Thermoresponsive carbohydrate polymer mucoadhesive gel for synergistic delivery of acarbose and fluconazole against Candida biofilms Authors: Helma David, Pooraniammal Balu, Sahana Vasudevan, Vedha Hari B. Narayanan, Adline Princy Solomon, D. Ramyadevi Published: 2026 ABSTRACT The emerging threat of drug-resistant Candida albicans, especially in recurrent vulvovaginal candidiasis (VVC) biofilms, has revealed weaknesses in existing monotherapeutic antifungal treatments. In this study, we investigated the antifungal activity of a new dual-drug in situ mucoadhesive gel system containing acarbose (AC), a pseudo-oligosaccharide sugar and FDA-approved α-glucosidase inhibitor, together with fluconazole (FLC). Although AC showed mild antifungal activity by itself, when combined with FLC, it dramatically potentiated fungicidal activity, disrupted biofilms, blocked morphogenetic switching, and arrested post-treatment hyphal regrowth. In checkerboard assays and time-kill kinetics, AC-FLC combination was highly synergistic against Candida spp., including clinical isolates, with > 2-log₁₀ decrease in CFU/mL. The thermosensitive gel (PCH16), formulated using poloxamer 407 along with the carbohydrate-based polymers chitosan and HPMC, exhibited sustained drug release, favorable rheological properties, and strong mucoadhesion suitable for vaginal delivery. Notably, the formulation retained Lactobacillus species viability, suggesting microbiome compatibility. These results affirm the promise of repurposing acarbose as a novel, non-traditional antifungal adjuvant that, in co-administration with FLC in vaginal gel, represents an encouraging, localized, and microbiota-conserving therapy for treating recurrent and resistant VVC.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-026-52764-5. Introduction Vaginal fungal infections by Candida species have been a global health concern, and among these, Candida albicans and non-albicans species such as C. glabrata, C. tropicalis, C. parapsilosis, and C. krusei are responsible for invasive candidiasis and for severe causes of nosocomial infection1,2. The treatment of such infections is increasingly complicated by diminishing therapeutic options and the emergence of antifungal resistance, especially to azoles. Fluconazole, an azole has been a primary antifungal for many years because it has oral bioavailability, broad activity against many species, and low toxicity3–5. Nonetheless, the in repeated use of antifungals, which has led to increased resistance, making therapeutic outcomes more difficult6,7. Mechanistically, Candida species have developed multiple strategies to resist azole action. These involve the overexpression of drug efflux pumps (e.g., CDR1, CDR2, MDR1), ERG11 gene mutations for the enzyme target of the azoles, and adaptive stress responses8,9. In addition, the roughbiofilms formed by Candida on mucosal surfaces and medical equipment enhances resistance, rendering treatment strategies developed against planktonic forms mostly ineffective in many cases10,11. Biofilms present physical and biochemical barriers that protect Candida cells against antifungal compounds and immune attack, calling for the use of different therapeutic approaches12,13. To mitigate these challenges, combination antifungal therapy is increasingly being explored as an effective therapeutic approach.The use of drug combinations can improve therapeutic outcomes, lower dose-dependent toxicity, expand antifungal coverage, and reduce the emergence of resistance14,15. These synergistic combinations have been successful in bacterial and viral diseases and are increasingly being explored in fungal infections16,17. In addition to pharmacologic approaches, interest has focused on the creation of locally administered drug delivery systems that may deliver sustained antifungal activity directly to the site of infection18,19. For VVC, vaginal route is suited for drug delivery because it has a large surface area, is richly vascularized, and avoids hepatic first-pass metabolism19,20. However, conventional dosage forms such as creams, foams, and tablets are associated with limitations including poor retention, leakage, and local irritation, which may compromise therapeutic efficacy and patient compliance21,19. Mucoadhesive and thermosensitive gel-based systems have been developed to address these limitations22. These systems increase drug residence time, preserve localized concentration, and minimize systemic side effects23. Having incorporated biocompatible polymers such as chitosan, poloxamer, and hydroxypropyl methylcellulose (HPMC), the gels provide enhanced mechanical strength and mucoadhesion24,25. Chitosan also provides intrinsic antifungal and antibiofilm activities, thus complementing the drug action of loaded drugs26. Thermo-responsive gels additionally improve performance by becoming gel at body temperature, enhancing ease of use and patient compliance27. The other important factor in VVC management is the influence of antifungal therapy on the vaginal microbiota. The normal vaginal flora is characterized by a predominance of Lactobacillus species, i.e., L. crispatus, L. jensenii, and L. gasseri, which create an acidic environment and produce antimicrobial substances such as lactic acid and hydrogen peroxide28,29. Alteration of this eubiotic status usually heightened by broad-spectrum antifungals may result in dysbiosis, recurrence of the infection, and colonisation by opportunistic pathogens. Thus, antifungal treatments that maintain healthy microbiota without acting against pathogenic Candida species are very much wanted30,31. In this respect, the current study understands the antifungal activity of acarbose (AC), an a-glucosidase inhibitor, in relation to the species of Candida, where its functionality is poorly studied. In particular, this study assesses whether AC can potentiate existing antifungal agents, to enhance its efficacy against robust biofilms of Candida species. To substantiate this, an in situ mucoadhesive gel is designed and evaluated on the basis of its capacity to deliver long-term drug delivery, eradicate biofilms, without inhibiting the commensal populations of Lactobacillus species, which can help better manage recurrent VVC. Results Synergistic enhancement of fluconazole by acarbose in Candida albicans To study the antifungal activity of acarbose (AC) when combined with fluconazole (FLC), checkerboard assays were performed on three clinical isolates of C. albicans (SC5314, ATCC90028, and BF1). AC had a potentiating role over FLC, where it increased the activity of FLC even when used below its MIC break point (Fig. 