The present study was conducted under a protocol approved by the Research Ethics Committee, University of Sharjah (REC-18-02-18-03).
Yeast isolates: Overall, a total of 85 Candida isolates, obtained from 48-yeast positive samples of carious dentin of children with S-ECC were selected for the current study and comprised ten isolates each of Candida albicans, C. krusei, and C. tropicalis, and five C. glabrata isolates (Fakhruddin et al., manuscript in preparation).
The identity of the isolates was reconfirmed by sub-culture on Sabouraud Dextrose Agar (SDA), and the characteristic growth on CHROMagar (HiCrome™ Candida Differential Agar, M1297A) and finally by multiplex PCR (see below).
Multiplex PCR:
The identity of the isolates were further confirmed by Multiplex PCR amplification method, which permitted the identification of six common pathogenic yeast species, namely C. albicans, C. glabrata, C. parapsilosis, C. tropicalis, C. krusei, and C. dubliniensis. The employed method was based on the amplification of two fragments from the ITS1 and ITS2 regions by the combination of two-yeast-specific and six-species-specific primers in a single PCR reaction [31], Table 1. All PCR-reaction products were evaluated by electrophoresis in 2.0% (w/v) agarose gels run at 90V for 60 mins
Antifungal susceptibility assay by disc diffusion:
The antifungal susceptibility of the yeast was evaluated against amphotericin B, fluconazole, and SDF. To prepare the SDF impregnated discs, stocks of sterile filter paper discs stored at 200C were allowed to reach room temperature, and then infused with SDF volumes of 2.5 µl, 5 µl, 10 µl, and 15 µl and dried in an oven for an hour at 600C before use.
To minimize batch to batch variation, plates were prepared on a single session at ambient temperature, with a constant agar volume of Sabouraud Dextrose Agar (SDA; MH063, Himedia).
All isolates of Candida albicans, C. krusei, C. tropicalis and C. glabrata strains were selected randomly and tested by agar-based diffusion test following CLSI M44-A2 standardized method with some modifications [32]. In brief, a yeast cell suspension of 106 cells/mL from a 24-hours old culture was grown in Sabouraud Dextrose broth (ME033, Himedia) and adjusted to 0.5 McFarland standard using a densitometer (Grant Instruments™ Grant Bio™ Densitometer) and spread uniformly using a glass spreader.
The inoculated SDA plate was allowed to dry for 20 minutes, followed by application, aseptically, of Amphotericin B-20mcg disks (SD233, AP 20 mcg, Himedia), Fluconazole-25 mcg disks (SD232, FLC 25 mcg, Himedia), sterile paper disks with SDF (Thermo Scientific™ Oxoid™ Blank Antimicrobial Susceptibility Discs) and a sterile paper disc (the negative control) using a pair of forceps. The plates were incubated at 370C within 15-20 mins, after the application of up to a period of 48-hours, before evaluation of zones of growth inhibition. Recommended CLSI quality assurance isolates Candida parapsilosis ATCC 22019 [25] was tested as a positive control, with each set of experiments.
After 48-hours, inhibition zone-diameters were measured to the nearest millimeter at the point where there was a noticeable growth reduction. All experiments were tested in triplicate, on three separate occasions.
Broth microdilution antifungal susceptibility assay:
Minimal inhibitory concentrations (MIC50 and MIC90) for silver diamine fluoride (SDF) was ascertained following standard methodology set out in CLSI M27-A3 broth microdilution procedure, with some modifications [33, 34] . The modifications include using flat-bottom 96-well microtiter plates (Corning, 3370 Polypropylene Flat Bottom 96 Well) and growth reading determined spectrophotometrically by a microplate reader. As brown discoloration of oxidized silver-ions in SDF makes it challenging to evaluate growth in the wells of the microtiter plates visually.
Amphotericin B (AMB, A2942, Sigma-Aldrich) and Fluconazole (FCZ, ≥98% (HPLC), powder F8929 Sigma-Aldrich) was prepared by dissolving AMB in 5% DMSO and FCZ in sterile distilled water. The solutions were added to RPMI-1640 w/ L-glutamine, 0.2% glucose and 0.165 moles/l MOPS buffer w/o sodium bicarbonate (AT180, RPMI-1640, Himedia) during the time of antifungal exposure to Candida spp. Minimum inhibitory effect of AMB and FCZ on clinical isolates were verified over the concentration range of 0.125-64 µg and 0.0312-16 µg, respectively.
Silver Diamine Fluoride complex (Topamine, 25%w/v of silver ions, Product code: DL160.9-1) was obtained from Dentalife Australia Pty. Ltd. Antifungal effects of Ag+ particles present in SDF volume-range of 0.156 µl-15 µl were tested over the concentration range between 0.039 mg to 3.75 mg of silver (0.039 mg; 0.078 mg; 0.156 mg; 0.313 mg; 0.625 mg; 1.25 mg; 2.5 mg; 3.75 mg).
