Antifungal resistance and herbal sensitivity of oral Candida isolates from HIV-infected patients in a rural community in Western Uttar Pradesh , India

Drug-resistant Candida species in HIV-infected patients are the result of the selective pressure of currently used azoles. In the course of screening for active plant products against drug-resistant Candida species, we paid special attention on the effects of commonly available herbal plants. A total of 123 oral Candida isolates which were previously isolated from 172 HIV-infected patients were included in this study. In vitro antifungal susceptibility to fluconazole (FCZ), itraconazole (ITZ) and amphotericin B (AMB) was evaluated using Clinical and Laboratory Standard Institute (CLSI) guidelines. Antifungal potency of garlic, neem, Aloe, Calendula, Citrus, mint, tea and ginger extracts were tested against drugresistant isolates. Out of 123 isolates, 26.8% were resistant to FCZ, 21.9% to ITZ and 8.1% to AMB. All drug-resistant isolates tested, were completely inhibited by garlic and neem leaves extracts with minimum fungicidal concentration (MFC) of 0.781 and 1.562 mg/ml, respectively. Aloe and Calendula extracts were also found to be effective with MFC of 3.125 mg/ml each. The observed growth inhibition zones and minimum inhibitory concentrations (MICs) showed that the isolates exhibited susceptibility. Our results provided scientific justification for the use of garlic, neem, Aloe and Calendula extracts in health products and herbal remedies against multidrug-resistant candidiasis.


INTRODUCTION
Oral candidiasis (OC) is the most frequent human immunodeficiency virus (HIV) infection-associated oral disease and can also act as a marker for immunosuppression (Moura et al., 2006;Chunchanur et al., 2009;Merçon et al., 2009;Moura et al., 2010;Thompson et al., 2010).Fluconazole (FCZ) is frequently used for the treatment of mucocutaneous candidiasis in patients with AIDS (Vazquez et al., 2001).In the 1990s, there was a significant increase in the prevalence of drug-resistant fungal infections due to Candida species in patients hospitalized for mucosal or systemic diseases.The widespread application of FCZ or related azole antifungal is postulated to promote selection of resistant subpopulations by shifting colonization to more naturally resistant species, such as Candida krusei or Candida glabrata.Alternatively, azole-resistant subspecies have arisen in vivo and in vitro that shows changes in the target enzyme lanosterol 14-α-demethylase, in expression of multidrug efflux pumps, or in both (Mathema et al., 2001).In India, despite the numerous reports of isolation of azole resistant strains of Candida species from patients with refractory mucosal candidiasis, only a few longitudinal prospective studies have evaluated the antifungal susceptibility of Candida isolates recovered from HIV-infected patients receiving long-term therapy with azole (Gautam and Garg, 2013a).Drug-resistant C. albicans and non-albicans Candida species (NACs) in immunocompromised patients are the result of the selective pressure of currently used azoles and have been well recognized as a global problem in recent years (Gautam and Garg, 2013a).As plants can produce antimicrobial compounds to protect themselves from biotic attack, the search for antimicrobial agents with different chemical structure and biological mode of action from plant sources is an alternative option in the fight against these microbes (Lai and Roy, 2004).In the course of screening for active plant products against drug-resistant Candida species, we paid special attention on the effects of commonly available herbal medicinal plants.Here, we reported the drug-resistant oral Candida isolates from HIV-infected patients in a rural community in western Uttar Pradesh, India together with antifungal activity of eight herbal plants extracts against these isolates.

Isolates
A total of 123 Candida isolates, which were previously isolated and identified from HIV-infected patients (Gautam and Garg, 2013b) in the Department of Microbiology, Chaudhary Charan Singh University, Meerut, India were included in this study.Out of 123 isolates, 91 were recovered from patients with oral lesions (Group 1) and 32 were from patients without oral lesions (Group 2).Informed consent obtained from participants and procedures were performed according to institutional board of ethical committee.

