A major multidrug transporter, MDR1 (multidrug resistance 1), a member of the MFS (major facilitator superfamily), invariably contributes to an increased efflux of commonly used azoles and thus corroborates their direct involvement in MDR in Candida albicans. The Mdr1 protein has two transmembrane domains, each comprising six transmembrane helices, interconnected with extracellular loops and ICLs (intracellular loops). The introduction of deletions and insertions through mutagenesis was used to address the role of the largest interdomain ICL3 of the MDR1 protein. Most of the progressive deletants, when overexpressed, eliminated the drug resistance. Notably, restoration of the length of the ICL3 by insertional mutagenesis did not restore the functionality of the protein. Interestingly, most of the insertion and deletion variants of ICL3 became amenable to trypsinization, yielding peptide fragments. The homology model of the Mdr1 protein showed that the molecular surface-charge distribution was perturbed in most of the ICL3 mutant variants. Taken together, these results provide the first evidence that the CCL (central cytoplasmic loop) of the fungal MFS transporter of the DHA1 (drug/proton antiporter) family is critical for the function of MDR. Unlike other homologous proteins, ICL3 has no apparent role in imparting substrate specificity or in the recruitment of the transporter protein.

INTRODUCTION

The overexpression of the drug efflux pumps encoding the genes CDR (Candida drug resistance) 1 and CDR2 of the ABC (ATP-binding cassette) family of transporters and MDR1 (multidrug resistance 1) transporters of the MFS (major facilitator superfamily) family of transporters, represents one of the most predominant mechanisms of azole resistance in the pathogenic yeast Candida albicans [14]. Among the ABC transporters, a high expression level of CDR1 and CDR2 and, among the MFS, an overexpression of MDR1 invariably contribute to the increased efflux of commonly used azoles, which confirms their direct involvement in MDR [5]. A detailed structural and functional analysis of ABC and MFS drug transporter proteins is necessary for the development of modulators/inhibitors for these exporters.

MFS is one of the two largest superfamilies of membrane transporters present in bacteria, archaea and eukarya and includes members that function as uniporters, symporters or antiporters. The computational analysis of the C. albicans genome revealed 95 putative (MFS) proteins that clustered into 17 families [2]. The Mdr1 protein of C. albicans (GenBank® accession number CAA37820), which belongs to the DHA1 (drug–proton antiporter) family, is a 564 amino acid protein with two TMDs (transmembrane domains), each comprising six TMHs (transmembrane helices) interconnected by ECLs (extracellular loops) and ICLs (intracellular loops) [6]. Several MFS proteins, including members of the fungal DHA1 family [7], possess a CCL (central cytoplasmic loop) large enough to form a cytosolic domain [713]. The helices in this domain are rich in non-polar amino acids, which impart considerable amphipathicity to the loop for establishing contact with the PM (plasma membrane) [11]. The CCL connecting the TMDs plays an important functional role in several MFS proteins. For example, the CCLs of the LacY (lactose permease of Escherichia coli) transporter [12], Dtr1p (di-tyrosine transporter) of Saccharomyces cerevisiae [14], human RFC1 (reduced folate carrier) [15] and hPEPT1 (human intestinal di-/tri-peptide transporter) [11] are necessary for their efficient membrane insertion or their proper maturation, expression or function. In human GLUT1 (glucose transporter 1), the conformational changes within the TMHs induced by the initiation of efflux activity have shown to be communicated through the CCL of the protein [8].

For the structural and functional analysis, the MDR1 gene of C. albicans was previously cloned and overexpressed as a GFP (green fluorescent protein)-tagged protein in S. cerevisiae [4]. The Mdr1 protein is an antiporter with a well-conserved ‘antiporter motif’ located within TMH5 [G(×6)G(×3)GP(×2)GP(×2)G], which is necessary for its drug/proton transport activity [4]. On the basis of theoretical measurements, Kapoor et al. [16] recently subjected the protein to rational site-directed mutagenesis and confirmed the role of several conserved residues located on different TMHs, which impart functional specificity to the Mdr1 protein. From the deduced three-dimensional homology model of the Mdr1 protein, TMH6 and TMH7 have been proposed to be located far apart from each other [17]. Whether similar to other MFS transporters, with the ICL3 of antiporter Mdr1 protein interconnecting TMH6 and TMH7 being a communicating link between the two halves of the protein, or whether they also have other roles in protein recruitment and function and/or in imparting substrate specificity, are some of the aspects that remain unexplored.

In the present study, to determine the role of ICL3, we employed a mutational strategy to evaluate the role of the entire stretch of amino acid residues of the loop. Thus we made two sets of mutant variants of the Mdr1 protein. One group of variants included nine progressive deletants of different stretches of six or seven residues, and the other group included variants in which the respective deletions were restored with hexa- or hepta-alanine. Our data clearly show that most of the deletions and substitutions of the ICL3 sequence stretches resulted in non-functional protein variants that completely eliminated their ability to confer multidrug resistance and substrate efflux. We show that the progressive deletions and their restoration lead to changes in protein conformation and surface-charge distribution that was evident from the accessibility of the protease digestion of variants, resulting in multiple tryptic fragments.

EXPERIMENTAL

Reagents

The drugs CYH (cycloheximide), 4NQO (4-nitroquinoline), MTX (methotrexate) and CER (cerulenin), and the proteolytic enzyme trypsin and protease inhibitors [PMSF, leupeptin, pepstatin A, aprotinin, TPCK (tosylphenylalanylchloromethane) and TLCK (tosyl-lysylchloromethyl ketone)] were obtained from Sigma. FLU (fluconazole) was from Ranbaxy Laboratories. The anti-GFP mAb (monoclonal antibody) (mAb-GFP) was purchased from BD Biosciences Clontech. The yeast anti-Dpm1 (dolichol phosphate mannose synthase) mAb (mAb-Dpm1) and ER-Tracker™ Red dye were purchased from Molecular Probes. The polyclonal yeast anti-Pma1 (PM ATPase) antibody was a gift from Professor Ramon Serrano (Universidad Politecnica de Valencia-CSIC, Valencia, Spain).

