BBA - Biomembranese-portal.ccmb.res.in/e-space/amit/Group Publications/2018 BBA.pdf · covery. FRAP...

12
Contents lists available at ScienceDirect BBA - Biomembranes journal homepage: www.elsevier.com/locate/bbamem Phosphatidylserine decarboxylase governs plasma membrane uidity and impacts drug susceptibilities of Candida albicans cells Nitesh Kumar Khandelwal a,b,1 , Parijat Sarkar c,1 , Naseem A. Gaur b , Amitabha Chattopadhyay c, , Rajendra Prasad d, a School of Life Sciences, Jawaharlal Nehru University, New Delhi 110067, India b International Centre for Genetic Engineering and Biotechnology, New Delhi 110067, India c CSIR-Centre for Cellular and Molecular Biology, Uppal Road, Hyderabad 500007, India d Amity Institute of Integrative Sciences and Health, Amity University Haryana, Amity Education Valley, Gurgaon 122413, India ARTICLE INFO Keywords: Phosphatidylserine decarboxylase Phosphatidylethanolamine Plasma membrane Fluorescence recovery after photobleaching Dierential scanning calorimetry Drug susceptibility ABSTRACT Plasma membrane (PM) lipid composition imbalances aect drug susceptibilities of the human pathogen Candida albicans. The PM fundamental structure is made up of phospholipid bilayer where phosphatidyletha- nolamine (PE) contributes as second major phospholipid moieties, which is asymmetrically distributed between the two leaets of the bilayer. PSD1 and PSD2 genes encode phosphatidylserine decarboxylase which converts phosphatidylserine (PS) into PE in C. albicans cells. Genetic manipulation of PSD1 and PSD2 genes is known to impact virulence, cell wall thickness and mitochondrial function in C. albicans. In the present study, we have examined the impact of PSD1 and PSD2 deletion on physiochemical properties of PM. Our uorescence recovery after photobleaching (FRAP) experiments point that the PM of psd1Δ/Δ psd2Δ/Δ mutant strain displays in- creased membrane uidity and reduced PM dipole potential. Further, the result of PSD1 and PSD2 deletion on the thermotropic phase behavior monitored by dierential scanning calorimetry (DSC) showed that in com- parison to WT, the apparent phase transition temperature is reduced by ~3 °C in the mutant strain. The func- tional consequence of altered physical state of PM of psd1Δ/Δ psd2Δ/Δ mutant strain was evident from observed high diusion of uorescent dye rhodamine 6G and radiolabelled uconazole (FLC). The higher diusion of FLC resulted in an increased drug accumulation in psd1Δ/Δ psd2Δ/Δ mutant cells, which was manifested in an increased susceptibility to azoles. To the best of our knowledge, these results constitute the rst report on the eect of the levels of phospholipid biosynthesis enzyme on physiochemical properties of membranes and drug susceptibilities of Candida cells. 1. Introduction Candida albicans is the most commonly isolated opportunistic human fungal pathogen. It is usually present as commensal in healthy humans but in immunocompromised individuals, it causes candidasis [1]. Approximately 912% of all hospital-acquired fungal infections are attributed to C. albicans with mortality rates rising up to 50% [2,3]. Dierent categories of drugs are used to combat invasive fungal infec- tions which mainly includes azoles, polyenes and echinocandins [4]. However, prolonged treatment with antifungals increases the frequency of the generation of MDR (multi drug resistance) strains. Various mechanisms are involved in the development of MDR in which rapid expulsion of drug by PM localized ABC (ATP-binding cassette) superfamily and MFS (major facilitator superfamily) drug transporters is one of the pronounce strategies [57]. Recently, it has been observed that apart from drug expulsion, drug diusion across the PM into the cells also play a crucial role in the development of drug tolerance [811]. Our earlier studies have conrmed that physical state of PM is a critical factor which impacts drug susceptibilities of C. al- bicans cells. For instance, it is earlier demonstrated that reduced drug diusion alone can decrease drug susceptibilities of Candida cells [811]. The phospholipids of C. albicans cells are mainly composed of phosphatidyl choline (PC), PE, phosphatidyl inositol (PI), PS, https://doi.org/10.1016/j.bbamem.2018.05.016 Received 7 March 2018; Received in revised form 1 May 2018; Accepted 27 May 2018 Corresponding authors. 1 These authors contributed equally to this work. E-mail addresses: [email protected] (A. Chattopadhyay), [email protected] (R. Prasad). Abbreviations: PM, plasma membrane; PS, phosphatidylserine; PE, phosphatidylethanolamine; FRAP, uorescence recovery after photobleaching; DSC, dierential scanning calori- metry; R6G, rhodamine 6G BBA - Biomembranes 1860 (2018) 2308–2319 Available online 29 May 2018 0005-2736/ © 2018 Elsevier B.V. All rights reserved. T

Transcript of BBA - Biomembranese-portal.ccmb.res.in/e-space/amit/Group Publications/2018 BBA.pdf · covery. FRAP...

Page 1: BBA - Biomembranese-portal.ccmb.res.in/e-space/amit/Group Publications/2018 BBA.pdf · covery. FRAP experiments were performed with Gaussian spot-photo-bleaching and line-scanning

Contents lists available at ScienceDirect

BBA - Biomembranes

journal homepage: www.elsevier.com/locate/bbamem

Phosphatidylserine decarboxylase governs plasma membrane fluidity andimpacts drug susceptibilities of Candida albicans cells

Nitesh Kumar Khandelwala,b,1, Parijat Sarkarc,1, Naseem A. Gaurb, Amitabha Chattopadhyayc,⁎,Rajendra Prasadd,⁎

a School of Life Sciences, Jawaharlal Nehru University, New Delhi 110067, Indiab International Centre for Genetic Engineering and Biotechnology, New Delhi 110067, Indiac CSIR-Centre for Cellular and Molecular Biology, Uppal Road, Hyderabad 500007, Indiad Amity Institute of Integrative Sciences and Health, Amity University Haryana, Amity Education Valley, Gurgaon 122413, India

A R T I C L E I N F O

Keywords:Phosphatidylserine decarboxylasePhosphatidylethanolaminePlasma membraneFluorescence recovery after photobleachingDifferential scanning calorimetryDrug susceptibility

A B S T R A C T

Plasma membrane (PM) lipid composition imbalances affect drug susceptibilities of the human pathogenCandida albicans. The PM fundamental structure is made up of phospholipid bilayer where phosphatidyletha-nolamine (PE) contributes as second major phospholipid moieties, which is asymmetrically distributed betweenthe two leaflets of the bilayer. PSD1 and PSD2 genes encode phosphatidylserine decarboxylase which convertsphosphatidylserine (PS) into PE in C. albicans cells. Genetic manipulation of PSD1 and PSD2 genes is known toimpact virulence, cell wall thickness and mitochondrial function in C. albicans. In the present study, we haveexamined the impact of PSD1 and PSD2 deletion on physiochemical properties of PM. Our fluorescence recoveryafter photobleaching (FRAP) experiments point that the PM of psd1Δ/Δ psd2Δ/Δ mutant strain displays in-creased membrane fluidity and reduced PM dipole potential. Further, the result of PSD1 and PSD2 deletion onthe thermotropic phase behavior monitored by differential scanning calorimetry (DSC) showed that in com-parison to WT, the apparent phase transition temperature is reduced by ~3 °C in the mutant strain. The func-tional consequence of altered physical state of PM of psd1Δ/Δ psd2Δ/Δ mutant strain was evident from observedhigh diffusion of fluorescent dye rhodamine 6G and radiolabelled fluconazole (FLC). The higher diffusion of FLCresulted in an increased drug accumulation in psd1Δ/Δ psd2Δ/Δ mutant cells, which was manifested in anincreased susceptibility to azoles. To the best of our knowledge, these results constitute the first report on theeffect of the levels of phospholipid biosynthesis enzyme on physiochemical properties of membranes and drugsusceptibilities of Candida cells.

1. Introduction

Candida albicans is the most commonly isolated opportunistichuman fungal pathogen. It is usually present as commensal in healthyhumans but in immunocompromised individuals, it causes candidasis[1]. Approximately 9–12% of all hospital-acquired fungal infections areattributed to C. albicans with mortality rates rising up to 50% [2,3].Different categories of drugs are used to combat invasive fungal infec-tions which mainly includes azoles, polyenes and echinocandins [4].However, prolonged treatment with antifungals increases the frequencyof the generation of MDR (multi drug resistance) strains.

Various mechanisms are involved in the development of MDR in

which rapid expulsion of drug by PM localized ABC (ATP-bindingcassette) superfamily and MFS (major facilitator superfamily) drugtransporters is one of the pronounce strategies [5–7]. Recently, it hasbeen observed that apart from drug expulsion, drug diffusion across thePM into the cells also play a crucial role in the development of drugtolerance [8–11]. Our earlier studies have confirmed that physical stateof PM is a critical factor which impacts drug susceptibilities of C. al-bicans cells. For instance, it is earlier demonstrated that reduced drugdiffusion alone can decrease drug susceptibilities of Candida cells[8–11].