Antifungal susceptibility test
The three antifungal agents used for susceptibility testing of Candida species isolates were fluconazole (FCZ), itraconazole (ITZ), and amphotericin B (AMB) (Sigma-Aldric, USA).Fluconazole was dissolved in sterile distilled water while ITZ and AMB were dissolved in dimethyl sulfoxide (DMSO) (Hi-Media, India) to make stock solutions.Antifungal susceptibility test using broth microdilution (BMD) method was performed as per Clinical and Laboratory Standard Institute (CLSI) (CLSI M27-A3, 2008).A completely synthetic medium, RPMI 1640 (Hi-Media, Mumbai) supplemented with glutamine but without bicarbonate, was used.Medium was supplemented to final concentration of 20 g/L (2.0%) glucose for better growth of yeast isolates.The range of concentrations tested were 0.125 to 64.0 µg/ml for FCZ and 0.0313 to 16.0 µg/ml for ITZ and AMB.The BMD test was performed using sterile, disposable multiwell microdilution plates (96 U-shaped wells) (Tarson, Mumbai).Aliquots of 100 µl of each antifungal agent at a concentration two times the targeted final concentration were dispensed in the wells of the plates.The suspension of yeasts after 48 h of incubation onto Sabouraud's dextrose agar was prepared in Gautam and Garg 1677 sterile saline, adjusted spectrophotometrically at 530 nm to match the turbidity of a 0.5 McFarland standard and was diluted in RPMI 1640 in order to obtain a final concentration of 0.5 × 10 3 to 2.5 × 10 3 CFU/ml (CLSI M27-A3, 2008).A constant volume (100 µl) of the inoculum was aseptically added to each microdilution well containing 100 µl of serial dilution of antifungal agents so as to make final desired concentration.For the members of the azole drug the MIC was defined as the lowest drug concentration that resulted in 80% growth inhibition.For AMB, this value was defined as the value in which 100% growth inhibition was observed (Brito et al., 2010).

Collection and processing of herbals
In vitro antifungal activities of eight herbal plants extracts were evaluated.These were garlic (Allium sativum), neem (Azadirachta indica), aloe (Aloe vera), Calendula (Calendula arvensis), Citrus (Citrus sinensis), mint (Mentha spicata), tea (Camellia sinensis) and ginger (Zingiber officinale).All were collected from crop fields and local market of Western Uttar Pradesh, India.The plant material was dried under shade and reduced to small pieces; plant material was grounded in electronic grinder to obtain fine powder.Dried herbal powder (100 g) was macerated with 1000 ml ethanol (95%) in 2000 ml Erlenmeyer flask.Flask was closed with aluminium foil and placed on orbital incubator shaker at 130 rpm for 5 days with intermittent agitation and soaking.After 5 days the mixture was filtered using Whatman's filter paper No. 1. Herbal extract was evaporated to semi solid state under vacuum at 40°C using a Buchi Rotavapour.Semi solid extract was dried in the crucible under a controlled temperature of 40°C to obtain solid extract.Prepared extract was stored in dark-coloured container at 4°C until used (Gautam and Garg, 2013a).All herbals were processed in same manner.

Agar well diffusion method
All extracts were first subjected to an antifungal susceptibility test for their growth inhibition zone diameter (in mm) following the previous agar well diffusion (AWD) method with a slight modification (Gautam and Garg, 2013a).In this method, Mueller-Hinton Agar (MHA) plates were prepared.Each plate was inoculated with test organism of 2.5 x 10 3 CFU/ml inoculum density using spread plate technique.Three wells at equidistance were cut aseptically with the help of sterile cork borer.Stock solution having 1250 mg/100 ml air dried powder of herbal extracts dissolved in DMSO was prepared.100 µl of each herbal extract was dispensed into 6 mm precut wells of MHA plates.These plates were incubated at 37°C for 48 h and inhibition zone diameter was measured using Hi Antibiotic Zone Scale-C PW297 (Hi-Media, Mumbai).Data are presented in the form of inhibition zone diameter (mm) which is an average of two independent replicates.