Strains and media

The S. cerevisiae strain used was AD1-8u(MATa pdr1-3 his1 ura3 Δ yor1::hisG Δsnq2::hisG Δpdr5::hisG Δpdr10::hisG Δpdr11::hisG Δycf1::hisG Δpdr3::hisG Δpdr15::hisG) which was provided by Professor Richard Cannon (University of Otago, Dunedin, New Zealand) [18,19]. The plasmids were maintained in competent Escherichia coli DH5α cells and cultured in Luria–Bertani medium obtained from Difco (BD Biosciences) and HiMedia, to which ampicillin was added (100 mg/ml). The yeast strains used in the present study are listed in Supplementary Table S1 (at http://www.BiochemJ.org/bj/445/bj4450313add.htm). The yeast strains were cultured in YEPD (yeast extract peptone dextrose) broth (Bio101) or in SD-ura dropout medium (0.67% yeast nitrogen base, 0.2% dropout mix and 2% glucose; Difco). For the agar plates, 2.5% (w/v) Bacto agar (Difco) was added to the medium.

Site-directed mutagenesis of the Mdr1 protein

The site-directed mutagenesis was performed using the QuikChange® site-directed mutagenesis kit (Stratagene) as described previously [20]. The mutations were introduced into plasmid pRPCaMDR1-GFP according to the manufacturer's instructions, and the desired nucleotide sequence alterations were confirmed by DNA sequencing of the ORF (open reading frame) [4]. The primers used for this purpose are listed in Supplementary Table S2 (at http://www.BiochemJ.org/bj/445/bj4450313add.htm). The mutated plasmids, after being linearized with XbaI, were used to transform AD1-8u cells for uracil prototrophy using the lithium acetate transformation protocol [21]. Integration was confirmed by Southern blot analysis (results not shown).

Confocal microscopy

The cells were grown to the late log phase in SD-ura medium, except for AD1-8u, which required the addition of uridine (0.02%). The harvested cells were then washed and resuspended in an appropriate volume of 50 mM Hepes (pH 7.0). To stain the ER (endoplasmic reticulum), live yeast cells were incubated with the ER-Tracker™ Red dye as described in the manufacturer's protocol [22]. The cells were placed on to glass slides and then imaged under an oil immersion objective at ×100 magnification (Radiance 2100, AGR, 3Q/BLD; Bio-Rad Laboratories).

Drug susceptibility assay

The drug susceptibilities of the yeast strains were measured by two independent methods: a micro-dilution spot assay and MIC80 (minimal inhibitory concentration for 80% of input cells) value determination. The MIC80 values for the strains were determined using a broth micro-dilution method as per the NCCLS 27A protocol [23]. For the spot assay, 5-fold serial dilutions (5 μl each) of each yeast culture resuspended in normal saline to D600=0.1 were spotted on to YEPD plates in the absence (control) or presence of the drugs [4]. The growth differences were recorded following the incubation of the plates for 48 h at 30°C [24].

PM preparation and immunodetection of Mdr1 protein

The PM fractions were prepared as described previously [20]. The PM protein concentration was determined using the bicinchoninic acid assay [25] with BSA as the standard. For Western blots, the PM containing Mdr1–GFP and its mutants was subjected to SDS/PAGE (12% gel), and the immunodetection was performed using mAb-GFP at a dilution of 1:5000. The same blots were stripped and reprobed with anti-Pma1 antibody at a dilution of 1:1000 followed by mAb-Dpm1 at a dilution of 1:2500. The immunoreactivity was detected using a HRP (horseradish peroxidase)-labelled secondary antibody at a dilution of 1:5000 and the HRP substrate for chemiluminescence (Immobilon HRP substrate, Millipore).

Transport of substrate

The accumulation of NR (Nile Red) in cells expressing WT (wild-type) or mutant variants of Mdr1–GFP was measured by flow cytometry with a FACSort flow cytometer (Becton Dickinson Immunocytometry Systems) [18]. Briefly, cells with a D600 of 0.1 were inoculated and allowed to grow at 30°C with shaking until the D600 reached 0.25. The cells were then harvested and resuspended as a 5% cell suspension in diluted medium (containing one part YEPD and two parts water). NR was added to a final concentration of 7 mM, and the cells were incubated at 30°C for 30 min. The cells were then harvested, and 10000 cells were analysed in the acquisition. The analysis was performed using the CellQuest software (Becton Dickinson Immunocytometry Systems).

Protease protection assay

Limited proteolysis was performed by digesting PM fractions that contained Mdr1–GFP or mutant variants of the protein (10 μg) with a ratio 20:1 (protein/enzyme) of purified trypsin (Sigma) in 50 mM Tris/HCl (pH 7.5). The digestions were incubated at 4°C for 5 min, and the reactions were stopped by adding PMSF. Immediately afterwards, the samples were loaded on to SDS/PAGE (10% gel) [26]. Mdr1–GFP and its mutants were immunodetected using mAb-GFP [16].

Sequence analysis and ab initio-based modelling of ICL3

The analysis of the amino acid sequence of ICL3 of the Mdr1 protein was performed using online servers such as PRALINE [27], PsiPred [28], HeliQuest [29] and AveHAS [30]. An efficient web-based loop-modelling method, FALC-Loop server [31], was used to model the ICL3. The Mdr1 protein structure generated by MODELLER 9v10 using GlpT (glycerol-3-phosphate transporter; PDB code 1PW4) as a template was submitted to the server with the positions and sequence of ICL3. Out of the several output models generated, the one with the minimum DFIRE energy score [32] was selected, and the PDB files were visualized using PyMOL v1.4 (http://www.pymol.org) to qualitatively generate the electrostatic surface potential. The resultant models were further validated by PROCHECK [34].

Statistical analysis

The data are expressed as the means±S.E.M. Where required, Student's t test was used to establish the statistical significance. The criterion for significance was set at P≤0.05.