The phospholipids of C. albicans cells are mainly composed ofphosphatidyl choline (PC), PE, phosphatidyl inositol (PI), PS,

https://doi.org/10.1016/j.bbamem.2018.05.016Received 7 March 2018; Received in revised form 1 May 2018; Accepted 27 May 2018

⁎ Corresponding authors.

1 These authors contributed equally to this work.E-mail addresses: [email protected] (A. Chattopadhyay), [email protected] (R. Prasad).

Abbreviations: PM, plasma membrane; PS, phosphatidylserine; PE, phosphatidylethanolamine; FRAP, fluorescence recovery after photobleaching; DSC, differential scanning calori-metry; R6G, rhodamine 6G

BBA - Biomembranes 1860 (2018) 2308–2319

Available online 29 May 20180005-2736/ © 2018 Elsevier B.V. All rights reserved.

T

Page 2: BBA - Biomembranese-portal.ccmb.res.in/e-space/amit/Group Publications/2018 BBA.pdf · covery. FRAP experiments were performed with Gaussian spot-photo-bleaching and line-scanning

phosphatidyl glycerol (PG), phosphatidic acid (PA), lysoPC, lysoPE andlysoPG [12,13]. PE is the second most abundant phospholipid in Can-dida cells [12,13], that contains a small polar head group and present asa conical molecular structure. This structural advantage helps it to formmembrane curvature which is important for various crucial functions[14,15]. In C. albicans, PE is synthesized by two pathways; (i) de novopathway, where PS is converted to PE by Psd1 or Psd2 enzymes, and (ii)Kennedy pathway in which PE can be formed by directly utilizing CDP-ethanolamine (Fig. 1). The levels of PSD1/PSD2 which encode phos-phatidylserine decarboxylase are known to impact virulence, cell wallthickness and mitochondrial functions in C. albicans [16]. PreviouslyChen et al. have performed the phospholipdome analysis of C. albicanspsd1Δ/Δ psd2Δ/Δmutant strain and observed significant decrease in PElevel in the mutant cells as compared to WT cells (in WT 14.5 ± 0.8%of total phospholipids) to (8.0 ± 1.1 in psd1Δ/Δ psd2Δ/Δ mutant). Nosignificant change in PI (WT 13.6 ± 0.9 to psd1Δ/Δ psd2Δ/Δ mutant14.0 ± 0.8% of total phospholipids) and PC level (WT 53 ± 0.6 topsd1Δ/Δ psd2Δ/Δ mutant 54 ± 0.8% of total phospholipids) was ob-served. Expectedly, the accumulation of Psd1 or Psd2 enzymes sub-strate PS was significantly higher in mutant cells (WT 6.1 ± 0.8 topsd1Δ/Δ psd2Δ/Δ mutant 11.2 ± 1.0) [16].

Considering the requirement of phospholipid as crucial buildingblock in the PM formation, in this present study, we tested the influenceof PE synthesis via de novo pathway on PM physicochemical propertiesand its effect on drug susceptibilities of C. albicans cells. For this, weemployed a homozygous double knockout mutant strain (psd1Δ/Δpsd2Δ/Δ) and observed that PM of mutant cells possesses increasedmembrane fluidity. The changes in membrane fluidity resulted in de-creased dipole potential and lowering of phase transition temperature,leading to enhancement of drug diffusion and susceptibility.

2. Materials and methods

2.1. Materials

The growth media YEPD (yeast extract/peptone/dextrose) and LBbroth were purchased from Himedia (Mumbai, India). The antifungalcompounds KTC (Ketoconazole), ITC (Itraconazole), MCZ (miconazole)and chemicals sucrose, Tris buffer, DMSO (Dimethyl sulfoxide), poly-L-lysine, Sorbitol, EDTA (Ethylenediaminetetraacetic acid), poly-L-lysineand protease inhibitor cocktails were purchased from Sigma ChemicalCo. (St. Louis, MO). FAST-DiI and Di-8-ANEPPS were purchased fromMolecular Probes/Invitrogen (Eugene, OR). FLC (Fluconazole) wasgenerously provided by Ranbaxy, India. Radioactive fluconazole ([3H]FLC: 2.8 Ci/mmol, 1 μCi/μl; 357.14 μM FLC) was custom synthesized byMoravek Biochemicals (Brea, CA).

2.2. Strains and culture conditions

The C. albicans strains used in the present study are listed in Table 1.

C. albicans strains stock were made in 15% glycerol at −80 °C andfreshly revived in YEPD before use. All strains were grown and main-tained in YEPD media. Bacterial strain Escherichia coli DH5α was used asa host to construction and propagation of plasmids. E. coli cells weregrown in LB medium containing 100 μg/ml ampicillin (Amresco,Radnor, PA), at 37 °C with shaking at 180 rpm for 16 h.

2.3. Labeling of C. albicans cells for experiments involving FRAP

Stock solution of FAST-DiI was prepared in methanol and con-centration was estimated from its molar extinction coefficient (ε) of148,000M−1 cm−1 at 549 nm [17]. C. albicans culture was suspendedat a density of ~108 cells/ml in 1M sorbitol-0.1 M EDTA buffer (bufferA) and labeled with FAST-DiI as previously described using final con-centration of 10 μM FAST-DiI in dark at 25 °C with mild shaking in glasstubes for 30min. [11]. After two washes with buffer A, cell pellet wasresuspended in 50 μl of buffer A. An aliquot of 10 μl of this suspensionwas mounted between a glass slide and a poly-L-lysine coated (0.01%,w/v) glass coverslip, sealed with nail enamel and used for imaging andFRAP experiments.

2.4. Fluorescence imaging of labeled C. albicans and FRAP experiments

Images were acquired at room temperature on an inverted ZeissLSM 510 Meta confocal microscope with a plan-apochromat 100×/1.4NA oil-immersion objective using the 561 nm laser. Fluorescenceemission was collected using the 575–630 nm bandpass filter with theconfocal pinhole set at 1 airy unit with a zoom factor of 7. The PMregion was selected for bleaching and monitoring recovery of fluores-cence. Data on diffusion coefficient (D) and mobile fraction (Mf) werecalculated from quantitative FRAP experiments in which just thebleached region was scanned over time to observed fluorescence re-covery. FRAP experiments were performed with Gaussian spot-photo-bleaching and line-scanning mode with circular ROI of 1 μm radius.Data representing the mean fluorescence intensity in the membraneregion (obtained using Zeiss LSM 510 software, version 3.2) within thebleached spot were corrected for background and analyzed. For a two-

Fig. 1. Schematic overview of PE synthesis in C. al-bicans.In C. albicans PE is synthesized by two pathways (i)De novo pathways; during de novo synthesis twoparallel pathways can synthesis PE by decarboxyla-tion of PS. PS can be converted to PE by Psd1 andPsd2 enzymes in mitochondria and Golgi vesiclesrespectively. (ii) Kennedy pathway; in this pathwayexogenous ethanolamine is converted to PE.

Table 1List of strains used in the study.

Strain Genotype/description Source/reference

WT (SC5314) Wild-type strain [53]psd1Δ/Δ psd2Δ/Δ SC5314 derivative, psd2Δ::FRT/

psd2Δ::FRT psd1Δ::FRT/psd1Δ::FRT[16]

CA-CDR6 SC5314 cells harboring extra copy of GFP-CDR6, integrated at RPS1 locus

[11]

NKKY104 psd1Δ/Δ psd2Δ/Δ cells harboring extracopy of GFP-CDR6, integrated at RPS1locus

The presentstudy

N.K. Khandelwal et al. BBA - Biomembranes 1860 (2018) 2308–2319

2309

Page 3: BBA - Biomembranese-portal.ccmb.res.in/e-space/amit/Group Publications/2018 BBA.pdf · covery. FRAP experiments were performed with Gaussian spot-photo-bleaching and line-scanning

dimensional diffusion model, FRAP data were fitted to determine thecharacteristic diffusion time (τd) [18]:

= − − + +∞F [F F ][exp( 2τ /t)(I (2τ /t) I (2τ /t))] F(t) ( ) (0) d 0 d 1 d (0) (1)

where F(t) is the mean background-corrected and normalized fluores-cence intensity at time t in the membrane region within the bleachedspot, F(∞) is the recovered fluorescence at time t→∞, and F(0) is thebleached fluorescence intensity at time t= 0. I0 and I1 are modifiedBessel functions. The effective two-dimensional diffusion coefficient (D)was determined from the equation [19]:

=D ω /2τ2d (2)

where ω is the radius of the bleached spot. Mobile fraction (Mf) wascalculated according to the equation:

= − −∞M [F F ]/[F F ]f ( ) (0) (p) (0) (3)

where F(p) is the mean background-corrected and normalized pre-bleach fluorescence intensity. Non-linear curve fitting of fluorescencerecovery data to Eq. (1) was carried out using the GraphPad Prismsoftware version 4.00 (San Diego, CA).