1). Importantly, synergy was found at low AC concentrations (0.03–0.24 µg/mL strain dependent) with FLC at 4 µg/mL. Bliss synergy scores were significantly high (37.8, 44.9, and 27.4 for SC5314, ATCC90028, and BF1, respectively), reflecting intense synergistic interactions. Mechanistic evidence implies that AC interferes with key virulence determinants such as biofilm formation and efflux pump functioning, resulting in enhanced intracellular accumulation of FLC and increased fungal cell lysis. These findings lend credence to the hypothesis that AC, as such, has no appreciable antifungal activity but is a powerful adjuvant upon combination with azoles. Fig. 1Bliss Independence model analysis of FLC-AC drug-drug interaction against C. albicans strains SC5314, ATCC90028, and BF1. (a) 3D synergy surface plot of SC5314 with a peak synergy score of 37.8, indicating strong synergism. (b) % inhibition bar graph corresponding to SC5314 at a range of concentrations of FLC and a constant AC concentration of 0.03 µg/mL. (c) Synergy plot of ATCC90028 with a synergy score of 44.9, representing the highest level of synergistic interaction. (d) Inhibition profile of ATCC90028 with AC at 0.12 µg/mL. (e) BF1 synergy surface plot with a synergy score of 27.4, confirming positive interaction. (f) Bar graph of % inhibition in BF1 for different FLC concentrations with constant AC at 0.24 µg/mL. Data are presented as mean ± SD (n = 3). Statistical analysis was performed using one-way ANOVA. p < 0.001 (*) and p < 0.0001 (**) were considered statistically significant compared to untreated control. Bliss Independence model analysis of FLC-AC drug-drug interaction against C. albicans strains SC5314, ATCC90028, and BF1. (a) 3D synergy surface plot of SC5314 with a peak synergy score of 37.8, indicating strong synergism. (b) % inhibition bar graph corresponding to SC5314 at a range of concentrations of FLC and a constant AC concentration of 0.03 µg/mL. (c) Synergy plot of ATCC90028 with a synergy score of 44.9, representing the highest level of synergistic interaction. (d) Inhibition profile of ATCC90028 with AC at 0.12 µg/mL. (e) BF1 synergy surface plot with a synergy score of 27.4, confirming positive interaction. (f) Bar graph of % inhibition in BF1 for different FLC concentrations with constant AC at 0.24 µg/mL. Data are presented as mean ± SD (n = 3). Statistical analysis was performed using one-way ANOVA. p < 0.001 (*) and p < 0.0001 (**) were considered statistically significant compared to untreated control. Inhibition of morphological switching in Candida albicans One of the characteristics of C. albicans virulence is plasticity of its morphology, particularly the yeast-hyphae transition. Through the in vitro hyphal inhibition assay, it was shown that AC, even at low doses, prevented hyphal growth when used alone. But with AC combined with FLC, the effect was much stronger, causing full inhibition of hyphal as well as pseudo-hyphal morphotypes in all test strains (Fig. 2; S1 and S2). This synergistic effect indicates dual therapy efficiently represses major signaling pathways that are implicated in hyphal induction and thereby compromises fungal invasiveness as well as pathogenesis. Fig. 2Phase contrast microscopy demonstrating the inhibitory activity of FLC, AC, and their combination on hyphal growth of C. albicans SC5314. Scale bar: 20 μm. Phase contrast microscopy demonstrating the inhibitory activity of FLC, AC, and their combination on hyphal growth of C. albicans SC5314. Scale bar: 20 μm. Accelerated and sustained fungicidal activity Time-kill assays conducted longer than 24 h revealed that FLC alone caused a decrease in levels of CFU initially, which reached a plateau after a certain time period, showing a plateaued effect (Fig. 3). Treatment with AC left the retained and functional fungicidal activity, more than 2-log CFU reduction, and no indication of regrowth. The sustained antifungal action demonstrates the potential of the AC-FLC combination to treat recalcitrant infections, particularly those that are non-susceptible to monotherapy. Fig. 3Time-kill kinetics of FLC, AC, and their combination against clinical isolates of C. albicans. (a) SC5314, (b) ATCC90028, and (a) BF1 were treated with FLC, AC, FLC + AC, or left untreated. Viable fungal counts (CFU/mL) were determined at designated time points over a 24-h period. Data are presented as mean ± SD from three independent experiments. Time-kill kinetics of FLC, AC, and their combination against clinical isolates of C. albicans. (a) SC5314, (b) ATCC90028, and (a) BF1 were treated with FLC, AC, FLC + AC, or left untreated. Viable fungal counts (CFU/mL) were determined at designated time points over a 24-h period. Data are presented as mean ± SD from three independent experiments. Disruption of mature biofilms The efficacy of the AC-FLC combination was also tested against mature C. albicans preformed biofilms for 24, 48, and 72 h (Fig. 4). Monotherapy with FLC resulted only in partial inhibition (~ 40%), whereas combination therapy resulted in a significant biomass reduction at all time points. This finding was consistent in both SC5314 and clinical strains, indicating the potential of the two-drug system to overcome one of the most formidable features of Candida pathogenesis: biofilm-linked resistance. Fig. 4Biofilm eradication of C. albicans strains after treatment with FLC, AC, and combination. (a) SC5314, (b) ATCC90028, and (c) BF1 were treated with single drugs or combination therapy, and biofilm biomass inhibition was measured at 24, 48, and 72 h. Biofilm eradication of C. albicans strains after treatment with FLC, AC, and combination. (a) SC5314, (b) ATCC90028, and (c) BF1 were treated with single drugs or combination therapy, and biofilm biomass inhibition was measured at 24, 48, and 72 h. Sustained post-treatment suppression of hyphal regrowth Post-treatment regrowth assays proved that FLC alone permitted recovery of hyphae within hours upon drug withdrawal, whereas AC alone maintained yeast morphology without inducing notable hyphae (Fig. 5; S3 and S4). The combination therapy, however, abolished altogether the regrowth of hyphae even after 6 h, suggesting a post-antihyphal effect that could prove pivotal for prevention of recurrence and enhancement of the strength of this combination strategy. Fig. 5Phase contrast microscopic imaging of the post-antihyphal effect of FLC, AC, and combination on C. albicans SC5314 at 2, 4, and 6 h after treatment. Scale bar: 20 μm. Phase contrast microscopic imaging of the post-antihyphal effect of FLC, AC, and combination on C. albicans SC5314 at 2, 4, and 6 h after treatment. Scale