Preclinical approaches in vulvovaginal candidiasis treatment with mucoadhesive thermoresponsive systems containing propolis

Vulvovaginal candidiasis (VVC) is a common vaginitis that affects women, especially in childbearing age, caused by Candida albicans in almost 80% of cases. Considering the limited drug arsenal available and the increasing fungal resistance profile, the search for new therapeutic sources with low toxicity and easy administration should be supported. Propolis has been used as a traditional medicine for multiple diseases, considering its particular composition and pharmaceutical properties that permits its wide applicability; it has also emerged as a potential antifungal agent. Thus, this study performed an in vitro and in vivo investigation into the efficacy of a new mucoadhesive thermoresponsive platform for propolis delivery (MTS-PRPe) in a preclinical murine model of VVC treatment caused by C. albicans. The methodologies involved chemical analysis, an assessment of the rheological and mucoadhesive properties of propolis formulations, in vitro and in vivo antifungal evaluations, histological evaluations and electron microscopy of the vaginal mucosa. The results demonstrated the antifungal activity of propolis extract and MTS-PRP against the standard strain and a fluconazole-resistant clinical isolate of C. albicans, in both in vitro and in vivo assays. These results were similar and even better, depending on the propolis concentration, when compared to nystatin. Thus, the formulation containing propolis exhibited good performance against C. albicans in a vulvovaginal candidiasis experimental model, representing a promising opportunity for the treatment of this infection.


Preparation and characterization of mucoadhesive thermoresponsive systems
The systems were prepared using poloxamer 407 (P407) (20%, w/w) (Sigma, St. Louis, USA) and Carbopol 934P (C934P) (0.15%, w/w) (BF Goodrich, Ohio, USA). The carbomer was dispersed in purified water, with mechanical stirring. After total dispersion of C934P, the polymer P407 was added and the preparation was stored at 4˚C for 12 h to ensure complete hydration. Afterwards, the preparation was stirred, ensuring the complete mixing of the polymers, and the pH was set to 7.0 using triethanolamine (Galena, Campinas, SP, Brazil). Propolis extract (PRPe) was prepared as described in S1 Data and added to the system at a concentration of 14 or 16% (w/w) with mechanical agitation. All formulations were stored in amber jars at 4˚C for at least 24 h before further analysis [14].
The formulations were analyzed regarding pH and relative density [18]. The preliminary sol-gel transition temperature (preliminary Tsol/gel) was determined according to Choi and collaborators [19].

Rheological analysis
The rheological analysis of formulations was accomplished using a controlled stress rheometer (MARS II, Haake Thermo Fisher Scientific Inc., Newington, Germany) with parallel steel cone-plate geometry (60 mm, cone code L09006 C60/1˚TiL, separated by a gap of 0.052 mm).
Continuous shear (flow) rheometry. Flow analysis of formulations was performed at 5˚C, 25˚C, and 37˚C ± 0.1˚C. The thermometer was set in flow mode, then the sample was carefully placed on the lower plate and allowed to equilibrate for at least 1 min prior to analysis. Flow curves were measured over shear rates ranging from 0 to 2000 s-1 (increased over a period of 150 s, held at the upper limit for 10 s, and then decreased over a period of 150 s). In each case, the continuous shear properties of at least five replicates were determined, and the up curves were modeled using the Oswald-de-Waele equation (power law) described in Eq 1 [14,20]: where τ is the shear stress (Pa), k is the consistency index [(Pa.s)n], γ is the shear rate (s-1 ) and n is the flow behavior index (dimensionless). The yield value of each formulation was evaluated by the Casson (Eq 2) and Herschel-Buckley (Eq 3) rheological models [19]. where τ is the shear stress (Pa), n is the flow behavior index (dimensionless), τ0 is the yield stress (Pa), γ is the shear rate (s-1 ) and ηp is the Casson plastic viscosity.
where τ is the shear stress (Pa), τ0 is the yield stress (Pa), k is the consistency index [(Pas)n], γ is the shear rate (s-1 ) and n is the flow behavior index (dimensionless). The hysteresis area of each system was also calculated using Rheo Win 4. 10.0000 (Haakes) software. Oscillatory rheometry. The oscillatory rheometry of all formulations was performed at 5˚C, 25˚C and 37˚C ± 0.1˚C. A sample of each formulation was carefully applied on the plate as previously described. After determination of the linear viscoelastic region of each formulation, frequency sweep analysis was evaluated from 0.1 to 10.0 Hz. The storage modulus (G'), loss modulus (G@), dynamic viscosity (η'), and the loss tangent (tan δ) were calculated using Rheo Win 4. 10.0000 (Haake) software. In each case, the dynamic rheological properties of at least three replicates were evaluated [26].
In addition, the sol-gel transition temperature (Tsol/gel) of the formulations was also performed in oscillatory mode using temperature ramp analysis. After the determination of the linear viscoelastic region of each formulation at 5˚C and 60˚C, a temperature sweep analysis was performed over the temperature range of 5-60˚C at a defined frequency (1.0 Hz) and rate of heating 10˚C/min using a controlled stress. G', G", viscosity η', and tan δ were calculated using Rheo Win 4.10.0000 (Haake) software. Tsol/gel was determined for all formulations in which the η' increased with increasing temperature. Moreover, the temperature at which G 0 was halfway between the values for the solution and the gel was considered Tsol/gel [14,15].