2.5. CDR6 overexpression strain construction

To construct the CDR6 overexpressing strain in C. albicans, we lin-earized the CIp-SAT-PTDH3-GFP-CDR6-SP plasmid with StuI. Next wetransformed C. albicans strain psd1Δ/Δ psd2Δ/Δ with linear DNA frag-ments by an electroporation method as reported previously [20]. Thescreening of C. albicans transformants were performed on YEPD platescontaining nourseothricin at 200 μg/ml. Further the putative transfor-mant colonies were checked by fluorescence microscopy to confirm theCdr6/Roa1-GFP expression. The resulting GFP-tagged CDR6 over-expression strain was named as NKKY104.

2.6. Confocal imaging of Cdr6/Roa1-GFP overexpressing C. albicans cellsand FRAP experiments

C. albicans culture was suspended at a density of ~108 cells/ml inbuffer A and drop of this suspension was mounted on glass coverslipspre-treated with 0.01% (w/v) poly-L-lysine. The coverslip was placedinverted on the microscope stage for imaging and FRAP experiments.Images were acquired at room temperature (~23 °C) on an invertedZeiss LSM 510 Meta confocal microscope (Jena, Germany), with a100×/1.4 NA oil immersion objective. The 488 nm line of an argonlaser was used for excitation, and emission was collected with a500–550 nm bandpass filter.

For FRAP analysis, images were acquired at room temperature(~23 °C) using the same set-up as mentioned above, with a zoom factorof 7. The distinct membrane fluorescence of the cell periphery wastargeted for bleaching and monitoring of fluorescence recovery.Analysis with a control ROI drawn at a certain distance away from thebleach ROI indicated no significant bleach while fluorescence recoverywas monitored. Data representing the mean fluorescence intensity ofthe bleached ROI (1 μm) were background subtracted using an ROIplaced outside the cell. Fluorescence recovery plots with fluorescenceintensities normalized to pre-bleach intensities were analyzed using Eq.(1). D and Mf were estimated using Eqs. (2) and (3), respectively. Non-linear curve fitting of fluorescence recovery data to Eq. (1) was carriedout using the GraphPad Prism software version 4.00 (San Diego, CA).

2.7. Isolation and confirmation of PM fraction

PM were prepared using sucrose density gradient centrifugationfrom WT and psd1Δ/Δ psd2Δ/Δ mutant C. albicans strains as describedpreviously [21]. The PM fractions (40 μg) were separated by 8%SDS-PAGE and electro-transferred to PVDF membranes. Pma1 was used toconfirm PM fraction and was detected using anti-Pma1p polyclonal

antibody. Pma1p antibodies were detected with HRP-conjugated goatanti-rabbit antibody (Santa Cruz Biotechnology Inc., Texas, USA).

2.8. Sample preparation for measurement of membrane dipole potential

Di-8-ANEPPS was added from a methanolic stock solution to PMcontaining 100 nmol total phospholipid in 1.5ml of 50mM Tris, pH 7.4buffer. The amount of di-8-ANEPPS added was such that the final probeconcentration was ~1mol% with respect to total phospholipid content.The concentration of the stock solution of di-8-ANEPPS in methanolwas estimated from its molar absorption coefficient of37,000M−1 cm−1 at 498 nm [22]. The final di-8-ANEPPS concentra-tion was 1 μM in all cases and methanol content was always low(0.05%, v/v). This ensures optimal fluorescence intensity with negli-gible membrane perturbation. Di-8-ANEPPS was added to membraneswhile being vortexed for 10 s at room temperature (~23 °C). Back-ground samples were prepared the same way except that di-8-ANEPPSwas not added to them. Samples were incubated in dark for 30min atroom temperature (~23 °C) for equilibration before measuring fluor-escence. Experiments were performed with three sets of samples induplicates at room temperature (~23 °C) and results are shown asmeans ± S.E.M. One-way ANOVA followed by Bonferroni test wasused to calculate statistical significance value.

2.9. Measurement of membrane dipole potential

Measurements were carried out by the dual wavelength ratiometricapproach using the voltage-sensitive fluorescence probe di-8-ANEPPS[23,24]. Steady state fluorescence measurements were performed witha Hitachi F-7000 (Tokyo, Japan) spectrofluorometer using 1 cm pathlength quartz cuvettes at room temperature (~23 °C). Excitation andemission slits with a bandpass of 5 nm were used for all measurements.Background intensities of samples were subtracted from each sample tocancel any contribution due to the scattering artifacts. Fluorescenceintensities were recorded at two excitation wavelengths (420 and520 nm). Emission wavelength was fixed at 670 nm. The fluorescenceratio (R), defined as the ratio of fluorescence intensities at an excitationwavelength of 420 nm to that at 520 nm (emission at 670 nm in bothcases) was calculated [25], which is a measure of membrane dipolepotential. The choice of the emission wavelength (670 nm) at the rededge of the fluorescence spectrum has previously been shown to ruleout membrane fluidity effects [26].

2.10. Thermotropic phase behavior analysis by differential scanningcalorimetry (DSC)

PM containing ~10mg of total protein were suspended in 1ml of50mM Tris buffer (pH 7.4) and used for DSC experiments. Each samplewas vortexed for 3min before carrying out experiments. Thermotropicbehavior of PM isolated from WT and psd1Δ/Δ psd2Δ/Δ was in-vestigated as described previously [27] using a MicroCal VP-DSC mi-crocalorimeter (Northampton, MA). Before running the DSC scan, eachsample was degassed for ~10min at 20 °C to avoid bubbles. All sampleswere subjected to two heating and two cooling scans between 1 and110 °C at a scan rate of 1 °C/min. In each experiment, the first heatingscan was considered for the determination of apparent phase transitiontemperature in the thermogram. Thermograms were overlaid to displaythe phase transition peaks and the overlaid plots were generated usingOrigin version 8.0 (OriginLab, Northampton, MA).

2.11. R6G accumulation assay

R6G accumulation in C. albicans strains were measured by flowcytometry. Briefly, log phase cells were starved for glucose in PBS for2 h. Now these cells were taken and OD600 was set of 0.1 in PBS. Then,R6G was added at a final concentration of 7 μM to the culture and the

N.K. Khandelwal et al. BBA - Biomembranes 1860 (2018) 2308–2319

2310

Page 4: BBA - Biomembranese-portal.ccmb.res.in/e-space/amit/Group Publications/2018 BBA.pdf · covery. FRAP experiments were performed with Gaussian spot-photo-bleaching and line-scanning

cells were incubated at 30 °C. Samples were collected at different timepoints, washed twice with ice cold PBS and analyzed with FACSort flowcytometer (Becton Dickinson Immunocytometry Systems, San Jose, CA)using FL2 filter. A total of 10,000 events were considered. The analysiswas performed using the CellQuest software (Becton DickinsonImmunocytometry Systems, San Jose, CA).

2.12. Spot based drug susceptibility assays

The susceptibilities of the C. albicans strains towards antifungaldrugs were determined by spot assay as described previously [13].Overnight grown strain on YEPD agar plates were diluted to an OD600 of0.1 in 0.9% saline. From this cell suspension, further 5-fold serial di-lutions were made, and 5 μl of cells from each dilution was spotted ondrug-containing or untreated YEPD agar plates, and the plates wereincubated at 37 °C for 36 h. Each experiment was repeated at least threetimes. Representative images are shown.

2.13. [3H]FLC accumulation assay

From 5ml overnight culture of C. albicansWT and psd1Δ/Δ psd2Δ/Δstrains 0.1 at OD600 was inoculated in 50ml YEPD and grown for 6 h at30 °C, 180 rpm shaking. After 6 h of growth cells were washed bycentrifugation and resuspended in PBS buffer. The concentration ofeach culture was checked by OD600 in Cary 100 Bio UV–Visible spec-trophotometer and set for 10 at OD600 in PBS and starved of glucose for2 h at 30 °C.

For the diffusion assay, 800 μl of glucose-starved cell culture wastaken and [3H]FLC was added to make final concentration is 100 nm.Now at different time points 100 μl of the treated cell solution wastaken out and added to a fresh 14ml round bottom tube containing 5mlice cold PBS. The solution was then filtered through a 0.45 μm cellulosenitrate filter using millipore manifold filtration assembly and washedtwice with ice cold PBS. The filters containing washed cells weretransferred to 15ml scintillation vials containing 5ml of scintillationcocktail was added and the radioactivity associated with the filter wasmeasured in liquid scintillation counter (Tri-Carb 2900TR LiquidScintillation Analyzer; Packard). All experiments were performed asbiological triplicates and results are shown as means ± S.E.M. A sta-tistical significance value was employed using one-way ANOVA fol-lowed by Bonferroni test.