Minimum inhibitory concentration assay
To measure MICs of active herbal extracts, macro-broth dilution (MBD) method was used with some modifications (Gautam and Garg, 2013a).In this method, 12 sterilized glass tubes with caps were taken.Tube 1 was filled with 2 ml of double-strength culture medium (Sabouraud's Dextrose broth) containing twice the final active herbal extract concentration (6.25 mg/ml).Tubes 2 -11 were filled with 1 ml of double strength Sabouraud's Dextrose broth.Tube 12 was considered as blank filled with distilled water. 1 ml amount was taken from tube 1 and diluted two-fold by transferring it to tube 2 with a micropipette.1 ml sample was then removed from tube 2 and transferred to tube 3 and so on to tube 10.The last 1 ml of diluted herbal extract was then discarded.Thus, each tube 1-10 contained 1 ml of double-strength Sabouraud's Dextrose broth containing twice the final herbal extract concentrations.Each tube was inoculated with 1 ml of 5 × 10 3 CFU/ml yeast suspension which will give the required herbal extract concentration and inoculums density of 2.5 × 10 3 CFU/ml.The growth control tube (tube 11), which contained 1 ml of sterile drug-free medium, was also inoculated with 1 ml of the same inoculum suspension.Tube 12 was filled with 2 ml of sterile distilled water, from the lot used to prepare the inoculum as a sterility control for media.Final volume of each tube was 2 ml.All tubes were incubated for 24 to 48 h at 37°C after which the turbidity was observed and measured with the spectrophotometer.On the basis of optical density, the tube showing critical inhibitory points were selected and 100 µl samples from each MIC assay tube with growth inhibition was spread onto Sabouraud's Dextrose Agar (SDA) plate.Plates were then incubated at 37°C for 48 h to count number of colony forming units (CFU).CFU/ml was determined for each selected well along with control well.Data are presented in the form of MIC 50 , MIC 80 and minimum fungicidal concentration (MFC).The MIC 50 was considered the lowest tested concentration with a 50% reduction in growth as compared to growth of positive control.The MIC 80 was considered the lowest tested concentration with 80% reduction in growth as compared to growth of positive control.The highest dilution of the plant extract that retained its inhibitory effect resulting in no growth (absence of turbidity) of a microorganism is recorded as the MFC value of the extract.

Phytochemical constituents of plant extracts
Herbal extracts were subjected to qualitative chemical tests for identification of various categories of organic chemical constituents such as alkaloids, tannins, steroids, flavonoids, saponins, glycosides, terpenoids, carbohydrates, antiquinones and phenols.The phytochemical analysis was done according to standard methods (Roopashree et al., 2008;Obasi et al., 2010;Raphael, 2012).Briefly the methods are given below.

Test for alkaloids
Extracts were dissolved individually in dilute hydrochloric acid and filtered.

Test for saponins
a) Froth test: Extracts were diluted with distilled water to 20 ml and this was shaken in a graduated cylinder for 15 min.Formation of 1 cm layer of foam indicates the presence of saponins.b) Foam Test: 0.5 g of extract was shaken with 2 ml of water.If foam produced persists for ten minutes it indicates the presence of saponins.

Test for phenols
Ferric chloride test: Extracts were treated with 3-4 drops of ferric chloride solution.Formation of bluish black colour indicates the presence of phenols.

Test for tannins
Gelatin test: To the extract, 1% gelatin solution containing sodium chloride was added.Formation of white precipitate indicates the presence of tannins.

Test for Steroids
2 ml of acetic anhydride was added to 0.5 g ethanolic extract of each sample with 2 ml H 2 SO 4 .The colour changed from violet to blue or green in some samples indicating the presence of steroids.

Test for flavonoids
a) Alkaline reagent test: Extracts were treated with few drops of sodium hydroxide solution.Formation of intense yellow colour, which

Statistical analysis
All the data were analysed in the worksheet of Statistical Package for Social Science (SPSS) software (Version 16.0) and analyzed accordingly.

Herbal activity against drug-resistant isolates
In vitro antifungal activity results of herbal plants extracts are summarized in Table 3.Using AWD method the highest growth inhibition zone diameter was recorded as 26±3.7 mm (mean) with garlic extract against all drugresistant isolates, followed by neem extract which demonstrated 24.0±2.9mm.Aloe and Calendula extracts demonstrated 18.2±3.5 and 17.5±3.2mm, respectively.Citrus peel, Mint leaves and ginger extracts demonstrated inhibition zone diameters of 15.8±4.1,13.3±2.9and 13.6±1.6mm, respectively, while tea leaves extract demonstrated only 11.1±2.2mm.Using MBD method, the lowest MIC 50 value was recorded as 0.097 mg/ml with garlic and neem extracts each against all isolates tested while Aloe and Calendula extracts demonstrated MIC 50 value of 0.390 mg/ml each.The isolates demonstrated low garlic extract, MIC 50 in which 66.7% ( 22) presented values of 0.048 mg/ml.Citrus peel extract demonstrated MIC 50 value of 0.390 mg/ml against all drug-resistant isolates while mint leaves and ginger extracts were comparatively less effective with MIC 50 value of 3.125 mg/ml each.Tea leaves extract was effective only to inhibit 37.23% growth of drug-resistant isolates.Garlic and neem leaves extracts completely inhibit the growth of all drug-resistant isolates with MFC value of 0.390 and 0.781 mg/ml respectively, while 72.7%(24) isolates demonstrated low neem extract's MFC value of 0.390 mg/ml.Aloe leaves and Calendula extracts demonstrated MFC value of 1.562 mg/ml each.Citrus peel, mint leaves and ginger extracts were unable to inhibit complete growth of any of the drug-resistant  isolates resulting to, no MFC value.Figure 1 summarizes the comparative antifungal activity analysis of all herbal extracts tested against 33 drug-resistant oral Candida isolates.