Strains and culture methods
The in vitro susceptibility tests were performed with PRPe and conventional antifungals against C. albicans ATCC 90028 from the American Type Culture Collection (ATCC) and 88 C. albicans clinical isolates from patients with VVC that belong to the archive collection of the Laboratory of Medical Mycology, Universidade Estadual de Maringa, Parana, Brazil. These isolates were divided into three groups according to clinical manifestations: 40 clinical isolates from the colonized group (asymptomatic), 27 from the VVC group (single symptomatic episode of VVC infection), and 21 from the recurrent VVC group (RVVC) that means more than three episodes of VVC infection in the last year. Prior to testing, the strains were subcultured in Sabouraud dextrose agar (SDA; Becton, Dickinson and Company, Sparks, MD) and incubated at 37˚C for 24 h.
The inoculum preparation was performed through the suspension of the yeast in phosphate-buffered saline (PBS; 0.1 M; pH 7.4), adjusting the concentration by counting in a Neubauer chamber to 2.5-5.0 x 106 CFU mL-1. Further dilutions were prepared until the desired final inoculum value was reached (2.5-5.0 x 103 CFU mL-1) [20,21].

Antifungal susceptibility assays
The minimum inhibitory concentration (MIC) of FCZ and NYS and the minimum fungicidal concentration (MFC) of PRPe were determined by the broth microdilution method for 88 clinical isolates and one ATCC strain, and the MFC of F14% and F16% were assessed by the broth macrodilution method for four clinical isolates. All assays were performed according to Clinical and Laboratory Standards Institute (CLSI) document M27-A3 [22], adapted for natural compounds [17]. The final concentration of both FCZ and NYS ranged from 0.125 to 64 μg/mL, PRPe 5.23 to 2680 μg/mL of total polyphenol content (TPC) expressed as gallic acid equivalents, MTF-PRPe at 14% (F14%) ranged from 138.57 to 2217.18 μg/mL of TPC expressed as gallic acid equivalents, and MTF-PRPe at 16% (F16%) from 158.40 to 2534.40 of TPC expressed as gallic acid equivalents.
The antifungal activity of PRPe, F14%, and F16% were evaluated according to the MFC. For this, aliquots from each well of the PRPe microdilution assay and the F14% and F16% macrodilution assay were withdrawn and seeded in Petri dished with SDA medium to determine the MFC and the quantitative reduction in colony forming units (CFU/mL). The MFC was defined as the lowest concentration of the antifungal agent solution that inhibited 100% yeast growth.

