Effect of β-Citronellol on Hydrolytic Enzyme Secretion, Ergosterol Biosynthesis and Biofilm Formation in C. albicans

Antifungal potential of β-citronellol was studied on C. albicans, a commensal that causes both superficial and systemic infection in immunocompromised humans. This non-toxic natural compound displayed a minimum inhibitory concentration (MIC) of 200 μg/ml and was fungicidal at 350 μg/ml. On YPD agar, it formed inhibition zones (ZOI) of 16.2 mm, 18.15 mm and 19.25 mm, respectively at MIC, 2MIC and 3MIC while fluconazole (FLC) formed ZOI of 20.43 mm at its MIC (disc diffusion assay). No growth was observed above 2MIC on solid media (spot assay) and ≥ MIC in liquid YPD media. Hydrolytic enzyme secretion decreased in the presence of β-citronellol and was concentration dependent. At MIC, the reduction in phospholipase activity (54.08 %) was greater than the reduction in proteinase activity (40.6 %). There was concentration dependent decrease in total ergosterol content by 19 %, 40 %, 91 % and 100 %, respectively at MIC/8, MIC/4, MIC/2 and MIC values of β-citronellol. FLC at MIC showed an inhibition of only 64%. Biofilm formation reduced by 71.13% at MIC. β-citronellol, hence has immense antifungal potential and significantly inhibits growth, ergosterol levels, hydrolytic enzyme secretion, adhesion and biofilm formation in C. albicans at MIC

There are treatment issues due to multidrug resistance, high health care costs and low drug efficacy (Gow and Yadav, 2017). Therefore, a lot of work has to be done in the field of antifungal drug discovery and development. β-citronellol, a natural cyclic monoterpenoid, possesses, anti-inflammatory (Brito et al., 2012), insect repellent (Semmler et al., 2014), larvicidal (Hierro et al., 2004), and anti-bacterial (Lopez-Romero et al., 2015) properties. Like geraniol, it is a major component of rose oil and a number of other plant essential oils. Chemically, β-citronellol (dihydrogeraniol) is related to geraniol whose anti-Candida activity and mode of action was studied recently (Sharma et al., 2016). Although its antifungal potential has been reported (Pereira Fde et al., 2015) its mode of action is not clearly discussed. In the present study, we observed the effect of varying concentrations of β-citronellol on growth pattern, secretion of proteinases and phospholipases secretion, membrane ergosterol content and biofilm formation in C. albicans ATCC 10261.

II.
MATERIAL AND METHODS C. albicans ATCC 10261 was maintained on YPD media constituting yeast extract, peptone, dextrose, in a percentage ratio of 1:2:2 supplemented with 2.5% agar at 4°C. All chemicals were of analytical grade purchased from Merck (India). The media components, β-citronellol and fluconazole (FLC) were purchased from Sigma Aldrich (Germany).

Antifungal drug susceptibility assays
Broth dilution method (CLSI, 2008) was used to determine the minimum inhibitory concentrations (MIC) defined as the least possible concentration that causes 90% decrease in absorbance in comparison to that of the control (without test compound). After getting the MIC value, 15 µL aliquots were removed from tubes that show absolutely no growth along with the last tube showing growth. These were subcultured on YPD agar plates and incubated at 35 ºC until growth was visible in the control samples. The minimum fungicidal concentration or MFC value was determined as the minimum concentration of the test compound for which there was no visible growth (Samber et al., 2015). Results were calculated as mean of the two separate experiments with three different values.
To study antifungal susceptibility to β-citronellol (Ansari et al., 2014) by spot assay, Candida cells were grown overnight in YPD media at 37 ºC, and suspended in 0.9% NaCl to achieve an optical density of 0.1 at 600 nm. Aliquots (5 µL) of five-fold serially diluted cultures were pipetted on YPD agar plates in the absence (control) and presence of MIC and sub-MIC values of β-citronellol. After 48 h of incubation at 30 ºC, growth variations were observed. Candida cells (1x10 5 cells/ml) were inoculated into YPD agar at 40 ºC and poured into 90 mm petriplates to perform disc diffusion assay. Sterile filter discs (4mm) were laden with three different concentrations of test compound (MIC, 2MIC and 3MIC) and placed on agar plates. After 48 h, the average diameter of inhibition zones (ZOI) was measured. A disc impregnated with FLC at its MIC value was used as positive control.
Candida cells were sub-cultured at least two times and grown till stationary phase is reached at 35 °C on SDA plates. To study growth pattern, cell culture (A595 = 0.1) was inoculated into fresh media along with β-citronellol at MIC and sub-MIC concentrations in 50 ml total volume. Growth was monitored at 37 °C, 200 rpm and recorded after every 2 h till 48 h. Absorbance was recorded at 595 nm for each concentration using Labo-med Inc. spectrophotometer (USA) and plotted against time in hours.