3. Results

3.1. Deletion of PSD1 and PSD2 leads to higher PM fluidity in C. albicans

PE is the most abundant phospholipid in the inner leaflet of the PMbilayer [28,29]. Based on the crucial role of phospholipids as buildingblock for the PM, we hypothesized that decreased levels of PE couldimpact physical state of PM of Candida cells. To test this hypothesis, weexploited psd1Δ/Δ psd2Δ/Δ double mutant strain in which both thephosphatidylserine decarboxylases genes (PSD1 and PSD2) were de-leted. Expectedly, the psd1Δ/Δ psd2Δ/Δmutant was reported to containsignificantly lower levels of PE in comparison to WT strain [16].

In order to check the impact of PSD1 and PSD2 deletion on plasmamembrane lipid dynamics, we performed FRAP-based lateral mobilityanalysis in WT and psd1Δ/Δ psd2Δ/Δ strains by employing FAST-DiIwhich we used earlier for measuring the PM fluidity in C. albicans [11].A schematic representation of the experiment is shown in Fig. 2A. TheWT and psd1Δ/Δ psd2Δ/Δ mutant cells were incubated with lipid probeFAST-DiI as described in Materials and methods section. A region ofinterest (ROI) was selected on the PM stained with FAST-DiI and pho-tobleached. The ROI was imaged over time and recovery of FAST-DiIfluorescence was measured. The lateral movement of FAST-DiI fromunbleached region of PM to the bleached region acts as a source offluorescence recovery. Fig. 2B depicts the representative fluorescence

recovery experiment images of WT and psd1Δ/Δ psd2Δ/Δ mutant. Therate of fluorescence recovery in the bleached area is dependent on thelater movement of FAST-DiI, which is regulated, by the physical state ofthe PM. For instance, decreased membrane fluidity would lead to slowfluorescence recovery and vice versa [30,31]. As shown in the over-lapping recovery plot, in comparison to WT strain, the psd1Δ/Δ psd2Δ/Δ strain showed faster recovery of FAST-DiI fluorescence (Fig. 2C). Thecalculations of diffusion coefficient (D) and mobile fraction (Mf) usingdifferent recovery data sets (as described in Materials and methodssection) revealed that D was significantly higher in case of psd1Δ/Δpsd2Δ/Δ (1.80 ± 0.24× 10−9 cm2/s) strain as compared to WT(0.80 ± 0.07× 10−9 cm2/s) strain (Fig. 2D). Mf value was also sig-nificantly high in psd1Δ/Δ psd2Δ/Δ strain (71.46 ± 1.82%) as com-pared to WT (50.43 ± 2.20%) (Fig. 2E).

3.2. Plasma membrane protein in psd1Δ/Δ psd2Δ/Δ mutant shows enhancelateral diffusion

Lipid and protein are the two major building blocks for the PM. Theplasma membrane localized proteins involve in various important vitalfunctions. The lateral movement of PM localized proteins is usuallyvery slow and considered as a plausible reason for the yeast cell polarity[32,33]. Previously, we showed that decrease in ergosterol biosynthesisincreases plasma membrane fluidity and lipid mobility, however, it didnot have any impact on lateral mobility of PM localized ABC transporterprotein [32].

To examine the impact of increased fluidity in the psd1Δ/Δ psd2Δ/Δstrain on PM protein lateral mobility, as a test case, we overexpressed aGFP tagged PM localized ABC transporter Cdr6/Roa1 in WT and psd1Δ/Δ psd2Δ/Δ mutant background strain, designated as CA-CDR6 andNKKY104, respectively. The strains CA-CDR6 and NKKY104 wereconfirmed by PCR and western blot (Fig. S1). We performed FRAP ex-periments by exploiting GFP fluorescence of tagged Cdr6/Roa1 fusionconstruct. As depicted in Fig. 3A, a particular region of the GFP taggedPM localized protein was selected and photobleached. The same regionwas imaged over time and recovery of GFP fluorescence was measuredwhich occurs due to lateral movement of Cdr6/Roa1-GFP protein fromunbleached region of PM to the bleached region (Fig. 3A).

In Fig. 3B, the panels depict the representative images of Cdr6/Roa1-GFP protein at pre bleached, bleached and recovery stages in CA-CDR6 (WT background strain) and NKKY104 (psd1Δ/Δ psd2Δ/Δ mu-tant background strain) cells. The recovery plot depicted in Fig. 3Cshows that PM localized ABC transporter protein display slower diffu-sion as compared to lipid probe FAST-DiI (Fig. 2). However, the rate offluorescence recovery was faster in NKKY104 strain (Fig. 3C). Thecalculation of the diffusion coefficient from recovery plots in CA-CDR6and NKKY104 strain revealed that D was significantly higher inNKKY104 strain (0.039 ± 0.004× 10−9 cm2/s) in comparison to CA-CDR6 (0.022 ± 0.003×10−9 cm2/s) (Fig. 3D). Mf of Cdr6/Roa1-GFPwas similar for both CA-CDR6 and NKKY104 strains (Fig. 3E). Theseresults indicate that the initial rate of recovery of Cdr6/Roa1-GFPfluorescence is faster in NKKY104 as compared to percentage of finalrecovery, which was similar. Our FRAP experiments support that theimbalances in PE levels and change in membrane fluidity therein couldalso impact protein mobility.

3.3. psd1Δ/Δ psd2Δ/Δ mutant show low PM dipole potential

The nonrandom arrangement of amphiphile dipoles and solvent(water) molecules at the membrane interface generates membrane di-pole potential due to electrostatic potential differences [25,34,35].Since the dipole potential is generated due to ‘nonrandom’ (restricted)arrangement of lipid and solvent dipoles, the change in membranefluidity could affect the population of nonrandom dipoles, leading tochange in membrane dipole potential.

To explore the effect of PSD1 and PSD2 deletion on PM dipole

N.K. Khandelwal et al. BBA - Biomembranes 1860 (2018) 2308–2319

2311

Page 5: BBA - Biomembranese-portal.ccmb.res.in/e-space/amit/Group Publications/2018 BBA.pdf · covery. FRAP experiments were performed with Gaussian spot-photo-bleaching and line-scanning

potential of C. albicans, we prepared PM fraction from WT and psd1Δ/Δpsd2Δ/Δ strains using sucrose gradient method as described inMaterials and methods section. The quality of the PM fraction waschecked by western blot using antibody against PM specific proteinPma1 (data not shown). For the measurement of dipole potential, thePM fractions of WT and psd1Δ/Δ psd2Δ/Δ strains were stained using avoltage sensitive fluorescent probe di-8-ANEPPS [pyridinium, 4-(2-(6-dioctylamino)-2-naphthalenyl)ethenyl)-1-(3-sulfopropyl)-inner salt](Fig. 4A). The useful parameter in this method of dipole potentialmeasurement is the fluorescence intensity ratio (R) which is the ratio offluorescence intensities at two different excitation wavelengths withemission wavelength being fixed at the same wavelength. R is sensitiveto any change in the dipolar field where the potential-sensitive probe

di-8-ANEPPS is localized. Fig. 3B shows that deletion of PSD1 and PSD2genes results in a red shift of di-8-ANEPPS fluorescence excitationspectra resulting in reduction of fluorescence intensity at 420 nm andincrease in fluorescence intensity at 520 nm (marked by arrows). Toobtain a quantitative estimate of R, we plotted R (averaged over threedifferent experiments) for WT and psd1Δ/Δ psd2Δ/Δ strain (seeFig. 3c). We observed that psd1Δ/Δ psd2Δ/Δ strain had significantlylow PM dipole potential (R=0.572 ± 0.006) as compared to WT(R=0.756 ± 0.010) strain (Fig. 4B and C).

Fig. 2. Deletion of PSD1 and PSD2 leads to higher PM fluidity and lipid mobility.(A) Schematic of FRAP. C. albicans cells were stained with FAST-DiI. PM bound fraction of FAST-DiI was photobleached and the same region was imaged over time,and recovery of FAST-DiI fluorescence is measured. Lateral diffusion of FAST-DiI from unbleached region of PM results in recovery of fluorescence intensity. The timeto recover the fluorescence intensity in the bleach area is dependent on the physical state of the PM; higher membrane fluidity would cause faster fluorescencerecovery.(B) Representative images of FAST-DiI labeled PM of WT and psd1Δ/Δ psd2Δ/Δ cells (left column, t= 0 s, pre-bleach). A region of interest (ROI, yellow circle) wasphotobleached and cells were imaged immediately thereafter (second column, ‘bleach’) and after 120 s post-bleach (third column, ‘recovery’).(C) Typical fluorescence recovery curves representing the qualitative estimation of rate of recovery for WT and psd1Δ/Δ psd2Δ/Δ strains. The curves are non-linearregression fits to Eq. (1) as described in Materials and methods section. The slope of the curves show faster fluorescence recovery in psd1Δ/Δ psd2Δ/Δ strain incomparison to WT.(D) Diffusion coefficient (D) was calculated for WT and psd1Δ/Δ psd2Δ/Δ strains using Eq. (2) as described in the Materials and methods section and plotted as ascatter plot. The experiment was performed in biological triplicates and values are the means ± S.E.M. A statistical significance values P < 0.001 was calculated byone-way ANOVA followed by Bonferroni test.(E) WT and psd1Δ/Δ psd2Δ/Δ strains mobile fraction (Mf) was calculated using Eq. (3) as described in Materials and methods section. Values are means ± S.E.M. Astatistical significance values P < 0.001 was calculated by one-way ANOVA followed by Bonferroni test.