Murine candidiasis vaginal model
The experiments were carried out according to the institutional regulations and were approved by the Institutional Ethics Committee for animal experimentation of the Universidade Estadual de Maringa, Brazil (approval number CEUA 4717180616, from 08/05/2016). The animals were treated according to the Guide for the Care and Use of Laboratory Animals from the Institute of Laboratory Animal Resources/USA [25]. Female Balb/c mice (6-8 weeks of age and weighing 20-22 g) were obtained from Bioterio Central of Universidade Estadual de Maringa and housed under standard conditions. First, pseudoestrus status was achieved in the animals, for optimal Candida colonization. For this, 0.1 mg of estradiol valerate (17 β-estradiol valerate) previously dissolved in sesame oil (Sigma-Aldrich) was injected subcutaneously in the dorsal area of each animal, followed by a massage to facilitate the dispersion of the suspension. This procedure was performed 72 h before infection and repeated once a week during the treatment period, to guarantee estrogenic vaginal conditions.
For the VVC infection, 30 μL of a suspension in PBS containing a standard inoculum of 1.0x106 CFU of C. albicans (ATCC 90028) or a clinical isolate resistant to FCZ from a woman with RVVC were inoculated by inserting the pipette tip about 5 mm deep into the vaginal lumen [26].
Treatment protocol. Each group was treated twice a day, with a 12 hour interval (except for the F16%-1x group, which was treated once a day and the control group) with 30 μL (0.123 μg of TPC expressed as gallic acid equivalents) for PRPe and 60 μL for F14% and F16% (0.266 and 0.304 μg of TPC expressed as gallic acid equivalents, respectively), for 7 and 14 days. In parallel, 16 non-infected animals were split into two groups of 8 animals each and treated with the two formulations (F14% and F16%) for 7 and 14 days to perform an in vivo toxicity evaluation.
After 7 and 14 days of treatment, the animals of each subgroup (n = 8) were euthanized (isoflurane overdose inhalation), the dissected vaginal tissue was evaluated for fungal burden (CFU) and a histological analysis in half of the animals, while the other half was evaluated by fungal burden (CFU) and scanning electron microscopy (SEM).
Fungal burden quantification. After establishing experimental vaginal candidiasis in mice and finishing the treatment protocol, the fungal burden quantification was performed. The vaginal tissues were aseptically removed, weighed and mechanically homogenized in sterile lysis buffer. Serial 10-fold dilutions of the homogenates in PBS were placed on SDA and incubated for 48 h at 35˚C to evaluate the efficacy of the treatment by a reduction in the vaginal tissue fungal burden, presented as log10 CFU/g of tissue.
Histopathological studies. The vaginal tissue removed from each animal was fixed in 4% paraformaldehyde, paraffin embedded and cut into thin sections (5 μm). The sections were stained with Grocott-Gomori's methenamine silver (GMS) to visualize fungi, and counterstained with hematoxylin and eosin (H&E) for the characterization of host cells [27]. Additionally, to assess the toxicity of the studied formulations, vaginal tissue from non-infected animals was stained with H&E. All stained slides were immediately washed, preserved with mounting medium and sealed with a coverslip. Then, they were observed and photographed using a binocular light microscope (Motic BA310), coupled to a camera (Moticam 5) and to a computer, using Motic Images Plus 2.0 software. The analyzed parameters (epithelial changes and presence of C. albicans) were descriptive and examined by a single researcher oriented by a pathologist and both were blinded to the analysis of the groups.
Cytotoxicity evaluation of mucoadhesive thermoresponsive formulations containing propolis. Two groups of four non-infected animals each were treated with mucoadhesive thermoresponsive formulations containing propolis at 14% (F14%) and 16% (F16%) to cytotoxicity test. These animals were treated for 7 and 14 days, then euthanized following the same protocol described above. The vaginal fluid from each animal was collected, centrifuged, and stored at -80˚C. The vaginal fluid from each animal was collected, centrifuged, and stored at -80˚C. A CBA Mouse Th1/Th2/Th17 Cytokine kit (BD™ Biosciences Mountain View, CA, USA) was used according to the manufacturer's instructions. Samples were tested for the following cytokines: interleukin-2 (IL-2), interleukin-4 (IL4), interleukin-6 (IL-6), interferon-γ (IFN-γ), tumor necrosis factor (TNF), interleukin-17A (IL-17A), and interleukin-10 (IL-10), and analyzed using a FACS Calibur flow cytometer and Cell Quest software (BD Biosciences, San Jose, CA, USA). In this phase, we tested cytotoxicity of MTS formulations in animal groups without infection. The results were analyzed according to a standard cytokine curve using the software FCAP Array 3.0, and the results are expressed as mean ± standard deviation. Only the detected cytokines and statistically different results are represented in the graphs.