bar: 20 μm. Alteration in colony morphology under hyphal-inducing conditions Colony morphology analysis in hyphal-inducing conditions further validated the anti-morphogenetic activity of the combination. Whereas control and FLC-treated colonies had wrinkled, filamentous morphologies, AC alone and the combination caused a smooth, compact colony morphology (Fig. 6). This phenotypic shift is correlated with inhibition of filamentation, extracellular matrix synthesis, and biofilm-associated gene expression, validating the multifactorial impairment of fungal pathogenic characteristics. Fig. 6C. albicans strains SC5314, ATCC90028, and BF1 colony morphology after treatment with FLC, AC, and their combination. C. albicans strains SC5314, ATCC90028, and BF1 colony morphology after treatment with FLC, AC, and their combination. Simultaneous quantification and compatibility of AC and FLC A dual-wavelength strategy was proven for simultaneous estimation of AC and FLC (Fig. 7). AC was subjected to chemical derivatization with KMnO₄/NaOH to form a stable chromophore that can be measured at 610 nm, whereas FLC was measured directly at 261 nm (Fig. 8). The method gave good linearity and specificity without cross-talk spectral interference. Physical and chemical compatibility of the two drugs was established by DSC and FTIR studies in favour of their co-formulation within a single gel matrix (Fig. S5 and S6). Fig. 7UV spectrophotometric scan of AC and FLC in distilled water. (a) UV spectrum of AC with an absorbance at 206 nm. (b) Acarbose calibration curve at 206 nm showing linearity (R² = 0.9994) within the concentration range tested. (c) UV spectrum of FLC with a strong and stable absorbance peak at 261 nm, ideal for direct quantification. (d) Fluconazole calibration curve at 261 nm with good linearity (R² = 0.9976). UV spectrophotometric scan of AC and FLC in distilled water. (a) UV spectrum of AC with an absorbance at 206 nm. (b) Acarbose calibration curve at 206 nm showing linearity (R² = 0.9994) within the concentration range tested. (c) UV spectrum of FLC with a strong and stable absorbance peak at 261 nm, ideal for direct quantification. (d) Fluconazole calibration curve at 261 nm with good linearity (R² = 0.9976). Fig. 8UV spectrophotometric determination of Acarbose and Fluconazole after derivatization and direct detection, respectively. (a) UV-Vis spectrum of derivatized Acarbose with intense and stable absorption in the visible range as a result of the formation of colored complex after reacting with KMnO₄ and NaOH. (b) UV spectrum of Fluconazole with intense and specific absorbance peak, ideal for direct quantification. (c) Calibration curve of Acarbose after derivatization, which is linear with absorbance and concentration and R² = 0.999. (d) Calibration curve of Fluconazole at 261 nm, which is linear with R² = 0.999. UV spectrophotometric determination of Acarbose and Fluconazole after derivatization and direct detection, respectively. (a) UV-Vis spectrum of derivatized Acarbose with intense and stable absorption in the visible range as a result of the formation of colored complex after reacting with KMnO₄ and NaOH. (b) UV spectrum of Fluconazole with intense and specific absorbance peak, ideal for direct quantification. (c) Calibration curve of Acarbose after derivatization, which is linear with absorbance and concentration and R² = 0.999. (d) Calibration curve of Fluconazole at 261 nm, which is linear with R² = 0.999. Optimization of thermoresponsive mucoadhesive gel formulation Of the 18 tested formulations, PCH16 proved to be optimal considering pH (3.64–4.95), sol-gel transition temperature (~ 32 °C), viscosity, and rheological behavior (Table S1). The formulation showed optimal thermoresponsive and mucoadhesive properties for vaginal use, with ease of administration and extended retention at the infection site (Fig. 9). The drug content of the optimized formulation (PCH16) was found to be 93.4 ± 1.8% for fluconazole and 91.7 ± 2.1% for acarbose (n = 3), indicating uniform drug distribution and minimal loss during formulation. The high drug content further confirms the efficiency of the incorporation process and suitability of the formulation for sustained vaginal drug delivery. Fig. 9Viscosity and Thermo-Responsive Profiles of Selected In Situ Gel Formulations. Viscosity profiles of five shortlisted formulations (PCH9, PCH10, PCH14, PCH15, and PCH16) at different shear rates and temperatures. All of the studied formulations have shown shear-thinning behavior, which is manifested as a decline in viscosity with increasing shear rate, supporting applicability for administration and retention at the application site. Another significant feature obtained from temperature-dependent viscosity curves is sol-to-gel transitions for each formulation: ~30 °C for PCH9 and PCH14, ~ 32 °C for PCH10 and PCH15, and ~ 34 °C for PCH16. These gelation temperatures lie within the physiological vaginal temperature range (33–37 °C), indicating that these are thermos-responsive for vaginal drug delivery. Viscosity and Thermo-Responsive Profiles of Selected In Situ Gel Formulations. Viscosity profiles of five shortlisted formulations (PCH9, PCH10, PCH14, PCH15, and PCH16) at different shear rates and temperatures. All of the studied formulations have shown shear-thinning behavior, which is manifested as a decline in viscosity with increasing shear rate, supporting applicability for administration and retention at the application site. Another significant feature obtained from temperature-dependent viscosity curves is sol-to-gel transitions for each formulation: ~30 °C for PCH9 and PCH14, ~ 32 °C for PCH10 and PCH15, and ~ 34 °C for PCH16. These gelation temperatures lie within the physiological vaginal temperature range (33–37 °C), indicating that these are thermos-responsive for vaginal drug delivery. The stability of the optimized formulation (PCH16) was evaluated under real-time and accelerated conditions. Under real-time storage, the formulation remained physically stable with no visible changes in appearance, pH, or viscosity over 3 months. Drug content of both fluconazole and acarbose showed minimal reduction, remaining above 89%, indicating good chemical stability. Under accelerated conditions, a slight decrease in viscosity and drug content was observed; however, no phase separation or significant degradation was noted. The pH remained within the acceptable vaginal range, suggesting formulation stability. These findings indicate that the formulation maintains both physical and chemical stability under tested conditions (Table S2 and S3). Contact angle measurements