For minimum inhibitory concentration (MIC) determinations, all yeast cell suspensions tested were adjusted to a turbidity of 0.5McFarland standard. Subsequently, the cell suspensions were further diluted to a final concentration of 103 cells/ml in the RPMI 1640 medium with 2X serial dilutions of Ag+ in SDF, AMB, and FCZ and were pipetted into well of 96-well plates in a standardized manner, and the plates incubated at 370C for 24-hours. Afterwards, the MICs were determined spectrophotometrically at 490nm with a microtiter plate reader (BIO-TEK, ELX800, USA). MIC50 and MIC90 were defined as the lowest drug concentrations that inhibited growth by 50% and 90% compared with drug-free wells, as determined by the absence of turbidity.
In each susceptibility test, QC strain C. parapsilosis ATCC 22019 was included. MIC range of 0.5µg-4µg of FCZ to C. parapsilosis ATCC 22019 was used as a reference.
All tests were replicated on three separate occasions with observations determined independently by two observers.
Effect of SDF on germ-tube formation:
The effect of SDF on the germ-tube formation in C. albicans was ascertained following a protocol described by Nair et al., [35] with some modifications. A fresh C. albicans inoculum was prepared for the experiment by harvesting a 24-hour growth in Sabouraud dextrose agar. A 107 yeast cells/ml suspension was added to fetal bovine serum (F2442, FBS, Sigma-Aldrich) in a microtube. Effects of SDF against germ-tube formation was tested in the volume-range of 0.156 µl-10 µl containing concentration range of silver ions between 0.039 mg to 3.75 mg (0.039 mg; 0.078 mg; 0.156 mg; 0.313 mg; 0.625 mg; 1.25 mg; 2.5 mg).
The mixture was then vortexed and incubated aerobically at 370C for 90 mins. At 90 minutes, the formalin solution (HT501640-neutral buffered 10%, Sigma-Aldrich) was added to the mixture to arrest further growth. A drop was removed from each mixture and placed on a glass-slide covered with a coverslip, and germ-tube positive cells were quantified under the microscope (34MP-2K HD-USB Microscope, under x 40 magnification)
A total of 100 microscopic fields with either germ tube positive or negative blastopores were counted, and the percentage of germ tube positive fields quantified as per the protocol of Nair et al [35]. The germ-tube experiment was repeated twice on different occasions.
Time-kill assay:
Time-kill curves were developed using a protocol described by Klepser et al., with some modifications [36]. The test concentrations of Ag+ (0.625 mg; 1.25 mg; 2.5 mg) in SDF volumes of 2.5 µl, 5 µl and 10 µl were assessed at pre-determined time points (0, 30mins, 1hr, 2hrs, 12hrs and 24hrs). A fresh yeast inoculum was prepared for each experiment by harvesting a 24-hour growth in Sabouraud dextrose broth, and yeast suspension adjusted, spectrophotometrically, to 0.5McFarland standard (1x106 to 5x106 CFU/ml).
Yeast suspension (1ml) was diluted in 9 ml of RPMI 1640 medium with and without SDF, which provides the starting inoculum of 1× 105 to 5×105 CFU/ml. After incubation at 370C with agitation, a 100µl aliquot was removed from each solution at predetermined time points and serially diluted (10-folds) in sterile water. From each dilution, a 30µl aliquot was plated on SDA. Colony counts were subsequently obtained after incubation of the plates at 37°C, at pre-determined time points. The fungicidal effect was defined as ≥99.9%, or 3-log10-unit, reduction in CFU/ml from the starting inoculum [36, 37]. Time-kill experiments were conducted in duplicate on different occasions.
Scanning electron microscopy:
The effects of the SDF on the ultrastructural features of a single strain of C. albicans was investigated using SEM. A randomly selected C. albicans strain was incubated in RPMI 1640 at 370C for 24- hours in a 12-well plate. A yeast suspension of 2.5x105 cells in RPMI was prepared into which 5µl SDF (1.25mg of silver-ions) was added. The morphological changes of the treated yeast cells were observed after 10- minutes, and 1-hour post-SDF treatment. The antifungal concentration of silver-ions was selected based on the results obtained in the susceptibility testing assays.
Standard methodology with minor modifications was used for SEM analysis of yeast as previously described [38]. Samples of the negative control (untreated cells) and treated cells for 10mins and an hour, respectively, were fixed in 2.5% glutaraldehyde and 4% paraformaldehyde in the presence of cacodylate buffer (pH 6.2) on ice for an hour. Post-fixation of the samples was carried out for 30-minutes with 1% osmium tetroxide. Samples were gently dehydrated in graded ethanol (30%, 50%, 70%, and 90%). The treated and control samples were then mounted on aluminum stubs and air dried at room temperature, and sputter coated with gold (Polaron SC7640 sputter coater; Thermo VG Scientific, United Kingdom) and observed with an environmental scanning electron microscope (FEI Co., Hillsboro, OR).
Finally, Energy Dispersive X-Ray Spectroscopy (EDS) of the Ag+ ion deposition was also carried out for the samples exposed to SDF.
Statistical Analysis:
The data obtained from SDF antifungal assays were presented as mean ± standard deviation (SD). MIC50, MIC90, and means of inhibition-zone diameters were calculated for each Candida strains. The group difference was analyzed using one-way analysis of variance (ANOVA). For all statistical analysis, a P value of ≤0.05 was considered statistically significant.