Qualitative assay of herbal extracts for phytochemical constituents
Phytochemical assay results indicated the presence of carbohydrates (reducing sugars, hexose sugars, non-reducing polysaccharides gums and mucilages), alkaloids and flavonoids as the main constituents in herbals tested (Table 4).

DISCUSSION
A variety of antifungal agents are now available for the treatment of Candida infections.However, worldwide reports indicated that pathogenic isolates of C. albicans have relatively high potentials for developing resistance  ( Espinel-Ingroff et al., 1998).Relatively, high resistance of C. albicans and non-albicans Candida to common antifungal agents tested was observed in the present study.Although triazole agents appear to be highly effective initially, the increase of resistance to them has been reported (Ellepola and Samaranayake, 2000;Pfaller and Diekema, 2007).

Antifungal resistance
Two different populations of individuals were studied, HIV-infected patients with oral lesions (symptomatic) and without oral lesions (asymptomatic).Isolates from symptomatic patients showed higher MIC range in comparison with isolates from asymptomatic patients.All these isolates were recovered from patients who had previously received same antifungal agent.This feature could be due to a secondary resistance produced by exposure to the drug.In past few decades, there have been numerous reports of Candida infections in India (Basu et al., 2003;Gugnani et al., 2003;Bharathi and Rani, 2011;Jain et al., 2011;Gautam and Garg, 2013b, c).Several studies have also reported FCZ resistance in C. albicans strains isolated from HIV-infected patients with oral candidiasis (Mane et al., 2010).
In the present study, 21 of 33 FCZ resistant, 17 of 27 ITZ resistant and 5 of 10 AMB resistant isolates were C. albicans while remainders were NACs isolates.NACs isolates showed higher resistance as compared to C. albicans.
Among 33 drug-resistant isolates, 80.8% ( 27) isolates (seventeen C. albicans, two C. tropicalis, five C. glabrata and three C. krusei) presented resistance to both azole tested.These findings suggest the possibility of a crossresistance phenomenon between these antifungal agents.Active flux and alteration in drug target are considered to be the most important mechanisms of azole resistance.
Although the breakpoint concentrations for AMB are not clearly defined, a strain is considered to be resistant when it represents an MIC ≥ 2 µg/ml (Luque et al., 2008).According to this criterion, only 10 isolates were found resistant to AMB and this feature is difficult to analyse from a clinical point of view.These isolates were recovered from patient, who had been previously treated with FCZ.These results agree with previous reports on the existence of strains resistant to AMB which had previously been in contact with azoles.The azoles affect the cellular membrane and this could be a factor in a selection of mutants (Carrillo-Muñoz et al., 1997).Most of the resistant isolates to AMB were recovered from immunocompromised individuals and this could be related to effects on cellular membrane of yeast, in relation to the multiple treatments received for these patients (Luque et al., 2008).

