Biosafety and ethical approval

The biosafety of the protocols of this study was approved by the Safety and Risk Management Taskforce Mahidol University (SI2023-006) and the Institutional Biosafety Committee Thammasat University (034/2564). Ethical approval for the isolation and use of human primary cells in this study was approved by the Ethics Review Sub-Committee for Research Involving Human Research Subjects of Thammasat University No. 3 (049/2564).

Antifungal susceptibility assay

The minimum inhibitory concentration (MIC) of Mz against C. albicans ATCC 24433 was determined according to the Clinical and Laboratory Standards Institute (CLSI) M27 standard [18]. This was performed as a baseline control for the antifungal efficacy of Mz. The medium used in this study was RPMI-1640 broth with glutamine, without bicarbonate and buffered to pH 7.0 with MOPS. A C. albicans stock suspension was prepared by selecting five colonies of C. albicans grown on yeast extract–peptone–dextrose (YPD) agar, subculturing into 5 mL of 0.85% saline, and adjusting to 0.5 McFarland standard at 530 nm wavelength.

For the 24-h microbroth dilution assay, C. albicans working suspension was made by preparing a 1:1000 dilution of the stock suspension in RPMI-1640 broth. Mz working solution was prepared with concentrations ranging from 0.015625–2 µg/ml in RPMI-1640 broth, and 0.05% Dex was added for the Mz + Dex group. The broth microdilution was performed by adding 50 µl of each miconazole working solution and 50 µl C. albicans working suspension. The plates were incubated at 37 °C for 24 h and checked for growth to determine the MIC.

For the 3-time exposure assay, a 1:50 dilution of the C. albicans was prepared in RPMI-1640 broth. Miconazole working solution was prepared with concentrations ranging from 0.25–64 µg/ml in RPMI-1640 broth. One milliliter of the diluted cell suspension was aliquoted into each Eppendorf tube and centrifuged to pellet the cells. The supernatant was removed, and the resuspended cells were incubated with different concentrations of Mz or Mz + Dex working solution. All tubes were incubated at 30 °C with shaking at 200 rpm for 5 min, then washed and resuspended the cells in fresh RPMI-1640 broth. Cells were then incubated at 30 °C without shaking for 4 h, repeated the same process twice, and then incubated all tubes at 30 °C for 18 h. Cells were diluted in a normal saline solution and spread on YPD agar. The number of colonies on each plate was counted after incubation at 37 °C for 2 days. The percentage of inhibition was calculated by dividing the number of colonies from each concentration by the number of colonies in the control plate.

Culture of human NOKs, GFs, and T lymphocytes

NOKs are spontaneously immortalized normal oral keratinocytes isolated from gingival tissues of healthy volunteers [19]. NOKs were maintained in a defined keratinocyte serum-free medium (KSFM: Gibco Life Technologies Ltd, Paisley, UK) containing 1% penicillin and streptomycin (both from Gibco) at 37 °C in a humidified atmosphere of 5% CO2 in the air. The medium was changed every 2–3 days. These cells have previously been reported to have characteristics resembling normal oral keratinocytes [19].

GFs were isolated from biopsies of healthy patients’ gingiva. They were cultured in Dulbecco’s modified Eagle medium (DMEM) containing 10% fetal calf serum (FCS) supplemented with 1% penicillin and streptomycin and 2 mM L-glutamine (all from Gibco) at 37 °C in a humidified atmosphere of 5% CO2 in the air. GFs between passages 3–5 were used.

T lymphocytes were isolated from buffy coats of fully anonymized healthy donors who provided blood donations to the Thammasat University Hospital Blood Bank. The diluted buffy coat was layered on Ficoll-Paque PLUS (GE Healthcare Life Sciences) and centrifuged at 800 g for 40 min at 18 °C. The peripheral blood mononuclear cell (PBMC) interface layer was collected, washed three times with phosphate-buffered saline (PBS), and incubated in the RPMI medium containing 10% FCS supplemented with 1% penicillin and streptomycin and 2 mM L-glutamine (all from Gibco) for 2 h. The non-adherent T cells were collected and activated with 5 µg/ml concanavalin A (ConA) for 72 h before use.

Keratinocyte-fibroblast Transwell co-culture model

The Transwell co-culture models of the oral mucosa were fabricated by preparation of the collagen mixture with type 1 rat tail collagen (3 mg/ml in 0.6% acetic acid; Gibco), 10 × minimum essential medium (Sigma-Aldrich), and neutralized with 1 N sodium hydroxide before an addition of GF suspension. 1.5 mg/ml collagen mixture was prepared with 2.7 × 106 GFs in 150 µl final volume. 50 µl of acellular 1.5 mg/ml collagen hydrogel was first added to the Transwell insert of 24-well insert (6.5 mm diameter with 0.4 µm pore size), and then followed by fibroblasts-containing collagen hydrogels. The hydrogels were incubated at 37 °C, 5% CO2 for 2 h to polymerize and for another 1 h after adding fibroblast media. Finally, 1 × 105 NOKs were seeded on top of the GFs-populated collagen hydrogels, and the construct was incubated and submerged in KSFM for 3 days. After 3 days, the insert was raised to an air–liquid interface for 14 days in the differentiation media (KSFM + 1.2 mM CaCl2 + 0.25 mmol/l ascorbic acid). The culture medium was changed every 2 days. The procedures are presented in Fig. 1.