RESULTS

Unlike the TMHs, which share some degree of conservation [17], the sequence alignment of ICL3 of the Mdr1 protein with all the other members of the fungal DHA1 family displayed no significant conservation (Supplementary Figure S1A at http://www.BiochemJ.org/bj/445/bj4450313add.htm). However, all of the ICL3 domains of the proteins of this family show homology in terms of the α-helix propensities and hydrophobicity (Supplementary Figures S1B and S1C). The ICL3 of the Mdr1 protein with 55 amino acids (Glu297 to Glu351) represents the largest of all of the inter-helical loops (Figure 1) [17]. Notably, all of the ICL3s from fungal members of the DHA1 family have charged residues, with the basic amino acid residues mostly clustered towards the N-terminal end (Figure 1 and Supplementary Figure S2C at http://www.BiochemJ.org/bj/445/bj4450313add.htm). Considering the homologous features, such as the secondary structure, distribution of charged residues and hydrophobicity, of the ICL3 domains of members of the fungal DHA1 family and other known MFS proteins, the functional relevance of ICL3 in Mdr1 could be extrapolated. In the following, we explored the role of the ICL3 in the structure and function of the Mdr1 protein by subjecting the protein to a series of deletion, insertion and site-directed mutational analyses (Supplementary Figure S2).

Topology of the Mdr1 protein with 12 TMHs and a large CCL or ICL3

Figure 1
Topology of the Mdr1 protein with 12 TMHs and a large CCL or ICL3

The ICL3 domain was progressively deleted by six or seven residues, and the deletion was restored by the insertion of polyalanines at the respective positions. The stretches subjected to mutation are indicated by bars under the sequence, and the charged residues are represented by + (basic amino acids) and − (acidic amino acids).

Figure 1
Topology of the Mdr1 protein with 12 TMHs and a large CCL or ICL3

The ICL3 domain was progressively deleted by six or seven residues, and the deletion was restored by the insertion of polyalanines at the respective positions. The stretches subjected to mutation are indicated by bars under the sequence, and the charged residues are represented by + (basic amino acids) and − (acidic amino acids).

Progressive deletion of ICL3 eliminated drug resistance

We constructed a set of nine deletant mutant variants covering the entire length of the loop; each resulting mutant protein was shorter by six or seven residues, and these were designated as ICL3-1Δ297–302, ICL3-2Δ303–309, ICL3-3Δ310–315, ICL3-4Δ316–321, ICL3-5Δ322–327, ICL3-6Δ328–333, ICL3-7Δ334–339, ICL3-8Δ340–345 and ICL3-9Δ346–351 (Supplementary Figure S2A). All of the GFP-tagged mutant variants were stably overexpressed from a genomic PDR (pleiotropic drug resistance) 5 locus in a S. cerevisiae AD1-8u host, which was earlier developed by Goffeau and colleagues, that was derived from a Pdr1-3 mutant strain with a gain-of-function mutation in the transcription factor Pdr1p, resulting in the constitutive hyper-induction of the PDR5 promoter [19]. A single copy integration of each transformant at the PDR5 locus was confirmed by Southern hybridization (results not shown).

The susceptibility of the ICL3 deletants to various structurally unrelated drugs, e.g. FLU, CYH, 4NQO, MTX and CER, was assessed by micro-dilution spot assays (Figure 2A). These assays clearly demonstrated that most of the deletants, including ICL3-1Δ297–302, ICL3-2Δ303–309, ICL3-3Δ310–315, ICL3-4Δ316–321, ICL3-5Δ322–327, ICL3-8Δ340–345 and ICL3-9Δ346–351, became highly susceptible to drugs, but a pair of deletants, ICL3-6Δ328–333 and ICL3-7Δ334–339, showed only moderately reduced resistance to various drugs (Figure 2A). The deletant pair ICL3-6Δ328–333 and ICL3-7Δ334–339 did not show any difference in the MIC80 values, which were similar to those of native protein-expressing cells. The general MIC80 values of all of the deletants corroborated the spot assay data (Supplementary Table S3A at http://www.BiochemJ.org/bj/445/bj4450313add.htm).

Drug resistance profile of yeast strains overexpressing WT and mutant Mdr1 proteins as determined by spot assay

Figure 2
Drug resistance profile of yeast strains overexpressing WT and mutant Mdr1 proteins as determined by spot assay

For the spot assay, 5-fold serial dilutions (5 μl), namely 1 (1:5), 2 (1:25), 3 (1:125) and 4 (1:625), of each strain were spotted on to YEPD plates as described previously in the absence (control) and presence of the following drugs: FLU (0.20 mg/ml), CYH (0.20 mg/ml), 4NQO (0.20 mg/ml), MTX (65 mg/ml) and CER (3 mg/ml), and the growth differences were recorded following incubation of the plates for 48 h at 30°C. Growth was not affected by the presence of the solvents used for the drugs (results not shown). The drug resistance profile of the yeast strains overexpressing the WT Mdr1 protein with the deletion mutants (A) and polyalanine insertion mutants (B).

Figure 2
Drug resistance profile of yeast strains overexpressing WT and mutant Mdr1 proteins as determined by spot assay

For the spot assay, 5-fold serial dilutions (5 μl), namely 1 (1:5), 2 (1:25), 3 (1:125) and 4 (1:625), of each strain were spotted on to YEPD plates as described previously in the absence (control) and presence of the following drugs: FLU (0.20 mg/ml), CYH (0.20 mg/ml), 4NQO (0.20 mg/ml), MTX (65 mg/ml) and CER (3 mg/ml), and the growth differences were recorded following incubation of the plates for 48 h at 30°C. Growth was not affected by the presence of the solvents used for the drugs (results not shown). The drug resistance profile of the yeast strains overexpressing the WT Mdr1 protein with the deletion mutants (A) and polyalanine insertion mutants (B).

The susceptibility profile of these deletant variants was corroborated by monitoring the efflux of an Mdr1 protein substrate, NR. In contrast to the native protein, cells expressing all the deletants, except ICL3-6Δ328–333 and ICL3-7Δ334–339, were unable to efflux the dye. The pair of deletants, ICL3-6Δ328–333 and ICL3-7Δ334–339, which showed no change in their drug susceptibility pattern, also exhibited a near-normal level of efflux activity (Figure 3A).

Substrate transport assays

Figure 3
Substrate transport assays

The intracellular accumulation assay of the fluorescent substrate NR in the different deletion (A) and insertion (B) mutant variants of Mdr1–GFP. The intensity scale for the fluorescence measurements is in arbitrary units (a.u.). The values are the means of at least two independent experiments. Results are means±S.E.M. *P≤0.05.