Herbal sensitivity
In this study, eight herbal extracts preparations were tested against drug-resistant oral Candida isolates.In the present study, garlic and neem leaves extracts exhibited highest anticandidal activity against all the drug-resistant isolates tested with MFC value of 0.390 and 0.781 mg/ml, respectively.Three C. glabrata isolates which were resistant to all the three drugs tested were completely inhibited with garlic and neem extracts with MFC value of 0.390 and 0.781 mg/ml, respectively.It has been proven over and over that garlic is an effective anti-fungal agent.Its compounds are very active against C. albicans and NACs isolates (Jafari et al., 2007;Bokaeian et al., 2010).In this study, A. vera and Calendula extracts were also found effective.In the present study, we confirm the inhibitory effect of Garlic, Neem, Aloe and Calendula extracts against drug-resistant Candida spp.isolateswhich has been described before by Agarry et al. (2005), George et al. (2009), Mahmoud et al. (2011), Doddanna et al. (2013) and Padhye et al. (2013).The observed growth inhibition zones and MICs showed that the isolates exhibited susceptibility.This indicates that garlic, neem, Aloe and Calendula extracts have broad spectrum of antifungal activity.The MFCs of garlic and neem extract reported in this study are in consonance with previous anticandidal studies on pathogenic and emerging drug-resistant Candida species such as C. albicans, C. glabrata and C. krusei (Gautam and Garg, 2013a).Although, Citrus peel extract was unable to inhibit the complete growth of any of the isolate tested but its MIC 50 and MIC 80 value of 0.390 and 3.125 mg/ml, respectively and intermediate growth inhibition zone diameter showed its anticandidal efficacy against drugresistant isolates.Mint leaves extract were found comparatively less effective (Doddanna et al., 2013).Tea leaves extract was unable to demonstrate signification reduction in growth of Candida isolates as previously reported by Doddanna et al. (2013).
The herbal extracts offer several advantages such as unlimited availability and possibility of minimal problem of drug resistance.The present study showed the presence of tannins, alkaloids and flavanoids as main constituents of the ethanolic herbal extracts.A correlative relationship has been reported between the phytochemicals such as tannins and flavonoids and the free radical scavenging activity and antimicrobial activity (Kaur et al., 2010).Tannins and flavanoids have therapeutic uses due to their anti-inflammatory, anti-fungal, antioxidant and healing properties (Thiago et al., 2008).An added advantage of using herbals to treat yeast infections is that no clinical strains of C. albicans have been known to be resistant to herbal therapy.

Conclusion
The high percentage of antifungal resistant against Candida isolates in the present study could be due to the fact that many of these patients had received previous treatments with these drugs.Therefore, azole prophylaxis should be considered only in special cases.Our study, suggest to avoid indiscriminate use of antifungal agents, which may ultimately decrease the incidence of candidiasis caused by resistant and non-albicans Candida species.The results of this study provided scientific justification for the use of neem, garlic, Aloe and Calendula extracts in health products and herbal remedies against multidrug-resistant candidiasis.Therefore, complementary and alternative medicine practices with herbal extract including neem, garlic, Aloe and Calendula as a means of decreasing the burden of drug resistance and reducing the cost of management of diseases would be of clinical and public health importance in any country.
becomes colorless on addition of dilute acid, indicates the presence of flavonoids.b) Lead acetate test: Extracts were treated with few drops of lead acetate solution.Formation of yellow colour precipitate indicates the presence of flavonoids.Test for terpenoids a) Copper acetate test: Extracts were dissolved in water and treated with 3-4 drops of copper acetate solution.Formation of emerald green colour indicates the presence of diterpenes.a) Salkowski's test: Extracts were treated with chloroform and filtered.The filtrates treated with few drops of concentrated sulphuric acid, shaken and allowed to stand.Appearance of golden yellow colour indicates the presence of triterpenes.

Figure 1 .
Figure 1.Comparative analysis of herbal extracts activity against drug-resistant oral Candida isolates.
Extracts were treated with ferric chloride solution and immersed in boiling water for about 5 min.The mixture was cooled and extracted with equal volumes of benzene.The benzene layer was separated and treated with ammonia solution.Formation of rose-pink colour in the ammonical layer indicates the presence of anthranol glycosides.
a) Mayer's test: Filtrates were treated with Mayer's reagent (potassium mercuric iodide).Formation of a yellow coloured precipitate indicates the presence of alkaloids.b) Wagner's test: Filtrates were treated with Wagner's reagent (iodine in potassium iodide).Formation of brown/reddish precipitate indicates the presence of alkaloids.a) Modified Borntrager's test: b) Legal's test: Extracts were treated with sodium nitropruside in pyridine and sodium hydroxide.Formation of pink to blood red colour indicates the presence of cardiac glycosides.

Table 1 .
Yeast distribution in different groups of the study population.
Data are shown in no.(%) of Candida spp.isolates; *

Table 2 .
Antifungal susceptibility of Candida spp.isolates using CLSI testing.

Table 3 .
Growth inhibition zone diameters and MICs of the herbal extracts against 33 drug-resistant oral Candida isolates.

Table 4 .
Phytochemical constituents of plant extracts.