Statistical analyses
Student's t-test was used for the comparison between the treated and untreated group and one-way analysis of variance (ANOVA) was used for multiple comparisons, applying the Bonferroni multiple-comparisons test. The data were analyzed using Prism 6.0 software (Graph-Pad, San Diego, CA, USA). Values of p<0.05 were considered statistically significant. All tests were performed at least in duplicate.

Mucoadhesive thermoresponsive system formulations: Physicochemical characterization
According to Table 1, the combination of thermoresponsive and mucoadhesive polymers P407 and C934P, respectively, showed good physicochemical characteristics.
The systems had their pH set to 7.0 using triethanolamine due to the presence of Carbopol 934P, a polymer with acidic characteristics. This pH value utilized for the best mechanical and rheological characteristics of the systems [14,17,21]. Considering that the mouse vagina has a neutral pH, while the environment of the human vagina id acidic, as well as the small amount of formulation administered, the formulation pH cannot change this physiological condition due the gradual swelling of Carbopol chains. Moreover, the neutral pH of the formulations (14% and 16%) may balance the vaginal pH, reducing the marked acidity of VVC.
When the T sol/gel of a vaginal formulation is lower than 20-25˚C, a gel can be formed at room temperature, hindering handling and administration. On the other hand, if the T sol/gel is higher than 37˚C, the liquid can remain inside the vaginal cavity that can lead to flow and loss of the formulation from the site of action (Pereira et al., 2013). In this study, the systems displayed a preliminary sol-gel transition temperature near 20˚C, indicating that they can be maintained under refrigerated temperatures, like many other medicines. However, the time spent preparing the formulation in the vaginal applicator was enough to increase the temperature, thereby reducing the risk of discomfort. Afterwards, the systems were applied before the sol-gel transition, ensuring a good distribution and contact with the vaginal mucosa.
For the T sol/gel , as shown in Fig 1, there was a decrease in temperature as the concentration of the extract increased in the mucoadhesive thermoresponsive system. Such mixtures are stored at low temperatures as aqueous dispersions, and when administered at body temperature the viscosity increases until reaching a plateau. These are stable systems and no phase separation was observed.
The polymer systems were characterized by rheology (oscillatory and continuous flow shear) to analyze their behavior at storage temperature (5˚C), room temperature (25˚C), and body temperature (37˚C).
In the continuous shear rheology, the system without PRPe (B) presented non-Newtonian flow behavior, as did the other formulations (F14% and F16%) containing PRPe. In addition, all systems displayed plastic characteristics. The formulations with different concentrations of PRPe, at 5˚C, presented a yield value where an initial stress is required for the formulation to flow [29,30]. Moreover, all formulations displayed thixotropy at all temperatures (Table 2), with a hysteresis area of 209,566.67 Pa/s and 305566.67 Pa/s for the formulations F14 and F16 at 37˚C, respectively; this was dependent on the temperature and presence of propolis extract. As the temperature increased, the thixotropy of formulation F16 increased. An increase in the amount of propolis extract increased the thixotropy of the formulations. According to Table 2, the consistency index (k) increased with an increase in the extract concentration in the systems. The flow behavior index (n) presented values below 1, which confirms that the formulations were non-Newtonian. All the systems presented yield values Each curve represents the average of at least three replicates. The standard deviation bars were omitted to clarify the graph, but all showed a coefficient of variation lower than 15%. B: mucoadhesive thermoresponsive system without propolis; F14%: mucoadhesive thermoresponsive system containing propolis at 14%; F16%: mucoadhesive thermoresponsive system containing propolis at 16%.