Figure 3
Substrate transport assays

The intracellular accumulation assay of the fluorescent substrate NR in the different deletion (A) and insertion (B) mutant variants of Mdr1–GFP. The intensity scale for the fluorescence measurements is in arbitrary units (a.u.). The values are the means of at least two independent experiments. Results are means±S.E.M. *P≤0.05.

Progressive insertion mutants of ICL3 fail to restore drug resistance

We evaluated whether the substitution of hexa- or hepta-alanine for the deleted native sequences would compensate for the abrogated drug resistance and efflux activity. We replaced the deleted stretches of these variants, and designated the resulting variants as ICL3-1Ala297–302, ICL3-2Ala303–309, ICL3-3Ala310–315, ICL3-4Ala316–321, ICL3-5Ala322–327, ICL3-6Ala328–333, ICL3-7Ala334–339, ICL3-8Ala340–345 and ICL3-9Ala346–351 (Supplementary Figure S2B). Most of the mutant variants with replacement stretches of alanine, which restored the length of the native loop, did not show any reversal in their drug susceptibility pattern and, similar to their deletant counterparts, remained highly susceptible to drugs (Figure 2B). Notably, the deletant pair, ICL3-6Δ328–333 and ICL3-7Δ334–339, which displayed drug resistance at normal levels, retained moderate resistance for one of its substituted mutants (ICL3-7Ala334–339) while completely eliminating resistance to all drugs for the other (ICL3-6Ala328–333). The variant ICL3-9Ala346–351, in contrast with its deletant counterpart ICL3-9Δ346–351, also partially restored resistance to a few drugs (Figure 2B). The determination of the MIC80 values of all the insertion mutants corroborated the spot assays results (Supplementary Table S3B).

The drug susceptibility profiles of the alanine replacement mutants were further validated with the NR accumulation studies, in which all other mutants, except for ICL3-7Ala334–339 and ICL3-9Ala346–351, showed no efflux (Figure 3B).

Both deletion and insertion mutants are expressed and properly surface localized

The possibility that the resulting phenotypes were not due to poor expression or localization of protein was ruled out by laser confocal microscopy. The confocal images of GFP-tagged protein revealed that both the deletion and insertion mutant variants of ICL3 were properly surface localized on the PM (Figure 4A). To further rule out that the GFP fluorescence does not represent the peri-nuclear ER and ‘cell surface’-like staining, which can constitute the cortical ER that lies underneath the PM, we also stained all the cells with an ER-specific dye, ER-Tracker™ [22]. The confocal images of the ER-Tracker™ staining clearly show red fluorescence of the dye across the entire yeast cell extending close to the PM. However, a superimposition of the GFP and ER-Tracker™ fluorescence images provided a clear distinction between the PM-localized Mdr1–GFP and the extent of the cortical ER just underneath it (Figures 4A and 4B). Together, both the GFP and ER-Tracker™ staining confirmed that the deletion and insertion mutants of the Mdr1 protein were properly surface localized.

Cell surface expression and membrane localization of the WT and mutants

Figure 4
Cell surface expression and membrane localization of the WT and mutants

Confocal microscopic images of all deletion (A) and alanine replacement (B) mutant stains showing the PM expression of GFP-tagged Mdr1 protein. The confocal image panel of each strain consists of four subpanels, representing fluorescence due to GFP (upper left-hand quadrant), fluorescence due to the ER-Tracker™ dye (upper right-hand quadrant), overlay of above both the fluorescence (lower left-hand quadrant) and superimposition of the fluorescence overlay on a differential interference contrast image (lower right-hand quadrant). The immunodetection of Mdr1 protein with deletion (C) and insertion (D) mutations by mAb-GFP, mAb-Dpm1 and by a polyclonal anti-Pma1 antibody.

Figure 4
Cell surface expression and membrane localization of the WT and mutants

Confocal microscopic images of all deletion (A) and alanine replacement (B) mutant stains showing the PM expression of GFP-tagged Mdr1 protein. The confocal image panel of each strain consists of four subpanels, representing fluorescence due to GFP (upper left-hand quadrant), fluorescence due to the ER-Tracker™ dye (upper right-hand quadrant), overlay of above both the fluorescence (lower left-hand quadrant) and superimposition of the fluorescence overlay on a differential interference contrast image (lower right-hand quadrant). The immunodetection of Mdr1 protein with deletion (C) and insertion (D) mutations by mAb-GFP, mAb-Dpm1 and by a polyclonal anti-Pma1 antibody.

The localization of all the mutant variants was further confirmed by Western blot analysis of PM fractions. As controls, we used an ER-specific yeast Dpm1 protein as an ER marker [35] and Pma1 as a PM marker. The Western blots clearly showed that the expression levels of the GFP-tagged proteins (probed with mAb-GFP) in all of the variants were similar. The loading control was probed with a polyclonal anti-Pma1 antibody, which ensured equal loading of the protein. Our purified PM fractions from all the strains had minor contamination of the ER fraction (approximately 10%). This level of contamination is not unexpected given the high density of ER, because obtaining PM free of ER membranes is extremely difficult. Taken together, the confocal imaging and Western analysis data demonstrated that the Mdr1 mutant variant proteins are properly localized in the PM.