were performed using the sessile drop method to evaluate the surface characteristics of the formulation. A total of six valid measurements were obtained. The mean contact angle was 38.97 ± 1.05°, indicating a hydrophilic surface favorable for mucoadhesion and spreading on biological surfaces (Figure. These findings suggest that the developed formulation possesses suitable wettability characteristics for effective mucosal interaction and retention. Comparative evaluation of drug release profiles among gel formulations Comparative study of drug release from different in situ gel formulations (PCH9, PCH10, PCH14, PCH15, and PCH16) showed remarkable differences in the release pattern, which were mostly controlled by the composition and concentration of polymers involved in the formulation. Not only do such polymers regulate the structure integrity and hydrating characteristics of the gel, but they also determine drug-polymer interaction and the kinetic rate of gelation. Even minimal modifications in ratios of polymers led to drastic changes in rates of diffusion of drugs and general release patterns (Fig. 10). Fig. 10In Vitro Drug Release Profiles of Acarbose (AC) and Fluconazole (FLC) from In Situ Mucoadhesive Gel Formulations. Drug release from five formulations (PCH9, PCH10, PCH14, PCH15, and PCH16) was assessed for 48 h at physiological and vaginal infection mimicking conditions (pH 6.8). Each figure illustrates a monotherapy (Gel_AC, Gel_FLC) and combination therapy (GelCom_AC, GelCom_FLC). PCH10 had the highest release profile, whereas PCH9, PCH14, and PCH15 had moderate profiles. PCH16 presented with a controlled, sustained drug release of both drugs, with > 80% cumulative release and no burst effect. This ensures that PCH16 will be a robust and efficient vaginal drug delivery formulation. Data are mean ± SD (n = 3). In Vitro Drug Release Profiles of Acarbose (AC) and Fluconazole (FLC) from In Situ Mucoadhesive Gel Formulations. Drug release from five formulations (PCH9, PCH10, PCH14, PCH15, and PCH16) was assessed for 48 h at physiological and vaginal infection mimicking conditions (pH 6.8). Each figure illustrates a monotherapy (Gel_AC, Gel_FLC) and combination therapy (GelCom_AC, GelCom_FLC). PCH10 had the highest release profile, whereas PCH9, PCH14, and PCH15 had moderate profiles. PCH16 presented with a controlled, sustained drug release of both drugs, with > 80% cumulative release and no burst effect. This ensures that PCH16 will be a robust and efficient vaginal drug delivery formulation. Data are mean ± SD (n = 3). For example, PCH10 formulation exhibited immediate and almost complete release of acarbose and fluconazole in the initial few hours. This implied a loosely packed matrix with low cross-linking density. The loose morphology would most probably allow water penetration and drug solubilization at a very rapid rate, creating an undesirable burst release profile. Such types of profiles are not favored in long-acting delivery systems, particularly for vaginal therapy where drug retention is desired for a prolonged period of time. Conversely, some formulations like PCH9 and PCH14 displayed extended and sustained release, especially for acarbose, which plateaued between 45% and 60% over 48 h. This could be due to greater intermolecular interactions, including hydrogen bonding or ionic interaction between HPMC and chitosan, which could have generated denser and more networked polymeric matrices. These more compact structures probably restrained early drug diffusion, particularly for hydrophilic drugs such as acarbose, thus restraining burst effects and predisposing towards sustained release. Interestingly, PCH15 displayed dual behavior in which the individual-drug acarbose gel (Gel_AC) showed complete release, while the composite drug gel (GelCom_AC) released only ~ 50% of acarbose. This is an indication for chances of drug-drug or drug-polymer interactions in the composite matrix that can restrict drug mobility. Highly water-soluble acarbose could have been subjected to competition for hydration or more potent retention in a viscous, dense network created in the presence of fluconazole. Among all the formulations that were investigated, PCH16 exhibited the most optimal drug release pattern. It released both the drugs in controlled and sustained fashion with cumulative drug release of more than 80% in both single and double systems within 48 h. The increased performance of PCH16 may be due to synergy among the polymers: Poloxamer 407 offers temperature-sensitive gelation and scaffolding, chitosan offers weak cross-linking and mucoadhesiveness, and HPMC controls viscosity and swelling behavior. Together, these offer a very well-hydrated, moderately viscous gel network that resists water penetration and erosion equally and thus ensures uniform and prolonged diffusion of the drug. The lack of burst release and efficient drug release at a wide range of pH values reflects the stability of PCH16 as a two-drug delivery system for best vaginal therapeutic use. Release kinetics and diffusion mechanism of active agents from in situ gels For better insight into the mechanism of release of fluconazole and acarbose from the gel matrix, the cumulative release data were mathematically modeled employing DDSolver software. The data were correlated to several kinetic models such as zero-order, first-order, Higuchi, Korsmeyer–Peppas, and Hixson–Crowell equations. Of these, the Korsmeyer–Peppas model gave the best fit for both the drugs according to the highest R² value and the lowest sum of squared residuals (SSR). The release exponent (n) obtained from the Korsmeyer–Peppas model was less than 0.5 for both drugs, proving Fickian diffusion to be the most dominating mechanism of drug release. This implies that the release is mostly controlled by the diffusion of drug molecules within the hydrated polymeric matrix and not by erosion of the matrix (Table S4). These results verify that the PCH16 gel system provides sustained drug availability and is an appropriate platform for long-term local therapy in vaginal candidiasis. Acarbose’s selective antibiofilm activity against non-albicans Candida Selective anti-biofilm action of acarbose against various non-albicans Candida species, i.e., C. glabrata, C. krusei, C. parapsilosis, C. tropicalis, which were not inhibited by the anti-biofilm compound under planktonic growth (Fig. 11 and Table S5). Dispersed and reduced hyphal forms were observed using microscopy techniques (Fig. 12). Distorted biofilm architecture was also demonstrated. AC was demonstrated by this research to influence biofilm-associated morphogenesis