PLOS ONE
(τ 0 ), therefore all were plastic, and the yield value increased with an increasing concentration of the extract in the formulation and with increasing temperature. At 5˚C, the loss modulus (G") was larger than the elastic modulus (G'), i.e. the formulation was elastic-viscous. At temperatures of 25˚C and 37˚C, the formulations maintained a pattern, where G' was greater than G", and the different concentrations of the extract did not interfere. The dynamic viscosity (η') of the formulations increased as the temperature increased. For example, the formulation at 37˚C increased in viscosity, which prevented flow. In addition, η' decreased with increasing frequency. The more concentrated formulation also showed an increase in viscosity. With an increase in frequency, the loss tangent (tan δ) decreased at 5˚C, confirming that with increasing frequency the loss modulus increased in relation to the elastic modulus.
The polymeric systems presented properties that were dependent on temperature and oscillatory frequency, as well as on the presence and concentration of the extract. An increase in frequency significantly decreased the viscosity η'. For the G' and G'' moduli, the formulations at 5˚C had smaller moduli than at 25˚C and 37˚C. In addition, the G'' modulus increased with increasing frequency at 5˚C, but the G'' modulus decreased at 25˚C and 37˚C.
In general, as the extract concentration in the formulation increased, the G' and G" moduli, dynamic viscosity η', and loss tangent (tan δ) also increased.

In vitro antifungal activity of PRP extract and mucoadhesive thermoresponsive systems containing propolis on C. albicans isolated from vulvovaginal candidiasis
For antifungal susceptibility testing with commercial drugs, 88 clinical isolates from women infected with C. albicans where divided into three groups according to the clinical manifestation: colonized group, VVC group, and RVVC group. These results were interpreted from the value of the MIC and are displayed in Table 3. Most of the clinical isolates showed susceptibility to NYS (97.5% of the colonized group; 100% of the VVC group; 95% of the RVVC group) and FCZ (97.5% of the colonized group; 100% of the VVC group; 95% of the RVVC group). Otherwise, 5% of isolates showed resistance to FCZ in colonized group, as well as in the RVVC group.
Based on the fungicidal effect of PRPe on the clinical isolates of C. albicans, we evaluated whether this extract incorporated in MTS was able to kill the Candida isolates. As shown in  Fig 3, the fungicidal evaluation of the F14% and F16% formulation, for selected clinical isolates, were able significantly reduce (3log10 CFU/mL) cell viability starting from 1108.59-1267.20 μg/mL of TPC expressed in gallic acid equivalents for F14% and F16%, respectively, compared to control (p<0.05). In addition, F16% was more effective than F14% since it inhibited 50% of isolate growth at the above concentration, with a substantial reduction in cell viability (3log10 CFU/mL) at 2534.40 μg/mL of TPC expressed in gallic acid equivalents compared to control (p<0.05); this reduction was not concentration-dependent. In comparison, the images (Fig 3) illustrate a qualitative analysis of the inhibition in fungal growth at these concentrations.

