Tragacanth and xanthan gum were purchased from Gol Darou Co, Tehran, Iran. Clotrimazole was obtained from Behvazan pharmaceutical Co, Rasht, Iran. Propylene glycol (PG), myrj52 and sodium benzoate were procured from Sigma-aldrich, Germany. Dialyzes bags with a molecular weight cutoff of 14kDa were purchased form Sigma (Steinheim, Germany). All gel formulations were prepared in deionized water.

Preparation and characterization of the mucoadhesive gel

According to the study design (Table 1) proper amount of each polymer was mixed with deionized water in a 250 mL beaker. The colloidal solution was stirred with magnetic stirrer at 500 rpm. Clotrimazole and myrj 52 and other excipients were dissolved in determined volume of deionized water, separately. Finally, clotrimazole and polymeric colloidal solutions were mixed and adjusted to 100 mL.

Table 1. Natural polymers and other excipient composition in formulations F1-F8.

- Visual examinations: formulations F1-F8 were visually evaluated among color, clarity, homogeneity, presence of particles or polymer clamps.

- Viscosity studies: viscosity of samples was measured using a DV-3 con viscometer (Brookfield, USA). 50 ml of samples were applied to viscometer container in 37°C (n=3) (Pandey et al. 2020).

- Gel spreadability: 0.5 g of each gel formulations was placed on a circle glass plate with 1cm diameter and another glass plate was placed on the gel. A 50 g weight was placed on upper glass for 5 minutes. The spreading area was calculated using the measurement of increase in gel diameter (n=3) (Deuschle, Deuschle, Bortoluzzi, & Athayde, 2015; Dantas et al., 2016).

- FTIR Spectroscopy: A FTIR spectrophotometer (Thermo Nicolet, model: Nexus670, USA) was used to examine the probable incompatibilities between clotrimazole and excipients. Clotrimazole, TG, XG and Formulations F1 (as the combination of API and excipients) where evaluated. FTIR Spectra were collected at a resolution of 4 cm−1and given as the ratio of 21 single beam scans to the same number of background scans in pure KBr (Fidalgo & Ilharco, 2001).

- Drug content: 1 g of gel was vigorously stirred with 100 ml citrate buffer, using sonicator and vortex resulting in a transparent solution after filtration. Then the filtrate was subjected to UV spectrometry in 230 nm (n=3) (Kumar & Verma, 2010).

- Scanning electron microscopy analysis: Polymer chain morphology was investigated by FESEM Sigma VP (Zeiss, Germany) with an accelerating voltage of 10 kV under vacuum conditions. For sample preparation, gels were air dried at 25°C in a desiccator and dried gels were gold sputter coated before FESEM (Schemehorn, González-Cabezas, & Joiner, 2004).

Mucoadhesion study was performed according to Tasdighi’s method (Tasdighi, Azar, & Mortazavi, 2012) with some modifications. 0.5 g of each gel formulations (A) was placed between two circle glasses (B) covered with sheep intestinal mucus (C). Bottom glass was fixed in a crystallizer and top glass was linked to a balance measuring the required force for detachment of the gel from mucosal membrane. The test was performed in phosphate buffer medium pH=4.5 and 37°C (n=3).

- In vitro release profile: The in vitro release was performed using 14 kDa Dialysis tubing Cellulose Membrane. Membrane was soaked in distilled water for 24hrs, before experiment. 5 g of each formulation was packed in dialysis tube and placed in 200 mL citrate buffer (pH=5) and ethanol (70:30) as receptor medium, in 37°C. Medium was stirred at 100 rpm during the release test and samples were withdrawn at certain time intervals of 0.5, 1, 2, 4, and 6hrs. Content of clotrimazole in each sample was analyzed with UV spectrophotometry (PerkinElmer, USA) in 230 nm (Kumar & Verma, 2010).