as an anti-biofilm compound but without exerting a fungistatic effect. Fig. 11Dose-dependent effect of AC on growth and biofilm formation by non-albicans Candida species. Each panel shows the reaction of a distinct non-albicans strain to different concentrations of AC. Growth (black line) and biofilm formation (blue line) were determined spectrophotometrically and shown as a function of AC concentration. Fungal growth was generally unaffected within the tested concentrations, whereas pronounced inhibition of biofilm formation occurred in a concentration-dependent fashion. Data points represent mean ± SD from three independent experiments. Dose-dependent effect of AC on growth and biofilm formation by non-albicans Candida species. Each panel shows the reaction of a distinct non-albicans strain to different concentrations of AC. Growth (black line) and biofilm formation (blue line) were determined spectrophotometrically and shown as a function of AC concentration. Fungal growth was generally unaffected within the tested concentrations, whereas pronounced inhibition of biofilm formation occurred in a concentration-dependent fashion. Data points represent mean ± SD from three independent experiments. Fig. 12Phase contrast microscopy images demonstrating the effects of Acarbose on biofilm morphology of non-albicans Candida strains. Representative images of untreated (left panels) and acarbose-treated (right panels) cells demonstrate a significant decrease in cell aggregation, filamentation, and hyphal formation in all strains. Scale bars: 20 μm. Phase contrast microscopy images demonstrating the effects of Acarbose on biofilm morphology of non-albicans Candida strains. Representative images of untreated (left panels) and acarbose-treated (right panels) cells demonstrate a significant decrease in cell aggregation, filamentation, and hyphal formation in all strains. Scale bars: 20 μm. Efficacy of AC on lactobacillus AC was found to be very selective and did not inhibit the growth of vaginal Lactobacillus species, like L. acidophilus, L. casei, and MTCC4183 and MTCC4184 (Fig. S7). This establishes a microbiome-friendly profile and the drug’s suitability for intravaginal administration without compromising the beneficial commensals. Combinatorial efficacy of acarbose and fluconazole against non-albicans Candida Checkerboard assays established that AC synergizes with FLC’s antifungal activity against a wide range of non-albicans Candida strains (Fig. 13; S7 and Table S5). Synergy scores based on the Bliss Independence Model were all > 10, with the exception of C. tropicalis MTCC38, which was inherently FLC-sensitive. Synergistic concentrations (AC 0.96 µg/mL, FLC 4 µg/mL) were tested and used in preparing the dual-drug gel (PCH16). The data validate that the AC-FLC combination is resistant-breaking and augments antifungal action in resistant strains. Fig. 13Combinatorial effect of AC and FLC on planktonic growth inhibition in non-albicans Candida strains. Bar graphs illustrate the dose-dependent inhibition of planktonic cells in eight non-albicans Candida strains treated with AC, FLC, and their combination using a checkerboard assay. A synergistic effect, evaluated via the Bliss Independence Model, was observed in all tested strains except C. tropicalis MTCC38. Synergy scores exceeding 10 confirm significant synergistic interaction. Statistical significance was determined using One-way ANOVA. Combinatorial effect of AC and FLC on planktonic growth inhibition in non-albicans Candida strains. Bar graphs illustrate the dose-dependent inhibition of planktonic cells in eight non-albicans Candida strains treated with AC, FLC, and their combination using a checkerboard assay. A synergistic effect, evaluated via the Bliss Independence Model, was observed in all tested strains except C. tropicalis MTCC38. Synergy scores exceeding 10 confirm significant synergistic interaction. Statistical significance was determined using One-way ANOVA. Dual-drug gel broad-spectrum antibiofilm activity (PCH16) The synergistic mixture PCH16 of concentrations of AC and FLC was compared to mature biofilms (24–72 h) of C. albicans and non-albicans Candida species. The formulation displayed constant better activity against monotherapies at every time point, including resistant 72-hour biofilms (Fig. 14). The broad-spectrum, time-dependent killing of biofilms demonstrates the potential of PCH16 to cure recurrent and persistent Candida infection. Fig. 14Antifungal efficacy of dual-drug loaded in-situ Mucoadhesive gel (PCH16) on preformed biofilms Candida biofilms matured for (a) 24 h, (b) 48 h, and (c) 72 h. CFU/mL counts of C. albicans and non-albicans Candida strains indicate that PCH16 had significantly reduced viable cells in biofilms at all stages of maturity compared with control and other treatment groups, and it exhibited strong and long-lasting antifungal activity. Antifungal efficacy of dual-drug loaded in-situ Mucoadhesive gel (PCH16) on preformed biofilms Candida biofilms matured for (a) 24 h, (b) 48 h, and (c) 72 h. CFU/mL counts of C. albicans and non-albicans Candida strains indicate that PCH16 had significantly reduced viable cells in biofilms at all stages of maturity compared with control and other treatment groups, and it exhibited strong and long-lasting antifungal activity. Fig. 15Impact of PCH16 (Gel_AC_FLC) on vaginal commensal Lactobacillus strain viability. CFU counts of L. acidophilus, L. casei, and Lactobacillus sp. MTCC4183 and MTCC4184 after being treated with different gel formulations, including the dual-drug gel. No noticeable decrease in CFU compared to control (p > 0.05, unpaired t-test), confirming that PCH16 has no negative impact on commensal bacterial viability. Impact of PCH16 (Gel_AC_FLC) on vaginal commensal Lactobacillus strain viability. CFU counts of L. acidophilus, L. casei, and Lactobacillus sp. MTCC4183 and MTCC4184 after being treated with different gel formulations, including the dual-drug gel. No noticeable decrease in CFU compared to control (p > 0.05, unpaired t-test), confirming that PCH16 has no negative impact on commensal bacterial viability. Compatibility of PCH16 gel with vaginal commensals PCH16 was very safe with vaginal Lactobacillus species (Fig. 16). CFU counts for all strains tested did not vary statistically after treatment, which verifies the gel preserves healthy microbial flora. This microbiome-sparing property further facilitates its application for extended, local vaginal therapy. Fig. 16Schematic representation of the development and evaluation of a dual-drug strategy against vaginal candidiasis. Schematic representation