Fig. 1

Schematic diagram showing the fabrication of NOKs-GFs Transwell co-culture model

Drug treatment

For the cell monolayers, cells were treated with 0.05% Dex in the presence of Mz at 0, 4, 6, 8, and 64 μg/ml [16, 20], following the clinical regimen recommended for patients with OLP, i.e., 5 min for 3 times per 24 h [21]. After 48 h, cells were assayed as indicated below. Non-treated cells and cells treated with Dex alone were used as controls.

For the Transwell co-culture model, drugs were diluted in 150 µl KSFM and added into the insert. Drugs were added to the co-culture models 3 times daily, treated every 4 h apart, for 5 min each at 37 °C, 5% CO2. The models were washed with fresh KSFM prior to each drug treatment. Fresh KSFM without drugs was used in the control group. This drug treatment was continued for 2 days before the samples were fixed in 4% paraformaldehyde (PFA) overnight at 4 °C, washed with PBS, and then stored at 4 °C until analysis.

Morphological analysis and cell viability assay of NOKs and GFs

Cells were cultured at a density of 5 × 103 cells/cm2 for 18 h. After the indicated time in culture, the cell morphology was examined under a light microscope (Nikon Eclipse TS100), and photomicrographs of cell appearances were taken using a Nikon Digital sight DS-L2. Cells were then fixed with 4% PFA for 15 min and stained with 0.05% crystal violet solution for 15 min. The cell morphology was examined under a light microscope, and photomicrographs of cell appearances were taken as above. For the cell viability assay, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay was performed using 0.2% MTT solution for 4 h at 37 °C, and the end-product color was subsequently analyzed by measuring absorbance at 490 nm (A490) which corresponds to the viability of cells. Cell viability is expressed as the mean percentage ± SD of control (100%) from three independent experiments.

Histological assessment and immunohistochemistry of Transwell co-culture model

The NOKs-GFs Transwell co-culture models were fixed with 4% PFA overnight at 4 °C. The organ cultures were submitted for routine tissue processing in the automatic tissue processor. First, the specimens were dehydrated through an alcohol series ranging from 70% to absolute and then cleared with xylene. Subsequently, the specimens were embedded in paraffin wax and the tissue blocks were sectioned using a microtome to produce 3-μm slices. Finally, the sections were stained with hematoxylin and eosin (H&E) and observed under a light microscope.

Immunohistochemistry (IHC) was performed on 3-µm thick sections with primary antibody against Ki67 (Ki67 mouse monoclonal antibody, dilution 1:300, Dako, Carpinteria, CA, USA) and E-cadherin (E-cadherin mouse monoclonal antibody, dilution 1:150, Novocastra, Leica Biosystems, Newcastle Upon Tyne, UK) for 40 min on the LEICA BOND-III IHC automated stainer (Leica Biosystems). Before mounting the slides, they were counterstained with hematoxylin for 20 min. To support the validity of the staining, positive controls for Ki67 (Appendix tissue) and E-cadherin (Breast carcinoma) were also included in each run. ImageJ software was used to count the number of Ki67-positive cells (4–5 regions per condition).

TUNEL assay

To assess the apoptotic cells of the Transwell co-culture models, 3-µm thick sections were subjected to indirect TUNEL staining using the ApopTag® Plus Peroxidase In Situ Apoptosis Detection Kit (S7101, Sigma-Aldrich). Micrographs of the NOKs-GFs Transwell co-culture model were captured using a microscope (Olympus). ImageJ software was used to count the number of TUNEL-positive cells (4–5 regions per condition).

Analysis of T-cell viability, activation, and proliferation

The viability, activation, and proliferation of T cells in each sample were assessed to determine the effect of Dex in the presence and absence of Mz. T cells were labeled with 1 μM 5-(and 6)-carboxyfluorescein diacetate succinimidyl ester (CSFE) (eBioscience™) and CFSE-labeled and non-labeled cells were treated with drugs as described above, with and without 5 μg/ml ConA activation before subjected to propidium iodide staining (for non-viable cell detection) and flow cytometry using the CytoFLEX Flow Cytometer (Beckman Coulter, CA, USA). The acquisition data were analyzed using the ModFit LT 6.0 software (Verity Software House, USA). Propidium iodide-positive cells were considered non-viable cells. Activated and non-activated T-cell populations were gated from propidium iodide-negative cells, with T cells cultured without ConA stimulation being used as a reference for non-activated cells in the flow cytometric dot plots. The proliferation index (PI) was analyzed using the ModFit LT 6.0 software, which automatically calculated the PI of T cells as the sum of cells in all generations divided by the number of original parent cells. The proportion of divided T cells was also obtained from the software.

Statistical analysis

All statistical analyses were conducted using GraphPad Prism version 9.0.0 (GraphPad Software, Boston, MA) and STATA version 16.0 software (StataCorp LLC, College Station, Texas). The data normality was assessed using the Shapiro–Wilk test prior to conducting one-way ANOVA. One-way ANOVA with Bonferroni’s post hoc test was performed to determine statistically significant differences between multiple groups. A p value < 0.05 was considered as statistically significant.