonferroni post-tests was used to analyze multiple groups. Log-rank test was used to analyze the equality of survival curves by SPSS 17.0. P < 0.05 was considered statistically significant.

Results

Construction and expression of mutant RTA2 reintegration in C. albicans

Multiple sequence alignment of Rta2p horologes in S. cerevisiae genome (Rsb1p, Rta1p, Yer185w, and Ylr046c) showed 16 conserved amino acids in the transmembrane domain (Fig. 1). The transmembrane domain is the important topology determined domain of membrane protein,²¹ all conserved amino acids residues were selected for site-directed mutagenesis according to the length of side chain, with charge or not, and amino acid hydrophobicity. The whole alterations (C150W, E155A, G158E, L182F, P186R, F220L, D224A, Q231H, A232D, G234S, G235R, G258C, Q262H, D382G, F396L, and F402V) were introduced into the vector pBES116 and the mutant plasmids were transformed to competent cells. Positive mutant cells were further confirmed by DNA sequencing by extracting plasmid DNA (Fig. S1). Except D382G and F402V, the original RTA2, vector (pBES116) and other mutant plasmids were reintegrated to rta2Δ/Δ mutant strain JXM101, confirmed by PCR method (Fig. S2).

Figure 1.

The sequence alignment of Rta2p, Rsb1p, Rta1p, Yer185w and Ylr046c. The alignment was generated using the Clustal X2 programs. The black boxes indicate identical residues, and the gray boxes show amino acid similarity. The asterisks represent the conserved amino acids in transmembrane domains of homogenous proteins. TM1-TM7: indicated transmembrane domains predicted by C. albicans and S. cerevisiae genome database.¹¹

Resistance to FLC of wild-type and mutant Rta2p in vitro

The susceptibilities of C. albicans to FLC were measured by determining the lowest drug concentrations that achieved MIC₈₀. Our previous study found that RTA2 was involved in the resistance to FLC mediated by calcineurin pathway in C. albicans.²² It was found in this study that 1 mM CaCl₂ was able to increase resistance to FLC in P-RTA2 dramatically, with MIC₈₀ increasing from 1 to 64 μg/ml. However, the effect of Ca²⁺ on resistance to FLC completely disappeared without RTA2, with MIC₈₀ remaining unchanged in P-B (Table 2). With the effect of Ca²⁺, the mutant strains of RTA2 increased resistance to FLC in different levels, while the MIC₈₀ only increased from 0.25 to 1 μg/ml in M3 and remained unchanged in M10 (Table 2). The appearance of FLC resistance was further confirmed in C. albicans killing assay. Compared with P-RTA2, M10 was more sensitive to high concentrations of FLC, while M3 formed drug resistance in shaking state. The addition of 1 mM Ca²⁺ dramatically decreased the fungicidal effect of FLC from 0.125 to 64 μg/ml in P-RTA2 and M3 without affecting M10 and P-B (Table 3).

Table 1.

C. albicans strains used in this study

Strain	Genotype	Parental strain	Reference
JXM101	RM1000§rta2Δ::hisG/rta2Δ::hisG	RM1000	Jia et al.12
P-B	RM1000§rta2Δ::hisG/rta2Δ::hisG ADE2/ADE2::URA3	JXM101	This study
P-RTA2	RM1000§rta2Δ::hisG/rta2Δ::hisG ADE2/ADE2::URA3::RTA2	JXM101	This study
M1	RM1000§rta2Δ::hisG/rta2Δ::hisG ADE2/ADE2::URA3::RTA2C150W	P-RTA2	This study
M2	RM1000§rta2Δ::hisG/rta2Δ::hisG ADE2/ADE2::URA3::RTA2E155A	P-RTA2	This study
M3	RM1000§rta2Δ::hisG/rta2Δ::hisG ADE2/ADE2::URA3::RTA2G158E	P-RTA2	This study
M4	RM1000§rta2Δ::hisG/rta2Δ::hisG ADE2/ADE2::URA3::RTA2L182F	P-RTA2	This study
M5	RM1000§rta2Δ::hisG/rta2Δ::hisG ADE2/ADE2::URA3::RTA2P186R	P-RTA2	This study
M6	RM1000§rta2Δ::hisG/rta2Δ::hisG ADE2/ADE2::URA3::RTA2F220L	P-RTA2	This study
M7	RM1000§rta2Δ::hisG/rta2Δ::hisG ADE2/ADE2::URA3::RTA2D224A	P-RTA2	This study
M8	RM1000§rta2Δ::hisG/rta2Δ::hisG ADE2/ADE2::URA3::RTA2Q231H	P-RTA2	This study
M9	RM1000§rta2Δ::hisG/rta2Δ::hisG ADE2/ADE2::URA3::RTA2A232D	P-RTA2	This study
M10	RM1000§rta2Δ::hisG/rta2Δ::hisG ADE2/ADE2::URA3::RTA2G234S	P-RTA2	This study
M11	RM1000§rta2Δ::hisG/rta2Δ::hisG ADE2/ADE2::URA3::RTA2G235R	P-RTA2	This study
M12	RM1000§rta2Δ::hisG/rta2Δ::hisG ADE2/ADE2::URA3::RTA2G258C	P-RTA2	This study
M13	RM1000§rta2Δ::hisG/rta2Δ::hisG ADE2/ADE2::URA3::RTA2Q262H	P-RTA2	This study
M15	RM1000§rta2Δ::hisG/rta2Δ::hisG ADE2/ADE2::URA3::RTA2F396L	P-RTA2	This study