Effect of the mucoadhesive thermoresponsive systems containing propolis in an experimental model of vulvovaginal candidiasis
Preclinical trials are essential for the evaluation of new therapeutic options. In this sense, we applied a robust VVC model using C. albicans with standard clinical isolates and additionally included a fluconazole-resistant clinical isolate to evaluate the efficiency of the MTS.
In agreement with the in vivo experiments, the formulations F14% and F16% showed no toxicity to the vaginal tissue. Non-infected animals and mice treated with the F14% and F16% formulations showed a vaginal epithelium without inflammatory characteristics and with keratinization similar to the control group (without infection and treatment), as shown in Fig 4. In addition, the cytokine levels were similar between the treated and untreated groups, and no evidence of inflammation was observed in the histopathological images. These acceptable safety results allowed us to develop the protocol in infected animals.
The treatment of experimental VVC due to C. albicans infection was efficient at both evaluated time points, according to Fig 5. The in vivo experiments showed a significant reduction in C. albicans (ATCC) burden (approximately 2log10 CFU/mL) in the PRPe, F14%, and F16%treated groups, with an NYS-like response, after 7 days of treatment in experimental VVC ( Fig  5A) compared to the control group (C1), (p<0.05). With an improved response, treatment for 14 days (Fig 5B) further reduced the fungal burden in animals treated with F16%, i.e. greater than 2 log10 CFU/mL compared to the control group (C1) (p<0.05). Moreover, it was observed that F14% was able to maintain antifungal activity until the last analyzed day, unlike nystatin with a small increase. In addition, we tested F16% with once daily dosing. Our data show that treating 1x per day was as efficient as twice per day, with no statistical significance between them (Fig 5C).
There was no statistically significant difference between the control groups (C1 and C2), as C2 animals were treated only with MTF without PRP (p<0.05) (Fig 5A and 5B).
Based on the good results with F16%, we tested this formulation on clinical isolates of C. albicans. Curiously, Fig 6 shows that the F16% was also efficient in the treatment of experimental VVC with a fluconazole-resistant clinical isolate, leading to a reduction in fungal burden of

PLOS ONE
approximately 2 log10 CFU/mL, at both analyzed time points. It must be considered that this result is better than that found for NYS. Even with satisfactory in vitro results for this antifungal, in vivo there was an apparent resistance to killing C. albicans, especially when compared to treatment with F16%.
Estrogen promotes modifications in the vaginal epithelium that are typical of women childbearing age and allow the infection, adhesion, and survival of the pathogen in the genital tract. S1 Fig in S1 Data illustrates the histological and SEM consequences of pseudoestrus status in the vaginal epithelium, where the organization and keratinization evaluated after 7 days of follow-up (S1B and S1D Fig in S1 Data) were maintained for up to 14 days, the final time point at which our animals were followed and treated, with similar quality (S1C Fig in S1 Data).   Fig 7B shows an image from the control group with penetration into the epithelium 7 days after infection, associated with edema and epithelial disorganization. These epithelial changes are directly related to the presence of pathogens. On the other hand, the reorganization of the epithelium and absence of fungi are evident in the treated groups (Fig 7C and 7C').
Furthermore, Fig 8 strengthens these results as a qualitative analysis of SEM images from vaginal epithelium, showing an expressive reduction in the fungal burden in treated groups compared to control, demonstrating NYS-like behavior.

Discussion
VVC is one of the most frequent vaginal infections, just behind bacterial vaginosis, and has a direct impact on quality of life in most symptomatic cases. C. albicans has been reported as the species most commonly involved in genital fungal infections [31]; the first-line treatment includes the use of topical or oral azoles and polyene agents [32]. Nevertheless, VVC treatment has proven to be a challenge in clinical practice, given the predisposing host factors (e.g. hormonal contraceptive and antibiotics use, pregnancy), the limited number of drugs available and the increasing antifungal resistance of pathogens driven by the indiscriminate use of antifungals [33]. In this context, the development of new therapies for VVC treatment should be encouraged in order to achieve greater pharmacological tolerance and low cost options. Thus, this study presented a therapeutic option for the treatment of VVC by means of a preclinical study with better results than the reference drug tested.
The majority of publications on new VVC therapies are based on in vitro results. Dalben-Dota et al [24] evaluated the in vitro antifungal activity of propolis extract against yeasts C. albicans and Candida non-C. albicans with a good response compared to nystatin; in the same way, the present study also began with in vitro screening and showed that PRPe had excellent antifungal activity on clinical isolates of VVC (Fig 3). Fig 6 shows that PRPe was also effective against a fluconazole-resistant clinical isolate.
We next sought to incorporate PRPe into an available pharmaceutical system for VVC treatment. For this, our group had already invested efforts in the technological development and characterization of an MTS-PRP for various uses [14,17,20,29,30,34]. The first step was to evaluate the antifungal activity of the formulations at different concentrations of propolis using in vitro assays. The results show that PRPe maintained its antifungal efficacy even when incorporated into the MTS, both for the standard strain and for clinical isolates of VVC (Figs 2  and 3).