N.K. Khandelwal et al. BBA - Biomembranes 1860 (2018) 2308–2319

2312

Page 6: BBA - Biomembranese-portal.ccmb.res.in/e-space/amit/Group Publications/2018 BBA.pdf · covery. FRAP experiments were performed with Gaussian spot-photo-bleaching and line-scanning

3.4. PSD1 and PSD2 deletion impacts on thermotropic phase behavior ofplasma membrane

Membrane phase is a crucial determinant of membrane organizationand function [36–38]. Since we observed that change in PE level affectmembrane microviscosity (Fig. 2), we explored if PE levels could alsoaffect phase transition temperature of the membrane. For this, weemployed DSC, which is widely used to characterize thermotropicphase behavior of biological membranes. The overlays of DSC ther-mogram of WT and psd1Δ/Δ psd2Δ/Δ strains are depicted in Fig. 5A.The thermograms of WT and psd1Δ/Δ psd2Δ/Δ strains exhibited asingle prominent peak (Fig. 5A). However, in psd1Δ/Δ psd2Δ/Δ strainthe peak center was shifted towards lower temperature in comparisonto WT (Fig. 5A). The average of three biological replicates shows

(Fig. 5B) that in psd1Δ/Δ psd2Δ/Δ strains, the apparent phase transitionpeak is centered at (54.1 ± 0.1 °C) which is around 3.5 °C lower thanWT (57.5 ± 0.1 °C).

3.5. Fluorescent dye rhodamine 6G show enhanced diffusion in psd1Δ/Δpsd2Δ/Δ mutant

Since changes in the PM fluidity have earlier been shown to influ-ence drug diffusion into yeast cells [8–11,30], we, explored if thechanges in PM fluidity observed in the case of psd1Δ/Δ psd2Δ/Δmutantcould impact diffusion through the PM. To examine this possibility, weused R6G, a fluorescent dye which is commonly used to check the di-rection of the movement of the molecules across the PM. Deenergizedlog phase cells were incubated with R6G and at different time points,

Fig. 3. Plasma membrane protein Cdr6/Roa1-GFP in psd1Δ/Δ psd2Δ/Δ mutant shows enhance lateral diffusion.(A) Schematic of the FRAP. GFP tagged PM localized ABC transporter protein Cdr6/Roa1 expressed in C. albicans WT and psd1Δ/Δ psd2Δ/Δ strains and named as CA-CDR6 and NKKY104 respectively. In these strains, PM associated Cdr6/Roa1-GFP is photo-bleached and same region is imaged over time and recovery of Cdr6/Roa1-GFP fluorescence is measured. The lateral movement of Cdr6/Roa1-GFP from unbleached region of PM to the bleached region causes fluorescence recovery. The timeto recover the fluorescence in the bleach area is dependent on the physical state of the PM; higher membrane fluidity would lead to faster fluorescence recovery.(B) Representative images of Cdr6/Roa1-GFP expressing C. albicans CA-CDR6 and NKKY104 cells (left column, t= 0 s, pre-bleach). A Region of Interest (ROI, yellowcircle) was photo-bleached and cells were imaged immediately thereafter (second column, ‘bleach’) and after 240 s post-bleach (third column, ‘recovery’).(C) A qualitative estimate of rate of recovery can be obtained by comparing the slopes of the overlapped recovery curve, demonstrating fast recovery in NKKY104strain in comparison to CA-CDR6 strain.(D) D was calculated for CA-CDR6 and NKKY104 strains using Eq. (3) as described in Materials and methods section and plotted as a scatter plot. The experiment wasperformed in biological triplicates and values are the means ± S.E.M. A statistical significance values P= 0.022 was calculated by one-way ANOVA followed byBonferroni test.(E) Mf was calculated for CA-CDR6 and NKKY104 strains using Eq. (3) as described in Materials and methods section. Values are means ± S.E.M. of biologicaltriplicates.

N.K. Khandelwal et al. BBA - Biomembranes 1860 (2018) 2308–2319

2313

Page 7: BBA - Biomembranese-portal.ccmb.res.in/e-space/amit/Group Publications/2018 BBA.pdf · covery. FRAP experiments were performed with Gaussian spot-photo-bleaching and line-scanning

aliquots were taken and intracellular accumulation of R6G was mea-sured by flow cytometry. As depicted in Fig. 6, the psd1Δ/Δ psd2Δ/Δmutant accumulated higher amount of dye in to the cells as comparedto WT cells and this difference was intensified with time.

3.6. PE levels affects azoles diffusion and susceptibility

Azoles are most commonly used drug to treat candidasis and it hasbeen observed that the frequencies of emerging drug resistance againstazoles are high in C. albicans [39,40]. Previous reports have shown thatlipid imbalances within PM affect azole susceptibility [8–11]. With thisbackground, we explored the impact of psd1Δ/Δ psd2Δ deletion ondrug susceptibility. We performed spot assays of WT and psd1Δ/Δpsd2Δ/Δ strains in presence of different azoles; FLC, itraconazole (ITR),miconazole (MCZ) and ketoconazole (KTC) in YEPD medium. Sincepsd1Δ/Δ psd2Δ/Δ strain has been shown to be cold sensitive, for this,the spot assays were performed at 37 °C as described previously byChen et al. [16]. As depicted in Fig. 7A the psd1Δ/Δ psd2Δ/Δ strainshowed increase susceptibility to azoles as compared to WT strain. Thesusceptibility of psd1Δ/Δ psd2Δ/Δ strain towards azoles was furtherconfirmed by microdilution broth assay (Fig. S3).

As discussed above and depicted in Fig. 6, we have shown that the

diffusion of R6G dye was higher in the psd1Δ/Δ psd2Δ/Δ mutant cells.We assumed that the increase sensitivity to azoles in the psd1Δ/Δpsd2Δ/Δ mutant could also be the result of high diffusion of azolesacross the PM into the cells. To confirm it, we examined the diffusion ofFLC in WT and psd1Δ/Δ psd2Δ/Δ strains by using radiolabeled [3H]FLCas described previously [11,41]. The intracellular accumulation ofradiolabeled [3H]FLC was quantitated in the WT and psd1Δ/Δ psd2Δ/Δstrains in the absence of glucose. As depicted in Fig. 7B, a significantlyhigh level of FLC accumulated in the psd1Δ/Δ psd2Δ/Δ mutant ascompared to the WT. At early time point (10min) the FLC accumulationdifference was not significant between WT and psd1Δ/Δ psd2Δ/Δ mu-tant which became significant at later time points and was maximum at50min (Fig. 7B).

4. Discussion

Drug resistance is an alarming and evolving problem in a widerange of pathogens. Various mechanisms are involved in this phe-nomenon where pathogen becomes resilient to the drug concentrationswhich were previously sufficient to exterminate the pathogen[39,42,43]. Among various mechanisms, enhanced efflux and decreaseinflux of drugs, represent two mechanisms that are tightly regulated by

Fig. 4. The psd1Δ/Δ psd2Δ/Δ mutant shows low PM dipole potential.(A) Chemical structure di-8-ANEPPS used for measuring membrane dipole potential in this study.(B) Representative normalized excitation spectra of di-8-ANEPPS in WT (___, blue) and psd1Δ/Δ psd2Δ/Δ (—, maroon) membranes. The ratio of di-8-ANEPPS tomembrane phospholipids was 1:100 (mol/mol), and the concentration di-8-ANEPPS was 1 μM in both cases. The arrows indicate changes in di-8-ANEPPS fluor-escence intensity corresponding to excitation wavelengths of 420 and 520 nm.(C) PM dipole potential measured using di-8-ANEPPS. Change in dipole potential measured utilizing di-8-ANEPPS after incorporation into the PM of WT (blue) andpsd1Δ/Δ psd2Δ/Δ strains (maroon). Fluorescence intensity ratio (R, a measure of membrane dipole potential) is defined as the ratio of fluorescence intensities at anexcitation wavelength of 420 nm to that at 520 nm (emission at 670 nm in both cases; see Materials and methods section) was calculated from Fig. 4B. Experimentswere performed at least in triplicate and results are shown as means ± S.E.M. A statistical significance values P < 0.001 was calculated by one-way ANOVAfollowed by Bonferroni test.