Table 2.

The activity of fluconazole with or without CaCl₂ (1 mM) against different Rta2p mutant strains in vitro (MIC₈₀, μg/ml). Data are representative of 3 independent experiments

	Strains	TMD	Mutagenic Site	FLC only	FLC with CaCl2
1	P-RTA2		RTA2 revertant	1	64
2	P-B		rta2Δ/Δ mutant	0.5	0.5
3	M1	TMD2	TGT→TGG(C150W)	2	64
4	M2	TMD2	GAG→GCG(E155A)	1	32
5	M3	TMD2	GGG→GAG(G158E)	0.25	1
6	M4	TMD3	TTA→TTC(L182F)	0.5	64
7	M5	TMD3	CCT→CGT(P186R)	0.25	32
8	M6	TMD4	TTC→TTG(F220L)	4	64
9	M7	TMD4	GAC→GCC(D224A)	2	64
10	M8	TMD4	CAA→CAC(Q231H)	0.5	64
11	M9	TMD4	GCT→GAT(A232D)	4	64
12	M10	TMD4	GGT→AGT(G234S)	0.25	0.25
13	M11	TMD4	GGA→CGA(G235R)	2	64
14	M12	TMD5	GGT→TGT(G258C)	1	32
15	M13	TMD5	CAA→CAC(Q262H)	2	64
16	M15	TMD7	TTC→TTG(F396L)	4	64

Table 3.

The C. albicans killing assay (%) by FLC with or without Ca²⁺. Data represent mean (±SD) of triplicates of one representive experiment of 3

FLC concentration (μg/ml)	P-RTA2		P-B		M3		M10
	FLC	FLC + Ca2+	FLC	FLC + Ca2+	FLC	FLC + Ca2+	FLC	FLC + Ca2+
0.125	4.92 ± 0.11	2.86 ± 0.09	12.34 ± 0.04	9.68 ± 0.07	9.45 ± 0.10	4.28 ± 0.01	11.26 ± 0.01	8.49 ± 0.01
0.25	15.16 ± 0.09	3.74 ± 0.02	24.81 ± 0.02	15.83 ± 0.29	18.37 ± 0.08	6.77 ± 0.03	23.23 ± 0.02	11.72 ± 0.02
0.5	17.72 ± 0.08	4.41 ± 0.05	30.53 ± 0.05	21.67 ± 0.02	21.16 ± 0.09	9.36 ± 0.05	34.07 ± 0.01	14.89 ± 0.06
1	22.07 ± 0.13	4.98 ± 0.06	34.38 ± 0.03	25.89 ± 0.04	28.26 ± 0.08	11.45 ± 0.08	36.38 ± 0.03	20.63 ± 0.01
2	26.53 ± 0.17	7.84 ± 0.05	34.70 ± 0.02	35.99 ± 0.17	29.37 ± 0.05	15.82 ± 0.04	37.03 ± 0.02	24.42 ± 0.05
4	27.45 ± 0.10	9.03 ± 0.11	42.21 ± 0.01	41.07 ± 0.07	30.54 ± 0.02	16.31 ± 0.01	44.58 ± 0.03	34.01 ± 0.04
8	27.01 ± 0.04	10.11 ± 0.31	50.37 ± 0.04	44.37 ± 0.16	29.78 ± 0.05	18.53 ± 0.01	52.02 ± 0.02	40.57 ± 0.01
16	27.21 ± 0.19	13.09 ± 0.28	51.62 ± 0.02	53.15 ± 0.12	30.79 ± 0.03	19.01 ± 0.02	53.11 ± 0.05	49.31 ± 0.35