PLOS ONE
Propolis has been studied and explored, since ancient times, in its various biological functions as an antifungal [14,16,24,34,35], antibacterial [36], anti-inflammatory [37], antineoplastic [38], immunomodulatory [39,40], tissue repair promoter [41,42], and antioxidant [43]. These results are attributed to the complex constitution of this extract that acts synergistically, highlighting the phenolic compounds as the main constituents with biological activity. In this context, this compound is attractive for VVC treatment, since it is clinically characterized as an infectious disease associated with an important inflammatory process that results in clinical manifestations (discharge, pruritus, burning, dyspareunia, dysuria) presented by women affected by VVC.
Investing in a pharmaceutical formulation such as MTS-PRPe is very important and useful, since it complements the use of this extract, considering its properties: viscoelastic formulation, resistance to deformation, good retention, thermoresponsivity, prolonged release for more than 24 hours, and easy administration and dispersion in the vaginal mucosa [17,23].
The systems had their pH set to 7.0 using triethanolamine due to the presence of Carbopol 934P, a polymer with acidic characteristics. This pH value utilized for the best mechanical and rheological characteristics of the systems [14,17,21]. Considering that the mouse vagina has a neutral pH, while the environment of the human vagina id acidic, as well as the small amount of formulation administered, the formulation pH cannot change this physiological condition due the gradual swelling of Carbopol chains. Moreover, the neutral pH of the formulations (14% and 16%) may balance the vaginal pH, reducing the marked acidity of VVC.
When the T sol/gel of a vaginal formulation is lower than 20-25˚C, a gel can be formed at room temperature, hindering handling and administration. On the other hand, if the T sol/gel is higher than 37˚C, the liquid can remain inside the vaginal cavity that can lead to flow and loss of the formulation from the site of action [17]. In this study, the systems displayed a preliminary sol-gel transition temperature near 20˚C, indicating that they can be maintained under refrigerated temperatures, like many other medicines. However, the time spent preparing the formulation in the vaginal applicator was enough to increase the temperature, thereby reducing the risk of discomfort. Afterwards, the systems were applied before the sol-gel transition, ensuring a good distribution and contact with the vaginal mucosa.
Knowing the in vitro profile of this formulation and considering the low toxicity of this extract [44,45]; the PRPe cytotoxic concentration was higher than the therapeutic concentration in this study. Thus, the use of this compound was considered safe for the next step of this study, which was to assess the in vivo antifungal activity of MTS-PRPe in a pre-clinical model of VVC. At this step, the results were even more promising, since the formulations tested at different concentrations of propolis (F14% and F16%) had expressive antifungal activity at both evaluated time points. Also, we noted that administration of the F16% formulation showed similar results in terms of decreased fungal load when treatment was given once or twice daily (Fig 5C). Finally, the anti-fungal action of the propolis formulations was well illustrated and exemplified in Figs 7 and 8. Moreover, according to Fig 7, a reduced inflammatory response after treatment with propolis could be inferred with the reorganization and tissue recovery of the vaginal mucosa. Therefore, this type of therapy is extremely interesting as it establishes ease of administration for future clinical studies. Moreover, it is important to note that the results obtained with the propolis formulations were similar or even superior to those with nystatin, a commercial reference treatment used as first line in VVC therapy [32].
These results are encouraging since they make possible the application of formulations in the clinical phase, considering the properties of adhesion and persistence of the compound in the vaginal mucosa, as well as the possibility of one application daily, meeting the interests of patients and facilitating adherence to treatment.

Conclusion
Our results support the use of PRPe as a promising option treatment for vaginal infections caused by C. albicans, as well as the MTS, especially because of its convenient physicochemical properties which makes its use easier and more comfortable, with better results than nystatin. Therefore, these results motivate us to invest efforts in the next steps of new antifungal agent development through clinical trials.