The ICL3 insertions and deletions make the protein amenable to tryptic digestion

There are instances, such as in the GLUT1 and hPEPT1 proteins, in which the central cytosolic domain has been shown to be an essential structural element [8,11]. In the following experiment, we explored this possibility by employing a limited protease protection assay and assessed whether the deletion and replacement mutations of ICL3 led to any changes in the native conformation of the Mdr1 protein. The PM fractions isolated from crude membrane extracts of native protein- and different mutant variant-expressing cells were subjected to limited proteolysis with trypsin [26]. The resulting peptide fragments were detected using mAb-GFP (Figure 5). Interestingly, despite the existence of several trypsin proteolytic sites, the native Mdr1 protein was resistant to protease treatment independent of time and temperature and gave a single band with an expected size of ~87 kDa (~60 kDa of the Mdr1 protein and ~27 kDa of the GFP tag) (Supplementary Figure S3 at http://www.BiochemJ.org/bj/445/bj4450313add.htm). In contrast, the deletion mutants ICL3-1Δ297–302, ICL3-2Δ303–309, ICL3-3Δ310–315, ICL3-4Δ316–321, ICL3-5Δ322–327, ICL3-8Δ340–345 and ICL3-9Δ346–351 gave two major bands in the region of ~55 kDa (~28 kDa of the C-terminal TMD and ~27 kDa of the GFP tag), indicating that the protein was cleaved largely into two parts and that the exposed tryptic sites were in the ICL3 loop itself (Figure 5A). Interestingly, the deletant mutants ICL3-6Δ328–333 and ICL3-7Δ334–339 remained inaccessible to trypsin. Most of the variants with replaced sequences also became amenable to trypsin digestion. Notably, in contrast with their deleted counterparts, ICL3-6Ala328–333 and ICL3-7Ala334–339 became susceptible to trypsin digestion (Figure 5B). To check whether a single amino acid mutant could cause any significant changes in the native confirmation, the purified PMs of randomly selected mutants, such as Mdr1-K301A and Mdr1-R306A, were subjected to limited trypsinization (Figure 5C). These mutant proteins remained intact after the limited trypsinization. Apparently, single amino acid mutations could not introduce any significant alterations in the native confirmation of the protein (discussed below).

Limited protease protection assay

Figure 5
Limited protease protection assay

Limited trypsinization of the PM fractions containing Mdr1–GFP and its mutant variants was performed as described in the Experimental section to monitor the differences in the native conformation due to deletion (A), insertion (B) and single amino acid residue (C) mutations. The banding pattern was immunodetected using mAb-GFP.

Figure 5
Limited protease protection assay

Limited trypsinization of the PM fractions containing Mdr1–GFP and its mutant variants was performed as described in the Experimental section to monitor the differences in the native conformation due to deletion (A), insertion (B) and single amino acid residue (C) mutations. The banding pattern was immunodetected using mAb-GFP.

The majority of the charged amino acids of ICL3 are functionally critical

A close examination of the ICL3 sequence reveals that the positively charged amino acid residues, i.e. K301A, R306A, K307A, K309A, R310A, R312A, R319A, K330A and H334A, are mostly confined to the N-terminal half of the loop, whereas the negatively charged amino acids, i.e. E297A, D318A, E323A, E325A, E327A, E335A, D339A, E346A and E351A, are predominantly clustered towards the C-terminal half (Figure 1 and Supplementary Figure S2C). To investigate the contribution of these charge residues, we constructed a series of mutant variants by replacing each residue individually with alanine by site-directed mutagenesis. Laser confocal microscopy (Figure 6A) and Western analysis (Figure 6B) ensured that all of the variants with replaced charges were properly recruited to the surface and expressed. The micro-dilution spot assays (Figure 7) and MIC80 values (Supplementary Table S4 at http://www.BiochemJ.org/bj/445/bj4450313add.htm) revealed that the replacement of most of the positively charged side-chain mutants with alanine completely eliminated their ability to tolerate drugs (Figure 7) and efflux NR (Figure 8). However, some of the substitutions of the basic residues with alanine (R306A, R312A and H334A) and of the acidic residues (E335A, D339A, E346A and E351A) did not affect the normal function of the protein.

Cell surface expression and membrane localization of the WT and charged residue mutant proteins

Figure 6
Cell surface expression and membrane localization of the WT and charged residue mutant proteins

(A) Confocal microscopic images of all the stains with single charged amino acid mutations of ICL3, showing the PM expression of the GFP-tagged Mdr1 protein. The confocal images for each strain consist of four subpanels, representing fluorescence due to GFP (upper left-hand quadrant), fluorescence due to the ER-Tracker™ dye (upper right-hand quadrant), overlay of above both the fluorescence (lower left-hand quadrant) and superimposition of the fluorescence overlay on a differential interference contrast image (lower right-hand quadrant). (B) Immunodetection of the Mdr1 protein in the PM by mAb-GFP to show the expression level of Mdr1–GFP.

Figure 6
Cell surface expression and membrane localization of the WT and charged residue mutant proteins

(A) Confocal microscopic images of all the stains with single charged amino acid mutations of ICL3, showing the PM expression of the GFP-tagged Mdr1 protein. The confocal images for each strain consist of four subpanels, representing fluorescence due to GFP (upper left-hand quadrant), fluorescence due to the ER-Tracker™ dye (upper right-hand quadrant), overlay of above both the fluorescence (lower left-hand quadrant) and superimposition of the fluorescence overlay on a differential interference contrast image (lower right-hand quadrant). (B) Immunodetection of the Mdr1 protein in the PM by mAb-GFP to show the expression level of Mdr1–GFP.

Drug susceptibility assays of charged residue mutant variants of the Mdr1 protein

Figure 7
Drug susceptibility assays of charged residue mutant variants of the Mdr1 protein

The drug resistance profile of S. cerevisiae strains overexpressing WT Mdr1 protein and the charged residue mutants were performed as described in Figure 2.

Figure 7
Drug susceptibility assays of charged residue mutant variants of the Mdr1 protein

The drug resistance profile of S. cerevisiae strains overexpressing WT Mdr1 protein and the charged residue mutants were performed as described in Figure 2.

Substrate transport assays

Figure 8
Substrate transport assays

The intracellular accumulation assay of the fluorescent substrate NR in the different mutant variants of Mdr1–GFP. The intensity scale for the intracellular fluorescence measurement is given in arbitrary units (a.u.). Results are the means±S.E.M. of at least two independent experiments. *P≤0.05.

Figure 8
Substrate transport assays

The intracellular accumulation assay of the fluorescent substrate NR in the different mutant variants of Mdr1–GFP. The intensity scale for the intracellular fluorescence measurement is given in arbitrary units (a.u.). Results are the means±S.E.M. of at least two independent experiments. *P≤0.05.