- Drug release kinetic: The in vitro release data was incorporated to investigate the release kinetics of formulations F1, F6, F7 and F8 using various mathematical kinetic models. Release kinetic was evaluated in formulations F1, F6, F7 and F8 due to higher variation in their polymeric proportion (referring to table 1), supposing probable difference in release mechanisms. Evaluated models included Zero order, First order, Higuchi and Korsmeyer-Peppas which are previously explained in (Ngwuluka, Kyari, Taplong, & Uwaezuoke, 2012)

For comparison between 8 formulations, data obtained from physico-chemical evaluations (gel spreadability, drug content and mucoadhesion) was subjected to one way ANOVA. In cases in which significant differences existed in the ANOVA test, Tukey post-hoc test was used to specify those samples having significant differences with each other. In all tests, p-value ≤0.05 was considered as significant.

- Visual examination: revealed that all formulations where homogenous in texture, with a creamy color which was due to presence of dispersed active pharmaceutical ingredient (API) particles.

- Viscosity: also influences the gel spreadability, drug release rate and mucoadhesion. Viscosity of all formulations are listed in Table 2. Results showed that from F1 to F6 viscosity gradually decreases due to decrease in total polymer concentration (from 4 to 1.5%) in which the TG percent is fixed (0.5%) and XG percent reduces. F7 and F8 showed higher viscosity as a result of higher polymer concentration (5%) and as other studies investigated, F7 with the highest viscosity confirmed that XG forms more viscous gel in comparison with TG in equal polymer percent (Mohammadian & Alavi 2016).

- Spreadability results: showing an exact reverse behavior with viscosity are presented in Table 2. Statistics showed the significant difference between formulations, except F1 and F2 in which F1 showed lower spreadability (Tukey Post Hoc) but the difference was not significant.

Table 2. Physical properties of formulations F1-F8.

*Due to very high viscosity, it was not measurable by the apparatus. **Cumulative drug release

- FTIR spectra: for pure clotrimazole, formulations F1 and F6 are presented in Figure 1 (a, b, c respectively) and positions of peaks are compared. Aromatic C-H bending (758 cm-1), C=N stretch (1488 cm-1), and aromatic C=C stretch (1534 cm-1) and aromatic C-H stretch (3167 cm–1) which are index peaks corresponding to clotrimazole are present in F1 and F6 final formulations spectrum. The FTIR spectra of pure API and final formulations showed that the position of characteristic peaks where not altered after incorporation in formulations confirming the absence of probable interactions between drug and excipients. Similarly, in other studies no significant interaction was found between clotrimazole and chitosan (Grimling,Karolewicz, Nawrot, Włodarczyk, & Górniak, 2020), hydroxypropylmethylcellulose, sodium carboxymethylcellulose and Carbopol (Gupta, Natasha, Getyala, & Bhat, 2013).

- Drug content: in an acceptable range of API, ensures the producer and consumer about receiving the adjusted or necessary dose. Since there is no defined monograph for vaginal clotrimazole gel in pharmacopeias, the prevalent range of 90-105% was considered as acceptable in this study. Drug content of each formulation is presented in Table 2. Statistical analysis for F1 to F8 revealed that there is significant difference between all formulations.

Figure 1. FTIR spectroscopy of a) pure clotrimazole, b) Formulation F1, c) formulation F6 with common peaks of 758, 1488, 1534 cm-1.

- Scanning Electron Microscopy (SEM): characterized the polymeric chain micro-morphology of hydrogel. Samples F7 and F8 which were simply made of XG and TG, respectively and F1 with highest amount of XG (3.5%) and 0.5% TG and F6 with lowest amount of XG (1%) and 0.5% TG, were subjected to SEM. Figure 2 shows a homogenous linear morphology of chains surrounding API for F7 (Figure 2a) and a relatively smoother morphology for F8 (Figure 2b) while in F1 and F6 a kind of polymeric chain integration is observed (Figure 2 c and d). It seems that formulations containing both polymers show special polymeric interactions due to probable hydrogen or van der Waals bonds. SEM images of prior investigations also showed higher polymeric entanglements while using two or more polymers in one formulations for example cationic tapioca starch–xanthan gum mixture in comparison with cationic tapioca starch (Chaisawang & Suphantharika, 2005) or incorporation of anionic tapioca starch-xanthan gum mixture compared to anionic tapioca starch which presents higher polymeric interactions in SEM images (Chaisawang & Suphantharika, 2006). As clotrimazole is a highly lipophile API with log P above 5, it’s dispersed in hydrogel and crystals are observed in all formulation’s SEM results.