of the development and evaluation of a dual-drug strategy against vaginal candidiasis. Materials and methods Chemicals and reagents The reagents and chemical used in this study were all of analytical grade. Fluconazole (FLC), amphotericin B (AMB), flucytosine (FLU), caspofungin (CAS), and acarbose (AC) were acquired at SRL, India (purities > 95%) and left untouched. Chitosan (degree of deacetylation of 85, viscosity of over 500 cP) and lactic acid-conjugated chitosan were sources of Marine Bio-Polymers LLP, New Delhi, India. As a viscosity modifier, Hydroxypropyl methylcellulose (HPMC, Mw of approximately 86 kDa, 28–30% methoxyl substitution, 7–12% hydroxypropyl substitution; HiMedia) was employed, and poloxamer 407 (Kolliphor(r), BASF; Mw approximately 12.6 kDa, an approximation of 70% PEO content) was utilised as the thermosensitive gel modifier. Sodium hydroxide, glacial acetic acid, potassium permanganate and potassium dihydrogen phosphate, among other chemicals, were bought in Merck Life Sciences (India). RPMI 1640 was purchased at HiMedia. Organisms, media, and culture conditions Candida albicans isolates (SC5314, ATCC90028, BF1), generously provided by Dr. Prasanna Neelakandan from the University of Hong Kong, as well as non-albicans like C. glabrata, C. tropicalis, C. krusei, and C. parapsilosis were used. Lactobacillus strains (such as L. acidophilus, L. casei, and two MTCC strains MTCC4183 and MTCC4184) were also used to check for microbiome compatibility. All bacterial and fungal strains were preserved in 20% glycerol at − 80 °C and sub-cultured on respective media (YPD for Candida spp., MHB for Lactobacillus) prior to use. Standardized cell suspensions of around 1 × 10⁷ CFU/mL were made by adjusting optical density (OD520 or OD595) after centrifugation and washing processes using PBS. In vitro antifungal activity Checkerboard drug interaction assay The synergistic effect of the combination of fluconazole and acarbose was evaluated by the broth microdilation checkerboard method. Double dilutions of both drugs were made in YPD broth and seeded with standardized Candida cell suspensions in 96-well microtiter plates. Optical density at 595 nm was read after incubation at 37 °C for 24 h using a BioRad i-Mark microplate reader. The Bliss Independence model was used with SynergyFinder Plus (https://synergyfinder.org/) software to analyze drug interactions, where synergy scores > 10 were deemed synergistic. Time-kill kinetics assay To assess time-dependent antifungal efficacy, aliquots were collected at 0, 2, 4, 6, 8, 12, and 24 h after incubation with each drug alone or in combination. Aliquots were seeded on YPD agar after dilution, and CFU were counted after 24-h incubation at 37 °C. ≥2 log₁₀ CFU/mL reduction from the initial inoculum was considered as fungicidal activity. Hyphal inhibition and post-hyphal regrowth assay Hyphal development was provoked by growing Candida in Spider medium with the addition of 10% sterile fetal bovine serum. FLC-, AC-, or combination-treated cells were inspected after 24 h at 37 °C by a Nikon Eclipse Ts2 microscope. To determine hyphal regrowth after drug treatment, the treated cells were washed and suspended in RPMI 1640 and left for another 2 h before they were visualized. Biofilm formation and disruption assays Biofilm formation was begun in 96-well or 24-well plates by culturing Candida in YPD with 1% glucose. Drugs were introduced during biofilm formation for inhibition assays, and for eradication assays, the biofilms were pre-formed for 24, 48, or 72 h before treatment. Quantification was done using crystal violet staining, and CFU count followed after mechanical biofilm disruption and plating. Colony morphology analysis YPD agar supplemented with 1% glucose was employed to analyze colony morphology as influenced by drug treatment. Drug-supplemented agar was seeded with 5 µL of overnight Candida culture and incubated at 37 °C for 5 days. The colony morphology was recorded under a Leica Ivesta 3 stereo microscope. Mucoadhesive in-situ gel formulation Preparation procedure Cold method was employed to prepare in situ gels. Poloxamer 407 was dissolved in cold water (4 °C) and continuously stirred and kept overnight. Chitosan was dissolved in 1% acetic acid and HPMC was dispersed in distilled water. These polymer solutions were mixed under cold conditions to create a uniform gel base. FLC and AC were added in the optimized concentration to the blend. Eighteen different formulations (PCH1–PCH18) with different ratios of polymers were prepared and kept at 4 °C. Pre-formulation studies The drug content was determined using UV-visible spectroscopy. For Fluconazole (FLC), direct UV analysis was by scanning standard solutions to determine its maximum absorbance. For Acarbose (AC), both direct UV and a derivatized colorimetric technique were utilized. In the latter, AC was oxidized with potassium permanganate (KMnO₄) with sodium hydroxide (NaOH) to obtain a coloured complex determinable at 610 nm. Calibration curves in pH 4.5 and pH 6.8 buffers were prepared to mimic vaginal and physiological conditions. Fourier-transform infrared (FTIR) spectroscopy was utilized to determine possible chemical interactions between the excipients and drugs. Pure compound and physical mixture characteristic functional group peaks were evaluated and compared. Differential scanning calorimetry (DSC) was utilized to investigate the thermal behavior and compatibility of AC, FLC, and the 1:1 physical mixture of AC and FLC. Thermograms were evaluated for melting points as well as any shift or disappearance of peaks that signify interaction or instability. Gel characterization pH of the gels was determined with a digital pH meter, and standard buffers were used to calibrate it. Sol-to-gel temperature transitions were measured by slow heating (25–40 °C) and observing the gelation point through the inverted tube method. Viscosity was determined at several shear rates with a Brookfield viscometer and validated with rheometric analysis at physiological temperatures. Drug content determination Drug content of the optimized mucoadhesive in situ gel formulation (PCH16) was determined by UV–visible spectrophotometric analysis. Accurately weighed 1 g of the gel formulation was transferred into a 10 mL volumetric flask and diluted with distilled water (or appropriate buffer, pH 4.5/6.8) to extract the incorporated drugs. The mixture was vortexed and sonicated for 20–30 min to ensure complete drug release from the polymeric matrix and then kept at 4 °C for 24 h to facilitate complete extraction. The resulting solution was centrifuged at 10,000 rpm for 10 min (or filtered through a 0.22 μm membrane filter) to remove any undissolved polymeric residues. The supernatant was suitably diluted and analyzed spectrophotometrically. Fluconazole (FLC) content was determined directly at 261 nm, whereas acarbose (AC) was quantified using a derivatization method involving potassium permanganate (KMnO₄) and sodium hydroxide (NaOH), with absorbance measured at 610 nm, as described in the pre-formulation studies. Drug concentrations were calculated using previously constructed calibration curves under identical conditions. Drug content was calculated using the following equation: Drug content (%) = (Actual amount of drug in formulation / Theoretical amount of drug) × 100. All measurements were performed in triplicate and expressed as mean ± standard deviation. Stability studies of mucoadhesive gel The stability of the optimized formulation (PCH16) was evaluated under real-time and accelerated conditions as per ICH guidelines. For real-time stability, the formulation was stored at 25 ± 2 °C and analyzed at predetermined intervals (0, 15, 30 days, and 3 months). Accelerated stability studies were conducted at 40 ± 2 °C and 75 ± 5% RH for 1 month. At each interval, samples were evaluated for physicochemical parameters including appearance, pH, viscosity, and drug content. Any signs of phase separation, color change, or consistency variation were visually inspected. Drug content was determined using the validated UV spectrophotometric method. Contact angle measurements were performed using the sessile drop method to assess surface wettability. A small volume of the formulation was placed onto a clean glass surface, and the angle formed between the droplet and the surface was measured using a contact angle goniometer. Measurements were recorded at ambient conditions, and the mean value was calculated from multiple replicates. In vitro release and kinetics of drug Release studies for the dual-drug-loaded in situ gel formulation were performed using a dialysis membrane method that is intended to mimic physiological vaginal conditions. The formulations were filled into pre-soaked dialysis bags having a molecular weight cut-off of 12,000–14,000 Da. These were placed in receptor compartments with buffer solutions that were pH-adjusted to 4.5 and 6.8 to reflect healthy and slightly different vaginal environments. The receptor compartments were kept at 37 ± 0.5 °C and stirred constantly at 100 rpm to simulate dynamic vaginal fluid conditions. At pre-set time intervals, aliquots were removed from the receptor media and replenished with fresh buffer to ensure sink conditions. The samples so obtained were quantified using UV-visible spectrophotometry at appropriate wavelengths for fluconazole and acarbose, with reference to prior calibration. Drug concentrations were measured by interpolating from standard curves prepared under similar buffer conditions. To interpret the release mechanism of the drugs from the polymeric gel matrix, the cumulative drug release data were best fitted to different kinetic models such as zero-order, first-order, Higuchi, Korsmeyer–Peppas, and Hixson–Crowell using the DD Solver software. The best-fit model for every drug was determined based on the coefficient of determination (R²), release exponent (n), and model fitting indicators such as the sum of squares of residuals (SSR). This evaluation allowed the identification of the prevailing mechanism governing drug diffusion and matrix degradation under sustained release. Microbiome compatibility tests To establish the microbiome compatibility of the dual-drug in situ gel formulation, a sequence of in vitro tests was performed to investigate the selective antifungal efficacy of acarbose (AC) and the PCH16 formulae against Candida species and safety against commensal Lactobacillus strains. Susceptibility testing Minimum inhibitory concentration (MIC) of AC against non-albicans Candida species (C. glabrata, C. tropicalis, C. parapsilosis, and C. krusei) and Lactobacillus species (L. acidophilus, L. casei, Lactobacillus sp. MTCC4183 and MTCC4184) was assayed by the broth microdilation technique in 96-well microtiter plates. Acarbose was diluted in suitable media (YPD for Candida, MHB for Lactobacillus) in a concentration gradient of 1000 µg/mL to 0.02 µg/mL. Each well was treated with 10 µL of stock microbial suspension (1 × 10⁷ CFU/mL), and plates were incubated for 24 h at 37 °C. Growth inhibition was measured by absorbance at 595 nm (OD₅₉₅) with a BioRad i-Mark ELISA plate reader. MIC₉₀ was defined as the lowest concentration with ≥ 90% inhibition compared to an untreated control. Filamentation and biofilm inhibition For the assessment of AC’s anti-biofilm activity, non-albicans Candida strains were grown in 1% glucose-supplemented YPD medium and plated into 96-well microtiter plates. These were treated with serial dilutions of AC and incubated at 37 °C for 24 h. Planktonic cells were discarded, and wells were washed gently with PBS. Attaching biofilms were stained with 0.1% crystal violet for 15–20 min and dissolved with 33% acetic acid. OD₅₉₅ was measured for absorbance and percent inhibition determined by using the following formula: [(control − test) / (control − blank)] × 100. For the determination of filamentation, non-albicans Candida species were grown under filamentation-induced conditions (YPD + 1% glucose) with or without AC. The cultures were centrifuged at 12,000 rpm for 10 min following incubation at 37 °C for 24 h and suspended in PBS before microscopic observation with 40× magnification using a Nikon Eclipse Ts2 microscope to look for morphological changes towards the development of hyphae. Selectivity evaluation of formulation To ascertain the selectivity of the in-situ gel formula that was optimized (PCH16), its action against Candida biofilms and viability of Lactobacillus was evaluated. For assessing PCH16 activity against established non-albicans mature Candida biofilms pre-cultured in YPD-glucose in sterile 24-well plates, treatment of wells was as follows. After incubation periods at 37 °C for 24, 48, and 72 h, gently non-adherent cells were removed by PBS washing. Biofilms of the adherent were removed from the well surface, dilutional series in PBS, and plated on YPD agar using drop plate method. Colony-forming units (CFU/mL) were counted after incubation at 37 °C for 24 h. ≥2 log₁₀ CFU/mL decrease from the initial inoculum was taken as significant fungicidal effect. Discussion The growing threat of C. albicans antifungal resistance has led to a great deal of interest in the limitations of current monotherapeutic strategies32. Since the