DISCUSSION

The Mdr1 protein, which belongs to the MFS family, is one of the major drug efflux proteins for which overexpression is often associated with frequently encountered azole resistance in Candida. The protein functions as a drug/proton antiporter, wherein it has a well-conserved ‘antiporter motif’ localized in TMH5. Interestingly, the protein appears to be divided into two halves interconnected with a large CCL. The large CCLs in several MFS proteins are a known feature that, however, apparently play no direct role in the transport mechanism. Interestingly, the conservation of the length of this CCL presents an interesting evolutionary predicament. Nonetheless, there are several studies on non-DHA1 family member proteins to show that it possesses many functionally critical amino acid residues. For example, the CCLs in the TetA (tetracycline efflux antiporter) and in RFC1 are involved in membrane trafficking [9,15]. Progressive deletions of the CCL of Lac permease and RFC1 result in reductions in the steady-state levels of transporter protein expression in the membrane [10,12]. In this context, the role of CCL or ICL3 of the Mdr1 protein of C. albicans or of any other DHA1 family member in membrane trafficking, stability and function is totally unexplored. To evaluate the role of ICL3 in the structure and function of the protein in the present study, we subjected it to progressive deletion and insertion mutational analysis.

ICL3 sequence deletions and insertions yield a non-functional protein

The progressive deletions of the entire sequence of ICL3 by six or seven residues and their substitution did not affect the membrane insertion of the protein variants; however, most of the variant proteins, when overexpressed in a heterologous host, eliminated the resistance to drugs and ability to efflux NR. Although the deletant results suggested that the length of the loop is critical, that hypothesis was not supported because the restoration of its length with stretches of hexa- or hepta-alanines did not reverse the loss of function. Of note, unlike other members of MFS, such as Lac permease [12], the deletion of different regions or subsequent substitution with an equivalent length of polyalanine in the ICL3 domain of the Mdr1 protein did not affect protein insertion and expression in the membrane (Figures 4A and 4B).

The deletion and insertion mutant variants result in the loss of the native conformation of the protein

The Mdr1 protein was totally inaccessible to limited protease digestion and showed no accessibility to trypsin. However, most of the deletions and their substitution with equivalent stretches of polyalanines made the protein variants amenable to trypsinization. Notably, the deletants ICL3-6Δ328–333 and ICL3-7Δ334–339 remained restricted to protease action, implying that the deletions of these stretches do not influence the native conformation. However, the substitution of these two stretches with hexa- and hepta-alanine resulted in mutant proteins that were accessible to tryptic digestion. Thus there appeared to be both size and sequence requirements for ICL3, which are able to drastically affect the native conformation and function of the protein. Why the replacement of ICL3-6Δ328–333 and ICL3-7Δ334–339 with alanines (6Ala328–333 and 7Ala334–339) allowed protease accessibility is not so apparent, but because the ICL3-6328–333 and ICL3-7334–339 stretches include both hydrophilic and hydrophobic residues, the replacement of these stretches with alanines could be speculated to perturb the orientation of the charged groups, which most likely results in protein accessibility to protease digestion. This argument could be extended to state that the predominant amphipathicity in the potential α-helix of ICL3, especially towards the C-terminus and the non-polar residues therein, might be involved in establishing hydrophobic contacts with the membrane to maintain the conformational stability of the native Mdr1 protein [36,37]. In hPEPT1, for example, the two putative amphipathic helices in loop 6–7 are predicted to interact with the membrane lipids for a stable conformational state [11].

For the ICL3-6328–333 and ICL3-7334–339 stretches, their deletion and replacement had similar effects with regard to trypsin digestion, which should be emphasized; however, these mutations had a different impact on the susceptibility to drugs, with ICL3-6Ala328–333 eliminating the resistance to all of the tested drugs, whereas the ICL3-7Ala334–339 variant remained resistant. The ICL3-6328–333 stretch is mostly hydrophilic (Supplementary Figure S1C), and its replacement with alanine residues can result in a loss of function (enhanced susceptibility to drugs). In contrast, the ICL3-7334–339 stretch is predominantly hydrophobic, and its deletion and replacement of the native residues appeared to be dispensable. The dispensability of the ICL3-7334–339 stretch was further evident from the observation that the site-directed mutagenesis of all the charged residues in this segment could not eliminate the resistance to drugs (Figure 7 and Supplementary Table S4).

Shortening the length of ICL3 results in a redistribution of the surface charges of the Mdr1 protein

Various mutant phenotypes can be correlated and explained on the basis of the redistribution of surface charges, which might get perturbed due to the loss of the involved amino acid residues upon deletion or insertion mutations. Indeed, our homology modelling of the Mdr1 protein supported this notion. The homology model of the Mdr1 protein and the deletion and insertion mutants show the perturbed distribution of surface charges (Supplementary Figure S4 at http://www.BiochemJ.org/bj/445/bj4450313add.htm). The deletion and insertion of ICL3 stretches result in conformational changes that became further evident when we modelled the ICL3 using an ab initio approach with a recently available web server [31]. The differences in the conformational states of the ICL3 domain in various mutants (Supplementary Figures S4A and S4B) leading to the redistribution of surface charges of the Mdr1 protein was obvious (Supplementary Figures S4C and S4D). The result of the individual replacement of charged residues in the ICL3 domain by site-directed mutagenesis reaffirmed their importance in protein functioning.

Taken together, the present study on the ICL3 domain of the Mdr1 protein not only explains the critical role of CCLs in a fungal MFS belonging to the DHA1 family of drug transporters, but also shows that the large inter-domain of central ICL3 connecting the two TMDs is an essential structural element of the protein.