Rta2p mutation affected the therapeutic efficacy of FLC against candidemia in vivo

The Ca²⁺ mediated calcineurin pathway was reported to be crucial for the virulence and FLC resistance of C. albicans in the systemic infection model.²³ Our previous study demonstrated that RTA2 was not involved in the virulence of C. albicans but could improve the survival rate of mice after FLC treatment.²² As the effect of Ca²⁺ failed to trigger FLC resistance completely in M10 in vitro, P-RTA2, P-B, M3 and M10 were selected to infect the mice. To examine the influences of mutant Rta2p on the efficacy of FLC against C. albicans in a murine model of systemic candidiasis, groups of 10 mice were intravenously injected with P-RTA2, P-B, M3 and M10, and the survival of mice was monitored for 30 days. As shown in Figure 2A, there was no significant difference in the survival rate between the groups of mice infected with P-RTA2, P-B, M3 and M10, while the mice infected by M10 after FLC treatment resulted in about 20% higher survival rate than the group infected by P-RTA2 in 30 days (P < 0.05; Fig. 2B). However, unlike the sensitivity to FLC with Ca²⁺ in settled state, the survival of M3 infected mice treated by FLC was similar with P-RTA2 infected mice (P > 0.05; Fig. 2B). These results suggested that G234, but not G158, was an important site of Rta2p affecting the therapeutic efficacy of FLC in the treatment of systemic candidiasis.

Figure 2.

Effects of FLC on survival curves of mice with systemic candidiasis. Groups of 10 mice were inoculated with 1 × 10⁶ cells/ml of each strain via the lateral tail vein. Survival and weight were monitored daily. The survival time of mice infected with P-RTA2, P-B, M3, M10 treated with FLC (1.0 mg/kg) for 7 days. Groups without treatment were inoculated with NaCl. Data are representative of 3 independent experiments.

G234S mutation of Rta2p increased accumulation of DHS by reducing the export

Like Rsb1p in S. cerevisiae, the function of Rta2p was to translocate DHS from the inner to the outer leaflet of the plasma membrane in C. albicans.^13,24 To study the crucial site of Rta2p function, we examined whether the mutant strains could affect the intracellular accumulation and export of DHS. When cultured with NBD-DHS, P-B accumulated fluorescence-labeled DHS in the cells efficiently. The intracellular fluorescence density was increased dramatically at 30 and 60 min, and the similar appearance occurred in M10. However, P-RTA2 and M3 showed a much lower accumulation of NBD-DHS (Fig. 3A). On the other hand, we determined the NBD-DHS export in C. albicans cells in vivo. It was found that only 0.32 μM DHS of P-B was exported into the supernatant at 10 min and the level increased to 0.39 μM at 30 min (Fig. 3B, black squares). M10 exported a little more than P-B, from 0.43 μM to 0.68 μM during 10–30 min (Fig. 3B, white squares). In contrast, P-RTA2 exported 1.12 μM DHS at 30 min, which was approximately threefold higher than that in P-B (Fig. 3B, black diamonds). M3 exported 1.01 μM at 30 min, which was slightly lower than P-RTA2 (Fig. 3B, white diamonds). Together, M3 is equally capable of pumping out NBD-DHS as the P-RTA2, suggesting that the residue G158 is not involved in transport function, while G234 site appears to have a more direct effect on transport function.

Figure 3.

The uptake and release of dihydrosphingosine (DHS) by wild and mutant Rta2p. (A) For uptake experiments, C. albicans cells were incubated 5 mM NBD-DHS for 10, 30, 60 min. At each time point, the cells were washed and immobilized on slides; pictures were acquired with a Leica confocal microscope. (B) For release experiments, cells were pre-incubated with NBD-DHS for 1 h, resuspended in PBS containing 1 mg/ml BSA, and then incubated for 0, 10, 30 min. At each time point, cells were centrifuged and the supernatant was measured fluorimetrically. Data are shown as means ± SDs from 3 independent experiments.

To get further verification, we tested the susceptibility of P-RTA2, P-B, M3 and M10 to myriocin and fumonisin B1, the upstream and downstream inhibitors of DHS in sphingolipid biosynthesis. As yeast cells could not survive in the presence of DHS over-accumulation,²⁵ the decrease of myriocin and increase of fumonisin B1 could result in the accumulation of DHS in C. albicans. Microbroth dilution assay was used to determine the sensitivity of C. albicans to myriocin or fumonisin B1. All strains were equally sensitive to myriocin (Fig. 4A). Compared with P-RTA2 and M3, P-B and M10 were more sensitive to fumonisin B1 due to more DHS accumulation (Fig. 4B). Above all, G234S mutation, but not G158E of Rta2p, significantly increased the DHS accumulation in cells by decreasing the export.