Mucoadhesion studies showed that similar to viscosity, increase in total polymer concentration, increases the mucoadhesion and there was a statistically significant difference between all formulation’s viscosity. This finding was confirmed by previous studies such as addition of XG to TG/chitosan polyelectrolyte complexes-based hydrogels which resulted in formation of a more viscous hydrogel with improved mucoadhesiveness and mechanical strength for buccal application (Potaś, Szymańska, Basa, Hafner, & Winnicka, 2021). In another study, XG was incorporated in formulation of mucoadhesive buccal patches of zolmitriptan. Results revealed that with an increase in concentration of XG, bioadhesion force increased (Shiledar, Tagalpallewar, & Kokare, 2014). Therefore, mucoadhesion decreased from F1 to F6 (77.71-54.49 dynecm-2). F7 and F8 with highest polymer concentration (5%) showed higher mucoadhesion, noting that TG revealed higher adhesion in comparison with XG in equal polymer concentration as was observed previously in (Parvinroo, Eslami, Ebrahimi-Najafabadi, & Hesari, 2020) (Table 2).

Figure 2. SEM image of hydrogel polymeric chain morphology in a) F7: XG (5%). b) F8: TG (5%). c) F1: XG (3.5%) and TG (0.5%). d) F6: XG (1%) and TG (0.5%).

- In vitro release profile: of clotrimazole was evaluated for all formulations in citrate buffer (pH=5) and ethanol (70:30) in 37°C. F1 to F8 release profiles are presented in Figure 3. Based on cumulative release percent, F6 showed the highest (95.29%), while F1 showed the lowest (63.13%) release rate in 6 hours. Since, F1 to F6 gels are composed of two polymers, decrease in total amount of polymers leads in increasing the release rate from F1 to F6. On the other hand, comparing the release rate of F7 and F8 reveals that TG has a more retardant effect on drug release than XG as reported by Salamanca CH (Salamanca et al., 2018). However, total polymer concentration in F7 and F8 is higher than F1 to F6, their drug release rate is higher than F1, F2 and F3. It maybe hypothesized that these two polymers show some synergistic effect in drug release rate control, when combined. Some synergistic effects were observed in XG in previous investigations including gelation with Locust Bean Gum or Konjac Glucomannan (Goycoolea,Richardson, Morris, & Gidley, 1995), viscosity with guar gum (Casas, Mohedano, & García‐Ochoa 2000), mechanical and barrier property with gellan gum (Zhang et al., 2020), viscosity with konjac-Mannan and TG (Mirzaei, Alimi, Shokoohi, & Golchoobi, 2018).

- Drug release kinetic: in formulations F1, F6, F7 and F8 showed the highest regression coefficient in with higuchi model (Table 3). Higuchi model describes the diffusion of API from the matrix and drug delivery system's (DDS) matrix erosion that has been observed repeatedly in XG based DDS (Kar, Mohapatra, Bhanja, Das, & Barik, 2010; Mughal, Iqbal, & Neau 2011; Zambrano-Zaragoza, Quintanar-Guerrero, Del Real, Piñon-Segundo, & Zambrano-Zaragoza, 2017). In this regard, main polymeric component of formulations F1, F6& F7 is XG which strongly navigates the release behavior toward higuchi model. However, F8 which is totally composed of TG, also was best fitted with higuchi model and describes the release of drug based on Fickian diffusion from insoluble matrix as a square root of time-dependent process which was also discovered in conjugated TG (Dehghan-Niri, Tavakol, Vasheghani-Farahani, & Ganji, 2015; Shafiee, Ahangar, & Saffar, 2019) or combination of TG and XG (Akhtar Rasul et al., 2010).

Figure 3. Cumulative drug release of formulations F1-F8 vs time.

Table 3. Release kinetic parameters for F1, F6, F7 and F8 formulations based on mentioned mathematical models.

z.hesari@gmail.com