Abbreviations

     
  • ABC

    ATP-binding cassette

  •  
  • CCL

    central cytoplasmic loop

  •  
  • CDR

    Candida drug resistance

  •  
  • CER

    cerulenin

  •  
  • CYH

    cycloheximide

  •  
  • DHA1

    drug/proton antiporter

  •  
  • Dpm1

    dolichol phosphate mannose synthase

  •  
  • ER

    endoplasmic reticulum

  •  
  • FLU

    fluconazole

  •  
  • GFP

    green fluorescent protein

  •  
  • GLUT1

    glucose transporter 1

  •  
  • hPEPT1

    human intestinal di-/tri-peptide transporter

  •  
  • HRP

    horseradish peroxidase

  •  
  • ICL

    intracellular loop

  •  
  • mAb

    monoclonal antibody

  •  
  • MDR1

    multidrug resistance 1

  •  
  • MFS

    major facilitator superfamily

  •  
  • MIC80

    minimal inhibitory concentration for 80% of input cells

  •  
  • MTX

    methotrexate

  •  
  • 4NQO

    4-nitroquinoline

  •  
  • NR

    Nile Red

  •  
  • PDR

    pleiotropic drug resistance

  •  
  • PM

    plasma membrane

  •  
  • Pma1

    PM ATPase

  •  
  • RFC1

    reduced folate carrier

  •  
  • TLCK

    tosyl-lysylchloromethyl ketone

  •  
  • TPCK

    tosylphenylalanylchloromethane

  •  
  • TMD

    transmembrane domain

  •  
  • TMH

    transmembrane helix

  •  
  • WT

    wild-type

  •  
  • YEPD

    yeast extract peptone dextrose

AUTHOR CONTRIBUTION

Ajeet Mandal generated the constructs, conducted the experiments and prepared the paper; Antresh Kumar performed some of the site-directed mutagenesis and yeast transformation experiments; Ashutosh Singh, Andrew M. Lynn and Khyati Kapoor contributed to the data discussion and manual editing of the paper; and Rajendra Prasad designed the experiments, analysed the data and wrote the paper.

We thank Professor Richard Cannon (University of Otago, Dunedin, New Zealand) for providing us with the plasmid and strains, and Professor Ramon Serrano (Universidad Politecnica de Valencia-CSIC, Valencia, Spain) for the anti-Pma1 antibody. We acknowledge the Advanced Instrumentation Research Facility (AIRF) and the Jawaharlal Nehru University (JNU) for providing the confocal microscope and technical assistance.

FUNDING

This work was supported in part by the Department of Biotechnology [grant numbers DBT No.BT/01/CEIB/10/III/02 and DBT No.BT/PR14879/BRB10/885/2010 (to R.P.)]. A.M. was supported by the University Grant Commission of India.

References

References
1
Gaur
M.
Choudhury
D.
Prasad
R.
Complete inventory of ABC proteins in human pathogenic yeast, Candida albicans
J. Mol. Microbiol. Biotechnol.
2005
, vol. 
9
 (pg. 
3
-
15
)
2
Gaur
M.
Puri
N.
Manoharlal
R.
Rai
V.
Mukhopadhayay
G.
Choudhury
D.
Prasad
R.
MFS transportome of the human pathogenic yeast Candida albicans
BMC Genomics
2008
, vol. 
9
 pg. 
579
 
3
Prasad
R.
De Wergifosse
P.
Goffeau
A.
Balzi
E.
Molecular cloning and characterization of a novel gene of Candida albicans, CDR1, conferring multiple resistance to drugs and antifungals
Curr. Genet.
1995
, vol. 
27
 (pg. 
320
-
329
)
4
Pasrija
R.
Banerjee
D.
Prasad
R.
Structure and function analysis of CaMdr1p, a major facilitator superfamily antifungal efflux transporter protein of Candida albicans: identification of amino acid residues critical for drug/H+ transport
Eukaryotic Cell
2007
, vol. 
6
 (pg. 
443
-
453
)
5
Prasad
R.
Kapoor
K.
Multidrug resistance in yeast Candida
Int. Rev. Cytol.
2005
, vol. 
242
 (pg. 
215
-
248
)
6
Paulsen
I. T.
Brown
M. H.
Skurray
R. A.
Proton-dependent multidrug efflux systems
Microbiol. Rev.
1996
, vol. 
60
 (pg. 
575
-
608
)
7
Saier
M. H.
Jr
Tran
C. V.
Barabote
R. D.
TCDB: the transporter classification database for membrane transport protein analyses and information
Nucleic Acids Res.
2006
, vol. 
34
 (pg. 
D181
-
D186
)
8
Monden
I.
Olsowski
A.
Krause
G.
Keller
K.
The large cytoplasmic loop of the glucose transporter GLUT1 is an essential structural element for function
Biol. Chem.
2001
, vol. 
382
 (pg. 
1551
-
1558
)
9
Sapunaric
F. M.
Levy
S. B.
Substitutions in the interdomain loop of the Tn10 TetA efflux transporter alter tetracycline resistance and substrate specificity
Microbiology
2005
, vol. 
151
 (pg. 
2315
-
2322
)
10
Sharina
I. G.
Zhao
R.
Wang
Y.
Babani
S.
Goldman
I. D.
Role of the C-terminus and the long cytoplasmic loop in reduced folate carrier expression and function
Biochem. Pharmacol.
2002
, vol. 
63
 (pg. 
1717
-
1724
)
11
Xu
L.
Li
Y.
Haworth
I. S.
Davies
D. L.
Functional role of the intracellular loop linking transmembrane domains 6 and 7 of the human dipeptide transporter hPEPT1
J. Membr. Biol.
2010
, vol. 
238
 (pg. 
43
-
49
)
12
Weinglass
A. B.
Kaback
H. R.
The central cytoplasmic loop of the major facilitator superfamily of transport proteins governs efficient membrane insertion
Proc. Natl. Acad. Sci. U.S.A.
2000
, vol. 
97
 (pg. 
8938
-
8943
)
13
Saraceni-Richards
C. A.
Levy
S. B.
Evidence for interactions between helices 5 and 8 and a role for the interdomain loop in tetracycline resistance mediated by hybrid Tet proteins
J. Biol. Chem.
2000
, vol. 
275
 (pg. 
6101
-
6106
)
14
Morishita
M.
Engebrecht
J.
Sorting signals within the Saccharomyces cerevisiae sporulation-specific dityrosine transporter, Dtr1p, C terminus promote Golgi-to-prospore membrane transport
Eukaryotic Cell
2008
, vol. 
7
 (pg. 
1674
-
1684
)
15
Sadlish
H.
Williams
F. M.
Flintoff
W. F.
Cytoplasmic domains of the reduced folate carrier are essential for trafficking, but not function
Biochem. J.
2002
, vol. 
364
 (pg. 
777
-
786
)
16
Kapoor
K.
Rehan
M.
Lynn
A. M.
Prasad
R.
Employing information theoretic measures and mutagenesis to identify residues critical for drug-proton antiport function in Mdr1p of Candida albicans
PLoS ONE
2010
, vol. 
5
 pg. 
e11041
 