WST-1 based cytotoxicity assay
The assay was performed as described previously (Khan et al., 2010). Cell culture (1×10 5 cells/ml) was taken along with β-citronellol (MIC and sub-MIC values) in 96-well plate (final volume ~ 100 µl/well) and incubated for 24 h. WST-1/CEC dye (10 µL) was added to each well and plates were again incubated at 37 ºC for 2 h with shaking. The reaction was stopped by adding 10 µL of 1% SDS. WST-1 salt was reduced to red coloured formazan by cellular dehydrogenases (Tsukatani et al., 2008), the absorbance of which was recorded at 450 nm using a micro plate Reader (BIORAD iMark, US) (reference was set at 655 nm). Experiment was repeated thrice and cytotoxicity was calculated using the following equation: % Cytotoxicity = [(cell control -cell with test compound)/(cell control)] ×100

Secretion of hydrolytic enzymes
For assessing the activity of proteinases and phospholipases, Candida cells were first inoculated into 5 ml YPD media and incubated for 18 h at 37 °C (Khan et al., 2014). Subsequently, cells were separated from culture media, washed twice and resuspended in 0.9% NaCl. Cell suspensions (MacFarland 0.5 index) were exposed to desired concentrations of β-citronellol ( 1 /8 MIC, ¼ MIC, ½ MIC and MIC). In case of proteinase assay, small aliquots (2 µl) were placed at equidistant points on agar plates (2% agar, 0.2 g BSA, yeast nitrogen base, 20 g glucose, and distilled water to a final volume of 1000 ml) while in case of phospholipase assay, aliquots were pipetted on agar peptone media (2 % agar, 10 g peptone, 30 g glucose, 57.3 g NaCl, 0.55 g CaCl2, and distilled water to a final volume of 900 ml) enriched with 10 % (v/v) egg yolk emulsion (HiMedia). Plates were incubated at 37 °C till noticeable growth (2-4 days). Enzyme secretory activity was estimated by measuring degradation/precipitation zones formed and calculated in terms of Pz values which is the ratio of the colony diameter to the colony diameter plus diameter of the zone of degradation/precipitation (Price et al., 1982).

Adhesion and biofilm formation
These experiments were performed according to Ramage et al., 2002 with modifications. An aliquot of 100 µl cell suspension (1x10 7 cells/ml) in RPMI 1640 media was placed into each well of a sterile 96-well microtiter plate. β-citronellol was added to each well at desired concentrations except control. Plates were incubated at 37ºC with gentle shaking to allow the cells to adhere to plate walls. After 90 min, nonadherent cells were washed out of each well with PBS (150µl). To see effect on biofilm formation, after the initial 1 h adhesion period, freshly prepared media (RPMI 1640) containing desired concentrations of βcitronellol were added to the wells containing adherent cells. The plates were then incubated for 24 h at 37°C. Estimation was done by semi-quantitative XTT reduction assay. A saturated solution of XTT (in PBS) was mixed with menadione-acetone solution (electron mediator). Pre-formed Candida bio films were washed with PBS first and taken in a 96-well plate with or without β-citronellol. A 100 μl aliquot of XTT was then added to each well and mixed gently. The plates were incubated in dark for 5 h at 37 °C after which absorbance of each sample was recorded at 450 nm using a micro titre plate reader (BIO-RAD, iMark, US). The wells containing PBS only or PBS + XTT were used as blank. The results were expressed as percentage viability.