N.K. Khandelwal et al. BBA - Biomembranes 1860 (2018) 2308–2319

2314

Page 8: BBA - Biomembranese-portal.ccmb.res.in/e-space/amit/Group Publications/2018 BBA.pdf · covery. FRAP experiments were performed with Gaussian spot-photo-bleaching and line-scanning

PM. In former case, PM localized drug exporters expel out the drug andin later case, the alterations in the physiochemical properties of the PMrestrict in inflow of drug (diffusion) into the cells. Human pathogenicyeast C. albicans possesses battery of multidrug efflux transporters, ofwhich several have been characterized and belong to either ABC su-perfamily (Cdr1, Cdr2 and Cdr6) or MFS superfamily (Mdr1 and Flu1)[11,44–47], however, how various factors such as the PM physical stateaffects drug susceptibility of Candida cells remain poorly understood.The present study focuses on impact of the nature of fluidity of PMassociated with imbalances in PE levels in drug susceptibility of C. al-bicans cells.

We have previously shown that the membrane ergosterol can affectthe PM fluidity and its interaction with sphingolipid is a crucial de-terminant of drug susceptibilities in C. albicans cells [30,48]. Interest-ingly, the increase in PM fluidity in erg mutants did not impact thelateral diffusion of PM localized ABC transporter Cdr1p [32]. However,our present study shows that the increase in membrane fluidity inpsd1Δ/Δ psd2Δ/Δ cells does impact the lateral diffusion of tested PMlocalized ABC protein (Fig. 3). This could be the result of specificchanges in physiochemical properties of PM of psd1Δ/Δ psd2Δ/Δ cells,

which in comparison to ergosterol depletion is more severe. Notably, ininsect cells, PE is considered as a key factor in regulating PM rigidity[49]. Recently, in C. albicans, we also observed that in a PM ABCtransporter knockout strain with high PM rigidity was linked to an in-crease in PE levels [11]. Together, apart from the imbalances in er-gosterol levels, fluctuations in PE contents also impact physiochemicalproperties of yeast membrane influencing lateral mobility of PM loca-lized proteins.

Our data show that apart from changes in fluidity, reduction indipole potential was also observed in psd1Δ/Δ psd2Δ/Δ cells. In highereukaryotic membranes, an impact of cholesterol on PM dipole potentialis known where an increase in cholesterol level has been shown to causeincrease in dipole potential [25,35]. Since PE level impacts on lipiddynamics and PM rigidity in psd1Δ/Δ psd2Δ/Δ cells, the changes inlipid packing density could be a plausible factor for observed change inthe PM dipole potential.

The thermotropic phase behavior is an important physical char-acteristic property of the PM and any change in PM lipid compositionhas marked effect on this property [50]. Fluctuations in membranesterol level are reported to impact thermotropic phase behavior. Wealso noticed that changes in PE levels in psd1Δ/Δ psd2Δ/Δ cells result inlower phase transition temperature (Fig. 5). As PE depletion increases

Fig. 5. PSD1 and PSD2 deletion in C. albicans causes decrease in membranephase transition temperature.(A) DSC first heating thermograms of WT (blue) and psd1Δ/Δ psd2Δ/Δ(maroon) strains membrane. Thermograms were overlaid to display the ap-parent phase transition peaks.(B) Apparent phase transition temperature of WT (blue) and psd1Δ/Δ psd2Δ/Δ(maroon) strain membranes. Apparent phase transition temperatures representmeans ± S.E.M. of three independent experiments. A statistical significancevalues P < 0.001 was calculated by one-way ANOVA followed by Bonferronitest.

Fig. 6. Fluorescent dye R6G show enhanced diffusion in psd1Δ/Δ psd2Δ/Δmutant.C. albicans WT and psd1Δ/Δ psd2Δ/Δ strain were incubated with 10 μM of R6G.After 30, 60, and 90min sample were collected and R6G accumulation in cellswere analyzed with flow cytometry. Data of one experiment is shown out ofthree biological experiments. The other two biological replicates are shown inFig. S2.

N.K. Khandelwal et al. BBA - Biomembranes 1860 (2018) 2308–2319

2315

Page 9: BBA - Biomembranese-portal.ccmb.res.in/e-space/amit/Group Publications/2018 BBA.pdf · covery. FRAP experiments were performed with Gaussian spot-photo-bleaching and line-scanning

the fluidity, this leads to enhancement in liquid ordered phase con-tributing to the observed decrease in the apparent phase transitiontemperature. The psd1Δ/Δ psd2Δ/Δ showed growth defect at lowtemperature because of mitochondria dysfunction [16] and decreasingin the phase transition temperature in psd1Δ/Δ psd2Δ/Δ cells could actas adaptation method to survive at low temperature stress. To the bestof our knowledge, this type of PE dependent change in phase transitionin fungal pathogen PM is a novel observation. Since membrane orga-nization represents an important determinant in drug resistance, ourobservation (PE dependent change in phase transition) could providecrucial insight.

This increase in fluidity shows impact on fluorescence dye R6Gdiffusion, where high accumulation of the dye is observed in psd1Δ/Δpsd2Δ/Δ mutant cells. It is well established fact that lipid imbalances

within PM could affect azoles susceptibility of the cell [8–11]. Here weobserved that homozygous psd1Δ/Δ psd2Δ/Δ mutant display decreaseresistance to azoles which was well corroborated by the observed high[3H]FLC accumulation in mutant cells. This would imply that an in-crease in PM fluidity in psd1Δ/Δ psd2Δ/Δ mutant results in enhanceddiffusion of drugs leading to increase susceptibility towards azoles.

A large number of drug resistance cases have been seen during C.albicans infection treatment and various clinical isolates have beenobtained from the patient over the time of therapy. The isolates, whichwere collected before treatment, are usually susceptible to drugs, whilethe surviving isolates obtained after drug therapy are often resistant todrugs. When these isolates collected from a single patient they areconsidered as clinical match pair isolates and considered a very usefultool to understand the development of drug tolerance mechanism in

Fig. 7. psd1Δ/Δ psd2Δ/Δ mutant shows increasedazoles susceptibility and drug diffusion.(A) A comparison of growth by spot dilution assaysbetween WT and psd1Δ/Δ psd2Δ/Δ strains. A 5-foldserial dilution of each strain was spotted on YEPDagar plates containing the triazoles (FLC, ITR) andimidazoles (KTC, and MCZ) at the indicated con-centrations, and grown for 36 h at 37 °C.(B) [3H]FLC accumulation levels in C. albicans WT,and psd1Δ/Δ psd2Δ/Δ strains were measured atmentioned time points in the absence of glucose inglucose-starved cells. Experiments were performedin biological triplicate and the results are shown asmeans ± S.E.M. A statistical significance value wasemployed using one-way ANOVA followed byBonferroni test.

N.K. Khandelwal et al. BBA - Biomembranes 1860 (2018) 2308–2319

2316

Page 10: BBA - Biomembranese-portal.ccmb.res.in/e-space/amit/Group Publications/2018 BBA.pdf · covery. FRAP experiments were performed with Gaussian spot-photo-bleaching and line-scanning

real clinical treatment scenario. Previously, our group have studied onesuch pair TW1 (drug susceptible) versus TW17 (drug resistant) andobserved that PSD1 and PSD2 transcript levels were found to be in-creased in TW17 isolate as compared to susceptible isolate (TW1) [51].We have checked the expression of PSD1 and PSD2 in another clinicalmatch pair isolates GU4 (drug susceptible) versus GU5 (drug resistant)and observed that in GU5 strain, the expression of PSD2 is up-regulatedas compared to GU4 (Fig. S4). The high expression of PSD genes in drugresistant isolate points that increased level of these enzymes may con-tribute to an increase in PE content which ultimately could impact azoleresistance.

Development of drug resistance not only depends on pathogenproperties, but also depends on the nature of the drug. In fungi, usuallyit has been observed that clinical drug resistance cases occur frequentlydue to the use of fungistatic drugs as compared to fungicidal drugs. Tohandle this problem, a combinatorial drug therapy is considered as oneof the rising solution in which fungistatic drugs are used along withother compounds and this combination of drug becomes fungicidal tothe pathogen. A recent work has reported a fluorescence based methodfor high throughput screening of phosphatidylserine decarboxylase in-hibitors [52] and can be further exploited for identification of novelinhibitors. Our present study provides a mechanistic insight thatphosphatidylserine decarboxylase inhibition can increase the suscept-ibility to azole drugs (Fig. 8). A combination of phosphatidylserinedecarboxylase inhibitor, along with azole drugs therefore can be a goodcombinatorial drug therapy option against candidiasis.

In summary, we present importance of PE in PM physiochemicalproperties and provide evidence that PE depletion increase PM fluidity,decrease in PM dipole potential and the thermotropic phase tempera-ture, thereby increasing drug diffusion and susceptibility (Fig. 8).

Transparency document

The Transparency document associated with this article can befound, in online version.