most widely used antifungals, fluconazole (FLC) is increasingly becoming less active against resistant Candida, particularly in biofilm-associated infections33. The ongoing excessive use and misuse of antifungal agents have hastened the rise in drug resistance by multiple mechanisms such as overexpression of efflux pumps, ERG11 and other target gene mutations, and alterations in metabolism resulting in drug sensitivity34,20. Further, the intrinsic potential of C. albicans to transition from yeast to hyphal cells and create structured biofilms further complicates the ability to treat, as biofilms provide a physical barrier to drug permeation and have persister cells that exhibit phenotypic resistance, irrespective of genetic mutations35–38. Our previous findings demonstrated that acarbose, a well-known inhibitor of α-glucosidase, exerts a significant effect on key virulence factors of Candida, including adhesion, morphogenesis, and biofilm formation, without affecting its overall growth. Notably, acarbose appears to operate through a novel mechanism by interfering with the processing of mannoproteins, which play a critical role in host–pathogen interactions. Therefore, to reduce the likelihood of recurrence, evaluating the potential of acarbose to enhance the efficacy of existing antifungal agents represents a promising therapeutic strategy worth further investigation39. To this end, the current study evaluates the therapeutic potential of the co-administration of existing antifungalswith acarbose (AC), anti-biofilm and anti-fungal properties. We demonstrate that although acarbose alone is not found to exhibit significant antifungal activity against planktonic cells, its co-administration with fluconazole has profound impacts on fungicidal activity. This synergistic effect was quantitatively established by time-kill kinetics, with a > 2-log drop in CFU/mL, and microscopically substantiated by irreversible disruption of hyphal morphology. This is in accordance with previous research where non-antifungal adjuvants such as piperine, thymol, and lovastatin were shown to potentiate azoles through the disruption of membrane integrity, inhibition of ergosterol biosynthesis, or the modification of stress response processes40,41. One of the more intriguing discoveries in our study is the inhibitory effect of AC-FLC combination on morphogenetic switching in C. albicans. While FLC on its own is not always an effective suppressor of filamentation and indeed can induce filamentation under stress or starvation conditions the addition of AC appears to reverse this action33. The most spectacular outcome of this combination therapy is activity against mature, 72-hour preformed Candida biofilms. AC-FLC combination was able to penetrate and break down the biofilms, reducing biomass and viable cells significantly. This result agrees with previous studies in which adjuvants such as antimicrobial peptides (e.g., ToAP2) or polyphenols (e.g., quercetin) enhanced biofilm killing when co-administered with fluconazole39,42,43. The increased activity is most likely to be due to increased drug penetration and interference with the matrix of the biofilm, due to AC’s anti-biofilm effect. Morphological examination of the colonies upon treatment showed a uniform transformation from the typical wrinkled and filamentous morphology to a smooth and dense one. The phenotypic transformation is generally associated with the downregulation of filamentation-associated genes like EFG1, TEC1, and HWP1, which are known to regulate hyphal growth in addition to biofilm organization44–46. Although other investigations have used gene deletion models for dissecting such pathways, our results indicate that pharmacological modulation through AC-FLC combination is able to reproduce the effect of genetic knockouts in clinical isolates of known tolerance. Another key finding of this study is the post-antihyphal effect, observed in that mature hyphae disaggregated in combination treatment. The post-antihyphal effect is of major therapeutic relevance, especially in chronic infection, as mature hyphal forms are involved in epithelial invasion and immune evasion. Reversal of hyphal growth may prevent recurrence and continued colonization, both characteristics of recurrent vulvovaginal candidiasis47,48. Concomitantly, the results here established firmly the potential of AC-FLC combination as a novel antifungal modality of wide-spectrum action (Fig. 16). Significantly, by enhancing antifungal efficacy without its detrimental impacts on commensal microbiota, this strategy is aligned with emerging trends in microbiome-friendly therapeutics. Conclusion and future perspectives A mucoadhesive, thermoresponsive gel co-delivering fluconazole and acarbose represents a promising localized therapeutic approach for vulvovaginal candidiasis (VVC). The formulation demonstrated synergistic antifungal activity by concurrently targeting key virulence and resistance determinants of Candida, including hyphal morphogenesis, biofilm formation, and efflux pump activity, while preserving Lactobacillus spp., suggesting preliminary microbiome compatibility. These findings indicate its potential as an effective strategy for the management of recurrent and drug-resistant VVC. The formulation offers distinct translational advantages, including site-specific delivery, prolonged residence time, and minimized systemic exposure. However, further considerations such as large-scale manufacturability, physicochemical stability, and batch reproducibility remain to be systematically addressed prior to clinical application. Although microbiome compatibility was evaluated using representative Lactobacillus species, comprehensive assessment of the vaginal microbial ecosystem is required. Future studies employing 16 S rRNA sequencing or metagenomic approaches will enable a more detailed understanding of microbial community dynamics and treatment-associated shifts. Subsequent investigations should focus on ex vivo permeation and retention studies to characterize pharmacokinetic behavior and optimize drug delivery, followed by in vivo validation in appropriate animal models to establish safety, efficacy, and dosing regimens. Additional evaluation against azole-resistant and high biofilm-forming Candida strains, as well as under polymicrobial conditions, will further substantiate its clinical relevance. Collectively, these findings support the potential of this dual-drug formulation for clinical translation as a targeted and microbiome-conscious therapeutic intervention for VVC and related candidiasis. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1