Sterol extraction and quantitation
Sterol content of treated and untreated Candida cells was evaluated as discussed earlier with slight changes (Sharma et al., 2016). Briefly, 50 ml of SD broth (along with varying concentrations β-citronellol) was inoculated with a single colony from a culture plate grown overnight. After an incubation of 16 h at 35 ºC (180 rpm), cells were harvested and washed with sterile distilled water. To each pre-weighed pellet, 3 ml of 25% alcoholic KOH was added and vortexed for 60 sec. Cell suspensions were then poured into sterile borosilicate glass tubes and incubated at 85 ºC for 1 h. Tubes were cooled to room temperature and sterols were extracted by adding sterile distilled water and n-heptane in the ratio of 1:3. The contents were mixed thoroughly before transferring the heptane layer to a clean borosilicate glass tube and stored at 20 ºC for 24 h. The extracted sterols were diluted fivefold in 100% ethanol and scanned spectrophotometrically between 240 and 300 nm using Labomed, Inc. spectrophotometer (USA). Both a positive control (FLC) and a negative control (without test compound) were also included. The ergosterol content (%) per wet weight of cells was calculated as done in previous studies (Sharma et al., 2016).

Statistical Analysis
All the assays were conducted in triplicate, and the results were exhibited as mean ± standard deviation. The student's t-test was used to verify statistical significance (p < 0.05).

III.
RESULTS AND DISCUSSION 3.1. Antifungal susceptibility of β-citronellol against C. albicans The antifungal efficacy of β-citronellol was estimated by performing drug susceptibility tests on C. albicans ATCC 10261. MIC was evaluated using broth micro dilution assay. Candida cells showed susceptibility to β-citronellol at an MIC of 200 µg/ml while the conventional drug FLC gave an MIC of 10 µg/ml. This natural monoterpenoid was fungicidal above 350 µg/ml (MFC). Disc diffusion assay was performed where sterile filter discs were impregnated with the test compound and placed on the YPD agar surface. After incubation the diameters of the zone of inhibition (ZOI) formed around the discs were measured and found to be concentration dependent (Fig. 1). The diameters were 16 The susceptibility of C. albicans to β-citronellol was confirmed by the spot assay also (Fig. 1). In comparison to control, a significant decrease in growth was observed at MIC/2. At MIC, growth was observed only for the first two dilutions. No growth was seen at and above 2MIC. Growth pattern of Candida cells was studied in the absence and presence of the test compound at different concentrations. The inhibitory effect was concentration dependent leading to significant decline in growth of cells with late lagphase, undifferentiated and delayed exponential-phase (Fig. 1). On the other hand, control cells showed a regular growth pattern with a 4 h lag-phase followed by an exponential-phase of 8-10 h and a stationaryphase. The growth was significantly suppressed in the presence of β-citronellol at sub-MIC concentrations of MIC/2 and MIC/4 while at MIC, the growth curve was seen as a flat line similar to that of FLC. At MIC/8 the growth pattern was similar to the control although the growth was inhibited to some extent. All the three growth studies corroborated well with each other.

3.2.
Effect of β-citronellol on secretion of hydrolytic enzymes Extracellular hydrolytic enzymes, especially proteinases and phospholipases play a vital role in fungal pathogenesis and tissue invasion (Silva et al., 2011). Proteinase and phospholipase activity was studied in the absence and presence of varying concentrations of β-citronellol. The Pz values were calculated and plotted for each concentration (Fig. 2). A concentration dependent inhibitory activity was observed for both the enzymes released by the fungus. Proteinase secretion decreased by 13.9 %, 21.3 %, 27.3 % and 40.6 % at MIC/8, MIC/4, MIC/2, and MIC values of β-citronellol, respectively (Fig. 2a). Similarly, secretion of phospholipases decreased by 9.4 %, 17.24 %, 24.75 % and 54.08 % in the presence of β-citronellol at the same concentrations respectively (Fig. 2b).