Figure 4.

The sensitivity to myriocin (A) or fumonisinB1 (B) of C. albicans P-RTA2 (▪), P-B(⧫), M3(□) and M10 (⋄) cells by microbroth dilution assays. The absorbance at 600 nm was recorded after 24 h incubation at 30°C. Data are means ± SDs of triplicate wells and representative of 3 independent experiments.

Discussion

In the present study, we selected conserved amino acids in the transmembrane domain of Rta2p for site-directed mutagenesis and demonstrated that G234S, but not G158E of RTA2, could affect the FLC resistance in vitro and in vivo, also could increase the accumulation of DHS by reducing its export.

Site-directed mutagenesis has been widely applied to find key residues of MDR-associated proteins such as Cdr1p, Cdr2p in C. albicans and Pdr5p in S. cerevisiae.^26-28 For example, mutation of C193, K901 and D327 of Cdr1p could change the ATPase activity of nucleotide-binding domains,^29,30 and the transmembrane domains of Cdr1p could specifically transport and bind substrates of different chemicals such as azoles, phospholipids and sterols.^31,32 The structure-functional relationship of Cdr1p determines the crucial amino acids and important domains of transport, which are considered as the base of designing suitable inhibitors or drugs targeting Crd1p of C. albicans. Like Cdr1p, Rta2p is also a membrane protein with several transmembrane domains, involved in azole resistance in clinical and sphingoid trasport in C. albicans.^12,13 Above all, it's a new approach to study structure and functional relationship of Rta2p in C. albicans.

In this study, we used the site-directed method to find functional amino acids of the drug-resistance-related protein Rta2p. Among the conserved amino acids in transmembrane domains, we found that the resistance of Rta2p mutant strains to FLC was increased in varying degrees upon the effect of 1 mM Ca²⁺, with MIC₈₀ of M3 (vector with RTA2-G158E) increasing from 0.25 to 1 μg/ml and that of M10 (vector with RTA2-G234S) remaining unchanged (Table 2). However, M3 formed FLC resistance with Ca²⁺ in shaking state in C. albicans killing assay (Table 3). We further found that the mutation of G234, but not G158, was more effective to treat systemic candidiasis by FLC in vivo (Fig. 2). With respect to Rta2p-mediated DHS translocation from inner to outer leaflet in C. albicans,¹³ we found that M10 could increase the accumulation of DHS by reducing the export, while M3 appeared not to have a more direct effect on transport function (Figs. 3 and 4). More importantly, the result of bioinformatics analysis suggested the interactions of key amino acids and D-erythro-dihydrosphingosine were completely different between the wild type and M10 (Fig. 5). The computer imitation results further suggested that G234 of Rta2p was an important amino acid involved in the binding between the nuclear structure and DHS.

Figure 5.

The computer modeling of amino acids and D-erythro-dihydrosphingosine between P-RTA2 (A) and M10 (B) by Discovery Studio™ software. Some key amino acids (violet) including Q90, Y121, R162 could directly link to DHS in the wild type strain, while the crucial amino acids were A233 and Y340 in the mutant strain.

n conclusion, we found that the mutant strain M10 was more sensitive to FLC in vitro and in vivo with Ca²⁺ than the other mutant strains as represented by decreased export of DHS. Our results suggest that the mutation in TMD4 (G234S) is only relevant in the presence of calcium when investigating FLC resistance in C. albicans. In addition, the concentration of 1 mM CaCl₂ used in this study was almost equal to that of ionized calcium in mammalian serum (about 1–2 mM). Inhibition of calcium-mediated azole resistance has been proposed as a new therapeutic target of antifungal therapy.³³ Sphingolipids, composed of dihydrosphingosine and phytosphingosine, are the primary lipids of plasma membrane in C. albicans. Our results also confirm that G234 is the functional site to affect DHS transport. Moreover, the susceptibility of C. albicans to FLC and transport of DHS are not merely dependent on the transmembrane domains. Further studies are needed to verify whether non-transmembrane domains and some amino acids residues of Rta2p have the function of lipid transport.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

We would like to thank Xin-Ming Jia for providing C. albicans strain JXM101 and plasmid BES116.

Funding

This work was supported by the National Key Basic Research Program of China (No 2013CB531602), the National Science Foundation of China (No 81273556, 81330083), and Shanghai Science and Technology Major Project (11JC1415400).

Supplemental Material

Supplemental data for this article can be accessed on the publisher's website.