Acknowledgement

Authors are thankful to Ranbaxy Laboratories for the generous giftof fluconazole. The authors also thank Dr. Todd B. Reynolds (Knoxville,USA) for kind gift of strains. Pma1p antibody was a kind gift from Dr.Ramon Serrano (Valencia, Spain). N.K.K. is thankful to Senior ResearchFellowship Award from University Grants Commission, India andResearch Associate Fellowship from ICGEB-INDIA. P.S. thanks theCouncil of Scientific and Industrial Research for the award of a ShyamaPrasad Mukherjee Fellowship. We thank Dr. S. Thirupathi Reddy forhelp with DSC experiments and Dr. Sarika Gupta for help with FACSexperiments.

Funding

The work has been supported in parts by grants to R.P. from theDepartment of Biotechnology DBT [No. BT/PR7392/MED/29/652/2012] and DBT [No. BT/PR14879/BRB10/885/2010]. This work wassupported by the Council of Scientific and Industrial Research, Govt. ofIndia. A.C. gratefully acknowledges J.C. Bose Fellowship (Departmentof Science and Technology, Govt. of India).

Conflict of interest

Authors declares no conflict of interest.

Fig. 8. Proposed model for C. albicans Psd1 and Psd2.PS is converted to PE in C. albicans in de novo by Psd1 and Psd2 in mitochondria and Golgi vesicle respectively. PE acts as a key regulator of PM rigidity and regulatesdiffusion of drugs in to the cells.

N.K. Khandelwal et al. BBA - Biomembranes 1860 (2018) 2308–2319

2317

Page 11: BBA - Biomembranese-portal.ccmb.res.in/e-space/amit/Group Publications/2018 BBA.pdf · covery. FRAP experiments were performed with Gaussian spot-photo-bleaching and line-scanning

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.bbamem.2018.05.016.

References

[1] R. Calderone, N. Sun, F. Gay-Andrieu, W. Groutas, P. Weerawarna, S. Prasad, et al.,Antifungal drug discovery: the process and outcomes, Future Microbiol 9 (2014)791–805, http://dx.doi.org/10.2217/fmb.14.32.

[2] M.A. Pfaller, D.J. Diekema, Epidemiology of invasive candidiasis: a persistentpublic health problem, Clin. Microbiol. Rev. 20 (2007) 133–163, http://dx.doi.org/10.1128/CMR.00029-06.

[3] T.E. Zaoutis, J. Argon, J. Chu, J.A. Berlin, T.J. Walsh, C. Feudtner, The epide-miology and attributable outcomes of candidemia in adults and children hospita-lized in the United States: a propensity analysis, Clin. Infect. Dis. 41 (2005)1232–1239, http://dx.doi.org/10.1086/496922.

[4] L.E. Cowen, The evolution of fungal drug resistance: modulating the trajectory fromgenotype to phenotype, Nat. Rev. Microbiol. 6 (2008) 187–198, http://dx.doi.org/10.1038/nrmicro1835.

[5] D. Sanglard, K. Kuchler, F. Ischer, J.L. Pagani, M. Monod, J. Bille, Mechanisms ofresistance to azole antifungal agents in Candida albicans isolates from AIDS patientsinvolve specific multidrug transporters, Antimicrob. Agents Chemother. 39 (1995)2378–2386.

[6] T.C. White, Increased mRNA levels of ERG16, CDR, and MDR1 correlate with in-creases in azole resistance in Candida albicans isolates from a patient infected withhuman immunodeficiency virus, Antimicrob. Agents Chemother. 41 (1997)1482–1487.

[7] C.N. Lyons, T.C. White, Transcriptional analyses of antifungal drug resistance inCandida albicans, Antimicrob. Agents Chemother. 44 (2000) 2296–2303.

[8] A. Kohli, K. Smriti, A. Mukhopadhyay, R. Prasad Rattan, In vitro low-level re-sistance to azoles in Candida albicans is associated with changes in membrane lipidfluidity and asymmetry, Antimicrob. Agents Chemother. 46 (2002) 1046–1052.

[9] K. Mukhopadhyay, A. Kohli, R. Prasad, Drug Susceptibilities of Yeast Cells AreAffected By Membrane Lipid Composition Drug Susceptibilities of Yeast Cells AreAffected By Membrane Lipid Composition, 46 (2002), http://dx.doi.org/10.1128/AAC.46.12.3695.

[10] T. Prasad, A. Chandra, C.K. Mukhopadhyay, R. Prasad, Unexpected link betweeniron and drug resistance of Candida spp.: iron depletion enhances membranefluidity and drug diffusion, leading to drug-susceptible cells, Antimicrob. AgentsChemother. 50 (2006) 3597–3606, http://dx.doi.org/10.1128/AAC.00653-06.

[11] N.K. Khandelwal, N. Chauhan, P. Sarkar, B.D. Esquivel, P. Coccetti, A. Singh, et al.,Azole resistance in a Candida albicans mutant lacking the ABC transporter CDR6/ROA1 depends on TOR signaling, J. Biol. Chem. (2017), http://dx.doi.org/10.1074/jbc.M117.807032.

[12] K.K. Mahto, A. Singh, N.K. Khandelwal, N. Bhardwaj, J. Jha, R. Prasad, An as-sessment of growth media enrichment on lipid metabolome and the concurrentphenotypic properties of Candida albicans, PLoS One 9 (2014) e113664, , http://dx.doi.org/10.1371/journal.pone.0113664.

[13] N.K. Khandelwal, P. Kaemmer, T.M. Forster, A. Singh, A.T. Coste, D.R. Andes, et al.,Pleiotropic effects of the vacuolar ABC transporter MLT1 of Candida albicans on cellfunction and virulence, Biochem. J. (2016), http://dx.doi.org/10.1042/BCJ20160024.

[14] D.P. Siegel, R.M. Epand, The mechanism of lamellar-to-inverted hexagonal phasetransitions in phosphatidylethanolamine: implications for membrane fusion me-chanisms, Biophys. J. 73 (1997) 3089–3111, http://dx.doi.org/10.1016/S0006-3495(97)78336-X.

[15] M.K. Storey, K.L. Clay, T. Kutateladze, R.C. Murphy, M. Overduin, D.R. Voelker,Phosphatidylethanolamine has an essential role in Saccharomyces cerevisiae that isindependent of its ability to form hexagonal phase structures, J. Biol. Chem. 276(2001) 48539–48548, http://dx.doi.org/10.1074/jbc.M109043200.

[16] Y.-L. Chen, A.E. Montedonico, S. Kauffman, J.R. Dunlap, F.-M. Menn, T.B. Reynolds,Phosphatidylserine synthase and phosphatidylserine decarboxylase are essential forcell wall integrity and virulence in Candida albicans, Mol. Microbiol. 75 (2010)1112–1132, http://dx.doi.org/10.1111/j.1365-2958.2009.07018.x.

[17] T.J. Pucadyil, A. Chattopadhyay, Effect of cholesterol on lateral diffusion of fluor-escent lipid probes in native hippocampal membranes, Chem. Phys. Lipids 143(2006) 11–21, http://dx.doi.org/10.1016/j.chemphyslip.2006.04.003.

[18] T.J. Pucadyil, S. Kalipatnapu, K.G. Harikumar, N. Rangaraj, S.S. Karnik,A. Chattopadhyay, G-protein-dependent cell surface dynamics of the humanserotonin1A receptor tagged to yellow fluorescent protein, Biochemistry 43 (2004)15852–15862, http://dx.doi.org/10.1021/bi0480887.

[19] D.E. Koppel, M.P. Sheetz, M. Schindler, Lateral diffusion in biological membranes.A normal-mode analysis of diffusion on a spherical surface, Biophys. J. 30 (1980)187–192, http://dx.doi.org/10.1016/S0006-3495(80)85087-9.

[20] O. Reuss, A. Vik, R. Kolter, J. Morschhauser, The SAT1 flipper, an optimized tool forgene disruption in Candida albicans, Gene 341 (2004) 119–127, http://dx.doi.org/10.1016/j.gene.2004.06.021.

[21] S. Shukla, P. Saini, S. Jha, V. Ambudkar, R. Prasad, S.V. Ambudkar, Functionalcharacterization of Candida albicans ABC transporter Cdr1p functional character-ization of Candida albicans ABC transporter Cdr1p, 2 (2003) 1361–1375, http://dx.doi.org/10.1128/EC.2.6.1361.

[22] G. Le Goff, M.F. Vitha, R.J. Clarke, Orientational polarisability of lipid membranesurfaces, Biochim. Biophys. Acta 1768 (2007) 562–570, http://dx.doi.org/10.

1016/j.bbamem.2006.10.019.[23] E. Gross, R.S.J. Bedlack, L.M. Loew, Dual-wavelength ratiometric fluorescence

measurement of the membrane dipole potential, Biophys. J. 67 (1994) 208–216,http://dx.doi.org/10.1016/S0006-3495(94)80471-0.