Effect β-citronellol on ergosterol biosynthesis
Ergosterol is unique to fungi and hence has been crucial as an antifungal target. The mechanism of action of some antifungal drugs involve either their binding to this sterol as in polyenes or inhibition of its biosynthesis as in azoles (Ghannoum & Rice, 1999). Although ergosterol is not found in the human host, these drugs besides being toxic with side effects also induce resistance in the long run. Hence total ergosterol levels were estimated in the presence of varying concentration of β-citronellol, a non toxic natural compound that has promising antifungal properties. Fig. 3 shows the sterol profiles of C. albicans in the presence of MIC and sub-MIC concentrations of β-citronellol. 10 µg/ml FLC was also taken as positive control. A significant percentage decrease was observed with increasing concentrations of the test compound. The decrease in total ergosterol content was 19%, 40%, 91% and 100%, respectively at MIC/8, MIC/4, MIC/2 and MIC concentration values. Interestingly, although FLC is a conventional antifungal drug whose mechanism of action involves inhibition of ergosterol biosynthesis, showed an inhibition of only 64% at its MIC value against C. albicans ATCC 10261 (Fig. 3).  3.4. Effect of β-citronellol on biofilm formation C. albicans associated biofilms are well protected in an extracellular matrix. They are hence difficult to treat with acquired resistance towards available antifungal drugs (Mukherjee et al., 2005;Hirota et al., 2017). Majority of the biofilms are commonly formed by C. albicans rather than the non-albicans Candida species (Kuhn et al., 2002). XTT reduction by the metabolic activity of cells was used to study the inhibitory activity of β-citronellol on Candida biofilms. Biofilm formation was inhibited by 71.13 % when treated with MIC concentration of β-citronellol. At sub-MIC concentrations of MIC/2 and MIC/4, the inhibition in biofilm formation was reduced to 42.94 % and 20.5 %, respectively (Fig. 4). Inhibitory effect of test compound on biofilm formation was clearly dose dependent. The effect of DMSO (solvent control) against biofilms was only 5 % inhibition.

Effect of β-citronellol on cell viability
The viability of Candida cells was studied in the presence of β-citronellol by a colorimetric method (WST1 cytotoxic assay) based on the metabolizing activity of mitochondria of living fungal cells (Kuhn et al., 2003). XTT is converted to formazan, a water soluble dark yellow coloured product that was measured in supernatants at 420-480nm (optimal at 440nm). This value directly correlates with the cell number. Cells cultured in micro plates were incubated with WST-1 and the assay was monitored spectrophotometrically. β-citronellol subjected ≥ 90% cytotoxicity at its MIC value (Fig. 5). It recorded a cytotoxicity of 70-82 % at MIC/2 and only 29-51 % at MIC/4. β-citronellol hence possesses high cytotoxic potential towards C. albicans and affects the metabolic activity of yeast cells leading to decreased growth and viability. The fact that β-citronellol causes only 1-2 % haemolysis in contrast to FLC and AmB, which causes ~ 6.48 % and 10.35 % respectively at 5MIC (unpublished data), shows that this compound has negligible toxicity at its MIC and MFC values.

III. CONCLUSION:
Infections caused by the pathogenic fungus C. albicans can be controlled by using β-citronellol, a plant essential oil constituents having significant antifungal potential. It has inhibitory effects on the growth pattern, adhesion and biofilm formation properties of this commensal. It also lowers the total ergosterol content of the cell and decreases its viability. Additional studies should to be performed to understand mechanism of antifungal action of βcitronellol both in vitro and in vivo.
ACKNOWLEDGEMENT Yamini Sharma acknowledges the Indian Council for Medical Research (ICMR), Government of India for awarding senior research fellowship.

CONFLICT OF INTEREST STATEMENT
There is no conflict of interest from the authors regarding publication of this article.