17
Kapoor
K.
Rehan
M.
Kaushiki
A.
Pasrija
R.
Lynn
A. M.
Prasad
R.
Rational mutational analysis of a multidrug MFS transporter CaMdr1p of Candida albicans by employing a membrane environment based computational approach
PLoS Comput. Biol.
2009
, vol. 
5
 pg. 
e1000624
 
18
Decottignies
A.
Grant
A. M.
Nichols
J. W.
de Wet
H.
McIntosh
D. B.
Goffeau
A.
ATPase and multidrug transport activities of the overexpressed yeast ABC protein Yor1p
J. Biol. Chem.
1998
, vol. 
273
 (pg. 
12612
-
12622
)
19
Nakamura
K.
Niimi
M.
Niimi
K.
Holmes
A. R.
Yates
J. E.
Decottignies
A.
Monk
B. C.
Goffeau
A.
Cannon
R. D.
Functional expression of Candida albicans drug efflux pump Cdr1p in a Saccharomyces cerevisiae strain deficient in membrane transporters
Antimicrob. Agents Chemother.
2001
, vol. 
45
 (pg. 
3366
-
3374
)
20
Shukla
S.
Saini
P.
Smriti
Jha
S.
Ambudkar
S. V.
Prasad
R.
Functional characterization of Candida albicans ABC transporter Cdr1p
Eukaryotic Cell
2003
, vol. 
2
 (pg. 
1361
-
1375
)
21
Altherr
M. R.
Quinn
L. A.
Kado
C. I.
Rodroguez
R. L.
Transformation and storage of competent yeast cells
Genet. Eng. Eukaryot.
1983
, vol. 
1983
 (pg. 
33
-
36
)
22
Hambrock
A.
Loffler-Walz
C.
Quast
U.
Glibenclamide binding to sulphonylurea receptor subtypes: dependence on adenine nucleotides
Br. J. Pharmacol.
2002
, vol. 
136
 (pg. 
995
-
1004
)
23
Espinel-Ingroff
A.
Barchiesi
F.
Hazen
K. C.
Martinez-Suarez
J. V.
Scalise
G.
Standardization of antifungal susceptibility testing and clinical relevance
Med. Mycol.
1998
, vol. 
36
 (pg. 
68
-
78
)
24
Kohli
A.
Smriti
Mukhopadhyay
K.
Rattan
A.
Prasad
R.
In vitro low-level resistance to azoles in Candida albicans is associated with changes in membrane lipid fluidity and asymmetry
Antimicrob. Agents Chemother.
2002
, vol. 
46
 (pg. 
1046
-
1052
)
25
Smith
P. K.
Krohn
R. I.
Hermanson
G. T.
Mallia
A. K.
Gartner
F. H.
Provenzano
M. D.
Fujimoto
E. K.
Goeke
N. M.
Olson
B. J.
Klenk
D. C.
Measurement of protein using bicinchoninic acid
Anal. Biochem.
1985
, vol. 
150
 (pg. 
76
-
85
)
26
Baumann
H.
Doyle
D.
Effect of trypsin on the cell surface proteins of hepatoma tissue culture cells. Characterization of a carbohydrate-rich glycopeptide released from a calcium binding membrane glycoprotein
J. Biol. Chem.
1979
, vol. 
254
 (pg. 
3935
-
3946
)
27
Simossis
V. A.
Heringa
J.
PRALINE: a multiple sequence alignment toolbox that integrates homology-extended and secondary structure information
Nucleic Acids Res.
2005
, vol. 
33
 (pg. 
W289
-
W294
)
28
McGuffin
L. J.
Bryson
K.
Jones
D. T.
The PSIPRED protein structure prediction server
Bioinformatics
2000
, vol. 
16
 (pg. 
404
-
405
)
29
Gautier
R.
Douguet
D.
Antonny
B.
Drin
G.
HELIQUEST: a web server to screen sequences with specific α-helical properties
Bioinformatics
2008
, vol. 
24
 (pg. 
2101
-
2102
)
30
Zhai
Y.
Saier
M. H.
Jr
A web-based program for the prediction of average hydropathy, average amphipathicity and average similarity of multiply aligned homologous proteins
J. Mol. Microbiol. Biotechnol.
2001
, vol. 
3
 (pg. 
285
-
286
)
31
Ko
J.
Lee
D.
Park
H.
Coutsias
E. A.
Lee
J.
Seok
C.
The FALC-Loop web server for protein loop modeling
Nucleic Acids Res.
2011
, vol. 
39
 (pg. 
W210
-
W214
)
32
Zhang
C.
Liu
S.
Zhou
Y.
Accurate and efficient loop selections by the DFIRE-based all-atom statistical potential
Protein Sci.
2004
, vol. 
13
 (pg. 
391
-
399
)
33
Reference deleted
34
Laskowski
R. A.
Macarthur
M. W.
Moss
D. S.
Thornton
J. M.
PROCHECK: a program to check the stereochemical quality of protein structures
J. Appl. Cryst.
1993
, vol. 
26
 (pg. 
283
-
291
)
35
Jackson
D. D.
Stevens
T. H.
VMA12 encodes a yeast endoplasmic reticulum protein required for vacuolar H+-ATPase assembly
J. Biol. Chem.
1997
, vol. 
272
 (pg. 
25928
-
5934
)
36
Vila
J. A.
Ripoll
D. R.
Villegas
M. E.
Vorobjev
Y. N.
Scheraga
H. A.
Role of hydrophobicity and solvent-mediated charge–charge interactions in stabilizing α-helices
Biophys. J.
1998
, vol. 
75
 (pg. 
2637
-
2646
)
37
Thiyagarajan
M. M.
Stracquatanio
R. P.
Pronin
A. N.
Evanko
D. S.
Benovic
J. L.
Wedegaertner
P. B.
A predicted amphipathic helix mediates plasma membrane localization of GRK5
J. Biol. Chem.
2004
, vol. 
279
 (pg. 
17989
-
17995
)

Author notes

1

Present address: Laboratory of Cell Biology, Centre for Cancer Research, National Cancer Institute, National Institute of Health, Bethesda, U.S.A.

Supplementary data