[24] P. Sarkar, H. Chakraborty, A. Chattopadhyay, Differential membrane dipolar or-ientation induced by acute and chronic cholesterol depletion, Sci. Rep. 7 (2017)4484, , http://dx.doi.org/10.1038/s41598-017-04769-4.

[25] T. Starke-Peterkovic, N. Turner, M.F. Vitha, M.P. Waller, D.E. Hibbs, R.J. Clarke,Cholesterol effect on the dipole potential of lipid membranes, Biophys. J. 90 (2006)4060–4070, http://dx.doi.org/10.1529/biophysj.105.074666.

[26] R.J. Clarke, D.J. Kane, Optical detection of membrane dipole potential: avoidanceof fluidity and dye-induced effects, Biochim. Biophys. Acta 1323 (1997) 223–239.

[27] S.T. Reddy, S. Shrivastava, K. Mallesham, A. Chattopadhyay, Cholesterol-dependentthermotropic behavior and organization of neuronal membranes, Biochim. Biophys.Acta 1858 (2016) 2611–2616, http://dx.doi.org/10.1016/j.bbamem.2016.07.007.

[28] H.I. Ingolfsson, M.N. Melo, F.J. van Eerden, C. Arnarez, C.A. Lopez, T.A. Wassenaar,et al., Lipid organization of the plasma membrane, J. Am. Chem. Soc. 136 (2014)14554–14559, http://dx.doi.org/10.1021/ja507832e.

[29] B. Fadeel, D. Xue, The ins and outs of phospholipid asymmetry in the plasmamembrane: roles in health and disease, Crit. Rev. Biochem. Mol. Biol. 44 (2009)264–277, http://dx.doi.org/10.1080/10409230903193307.

[30] K. Mukhopadhyay, T. Prasad, P. Saini, T.J. Pucadyil, A. Chattopadhyay, R. Prasad,Membrane sphingolipid-ergosterol interactions are important determinants ofmultidrug resistance in Candida albicans, Antimicrob. Agents Chemother. 48 (2004)1778–1787.

[31] A. Chattopadhyay, M. Jafurulla, Novel insights in membrane biology utilizingfluorescence recovery after photobleaching, Adv. Exp. Med. Biol. 842 (2015)27–40, http://dx.doi.org/10.1007/978-3-319-11280-0_3.

[32] S. Ganguly, P. Singh, R. Manoharlal, R. Prasad, A. Chattopadhyay, Differentialdynamics of membrane proteins in yeast, Biochem. Biophys. Res. Commun. 387(2009) 661–665, http://dx.doi.org/10.1016/j.bbrc.2009.07.054.

[33] J. Valdez-Taubas, H.R.B. Pelham, Slow diffusion of proteins in the yeast plasmamembrane allows polarity to be maintained by endocytic cycling, Curr. Biol. 13(2003) 1636–1640.

[34] R.J. Clarke, The dipole potential of phospholipid membranes and methods for itsdetection, Adv. Colloid Interf. Sci. 89-90 (2001) 263–281.

[35] S. Haldar, R.K. Kanaparthi, A. Samanta, A. Chattopadhyay, Differential effect ofcholesterol and its biosynthetic precursors on membrane dipole potential, Biophys.J. 102 (2012) 1561–1569, http://dx.doi.org/10.1016/j.bpj.2012.03.004.

[36] G. Van Meer, D. Voelker, G. Feigenson, Membrane lipids - where they are and howthey behave.pdf, Nat. Rev. Mol. … 9 (2008) 112–124, http://dx.doi.org/10.1038/nrm2330.Membrane.

[37] G.W. Feigenson, Phase behavior of lipid mixtures, Nat. Chem. Biol. 2 (2006)560–563, http://dx.doi.org/10.1038/nchembio1106-560.

[38] S. Shrivastava, D. Dutta, A. Chattopadhyay, Effect of local anesthetics on the or-ganization and dynamics in membranes of varying phase: a fluorescence approach,Chem. Phys. Lipids 198 (2016) 21–27, http://dx.doi.org/10.1016/j.chemphyslip.2016.04.009.

[39] R. Prasad, A. Banerjee, N.K. Khandelwal, S. Dhamgaye, The ABCs of Candida albi-cansmultidrug transporter Cdr1, Eukaryot. Cell (2015), http://dx.doi.org/10.1128/EC.00137-15 (EC.00137–15).

[40] A. Coste, A. Selmecki, A. Forche, D. Diogo, M.-E. Bougnoux, C. d'Enfert, et al.,Genotypic evolution of azole resistance mechanisms in sequential Candida albicansisolates, Eukaryot. Cell 6 (2007) 1889–1904, http://dx.doi.org/10.1128/EC.00151-07.

[41] B.E. Mansfield, H.N. Oltean, B.G. Oliver, S.J. Hoot, S.E. Leyde, T.C. White, Azoledrugs are imported by facilitated diffusion in Candida albicans and other pathogenicfungi, 6 (2010) 1–11, http://dx.doi.org/10.1371/journal.ppat.1001126.

[42] D. Sanglard, Emerging threats in antifungal-resistant fungal pathogens, Front. Med.3 (2016), http://dx.doi.org/10.3389/fmed.2016.00011.

[43] L.E. Cowen, D. Sanglard, S.J. Howard, P.D. Rogers, D.S. Perlin, Mechanisms ofantifungal drug resistance, Cold Spring Harb. Perspect. Med. 5 (2015) a019752, ,http://dx.doi.org/10.1101/cshperspect.a019752.

[44] R. Prasad, P. De Wergifosse, A. Goffeau, E. Balzi, Molecular cloning and char-acterization of a novel gene of Candida albicans, CDR1, conferring multiple re-sistance to drugs and antifungals, Curr. Genet. 27 (1995) 320–329.

[45] D. Sanglard, F. Ischer, M. Monod, J. Bille, Cloning of Candida albicans genes con-ferring resistance to azole antifungal agents: characterization of CDR2, a newmultidrug ABC transporter gene, Microbiology 143 (Pt 2) (1997) 405–416, http://dx.doi.org/10.1099/00221287-143-2-405.

[46] D. Calabrese, J. Bille, D. Sanglard, A novel multidrug efflux transporter gene of themajor facilitator superfamily from Candida albicans (FLU1) conferring resistance tofluconazole, Microbiology 146 (Pt 1) (2000) 2743–2754, http://dx.doi.org/10.1099/00221287-146-11-2743.

[47] R. Ben-Yaacov, S. Knoller, G.A. Caldwell, J.M. Becker, Y. Koltin, Candida albicansgene encoding resistance to benomyl and methotrexate is a multidrug resistancegene, Antimicrob. Agents Chemother. 38 (1994) 648–652.

[48] K. Mukhopadhyay, A. Kohli, R. Prasad, Drug susceptibilities of yeast cells are af-fected by membrane lipid composition, Antimicrob. Agents Chemother. 46 (2002)3695–3705.

[49] R. Dawaliby, C. Trubbia, C. Delporte, C. Noyon, J.-M. Ruysschaert, P. VanAntwerpen, et al., Phosphatidylethanolamine is a key regulator of membranefluidity in eukaryotic cells, J. Biol. Chem. 291 (2016) 3658–3667, http://dx.doi.org/10.1074/jbc.M115.706523.

[50] M.F. Renne, A.I.P.M. de Kroon, The role of phospholipid molecular species in de-termining the physical properties of yeast membranes, FEBS Lett. (2017), http://dx.

N.K. Khandelwal et al. BBA - Biomembranes 1860 (2018) 2308–2319

2318

Page 12: BBA - Biomembranese-portal.ccmb.res.in/e-space/amit/Group Publications/2018 BBA.pdf · covery. FRAP experiments were performed with Gaussian spot-photo-bleaching and line-scanning

doi.org/10.1002/1873-3468.12944.[51] A. Singh, V. Yadav, R. Prasad, Comparative lipidomics in clinical isolates of Candida

albicans reveal crosstalk between mitochondria, cell wall integrity and azole re-sistance, PLoS One 7 (2012) e39812, , http://dx.doi.org/10.1371/journal.pone.0039812.

[52] J.-Y. Choi, Y.V. Surovtseva, S.M. Van Sickle, J. Kumpf, U.H.F. Bunz, C. Ben

Mamoun, et al., A novel fluorescence assay for measuring phosphatidylserine dec-arboxylase catalysis, J. Biol. Chem. 293 (2018) 1493–1503, http://dx.doi.org/10.1074/jbc.RA117.000525.

[53] A.M. Gillum, E.Y. Tsay, D.R. Kirsch, Isolation of the Candida albicans gene for or-otidine-5′-phosphate decarboxylase by complementation of S. cerevisiae ura3 and E.coli pyrF mutations, Mol Gen Genet 198 (1984) 179–182.

N.K. Khandelwal et al. BBA - Biomembranes 1860 (2018) 2308–2319

2319