Design, Development, and Optimization of Sterculia Gum-Based Tablet Coated with Chitosan/Eudragit RLPO Mixed Blend Polymers for Possible Colonic Drug Delivery

The purpose of this study is to explore the possible applicability of Sterculia urens gum as a novel carrier for colonic delivery system of a sparingly soluble drug, azathioprine. The study involves designing a microflora triggered colon-targeted drug delivery system (MCDDS) which consists of a central polysaccharide core and is coated to different film thicknesses with blends of chitosan/Eudragit RLPO, and is overcoated with Eudragit L00 to provide acid and intestinal resistance. The microflora degradation property of gum was investigated in rat caecal medium. Drug release study in simulated colonic fluid revealed that swelling force of the gum could concurrently drive the drug out of the polysaccharide core due to the rupture of the chitosan/Eudargit coating in microflora-activated environment. Chitosan in the mixed film coat was found to be degraded by enzymatic action of the microflora in the colon. Release kinetic data revealed that the optimized MCDDS was fitted well into first-order model, and apparent lag time was found to be 6 hours, followed by Higuchi release kinetics. In vivo study in rabbits shows delayed T max, prolonged absorption time, decreased C max, and absorption rate constant (Ka), indicating a reduced systemic toxicity of the drug as compared to other dosage forms.


Introduction
In the recent times, colon-speci�c technologies have utilized single or combination of the following primary approaches, with varying degrees of success: (1) pH-dependent systems, (2) time-dependent systems, (3) prodrugs, and (4) colonic micro�ora-activated systems [1,2]. Among the different approaches to achieve colon speci�c drug delivery system, the use of polymers speci�cally degraded by colonic bacterial enzymes (such as -glucoronidase, -xylosidase,galactosidase, and azoreductase) holds promise. Microbially activated delivery systems for colon targeting are being developed to exploit the potential of the speci�c nature of diverse and luxuriant microbiota associated with the colon compared to other parts of the gastrointestinal (GI) tract. ese colonic microbiotas produce a large number of hydrolytic and reductive enzymes which can potentially be utilized for colonic delivery [1,2]. Most of these systems are based on the fact that anaerobic bacteria in the colon are able to recognize the various substrates and degrade them with their enzymes. Natural gums are oen preferred to synthetic materials due to their low-toxicity, low-cost, and easy availability. A number of colon-targeted delivery systems based both on combination of pH, polysaccharides and biodegradable polymers have been designed and developed by various research groups for successful delivery of drugs to the colonic region [3,4]. Sterculia gum has not yet been used as drug carrier speci�cally to the colon. It is insoluble in water, hydrates quickly, and swells into a homogenious hydrogel consistency or mass which pose difficulty for its use as polysaccharide coat [5]. But, it seemed to be an interesting polymer for the preparation of hydrophilic matrix tablets [6,7]. However, sterculia gum in the form of hydrophilic matrix cannot protect the drug from being released in stomach and small intestine. Besides, sterculia gum is expected to retard drug release due to its higher swelling index, and at the same time its degradation by the colonic micro�ora would make it ideal to deliver drugs in the colon. e property of higher swelling index would provide greater surface area for more bacterial enzymatic attack. is property of the gum could be used to produce hydrostatic pressure in the design of micro�ora triggered colon targeted drug delivery system (MCDDS). In this system, the hydrostatic force is produced by osmotic agents and polymer swelling which concurrently drives the drug out of the system through the pores created by the pore-forming agent in the inner coating aer exposure of the system to the colonic �uid [8,9]. In addition, eudragit RLPO polymer has been reported to increase the permeability to colonic �uid due to the presence of higher number of quaternary ammonium groups.
Hence, the objectives of present investigation was to design MCDDS based on swelling property of sterculia gum and to study the in�uence of different independent variables on dependent variables. e design of MCDDS comprises of an osmotic tablet core containing model drug azathioprine (AZA), sterculia gum as binder, and other excipients; an inner semipermeable coating which is over coated with enteric layer to provide acid and intestinal resistance. e study includes the optimization of chitosan/eudragit RLPO mixed �lm coating for colonic delivery of polysaccharide core and to investigate the effects of the polymer blend ratio, concentration of pore former in the coat and coating thickness on the resulting drug release and to propose the drug release mechanism of the system. e innermost layer of chitosan/eudragit RLPO provides desired intestinal resistance, but controlling drug release in the colon [10]. Eudragit L100 was deposited in order to protect the delivery system from the gastric acidic conditions. A multilayered approach was selected, since such a dosage form was less likely to undergo dose dumping, and also, it may facilitate the spreading of the drug over the in�amed regions of the colonic lumen. e feasibility of the novel MCDDS was studied using AZA as a model anti-in�ammatory drug via in vitro evaluation of drug release characteristics and in vivo assessment of pharmacokinetics in rabbits [11,12].  [13,14]. Caecal contents were collected from male Wistar rats weighing 250-300 g each. e caecal contents were dispersed in PBS under anaerobic environment (bubbled with CO 2 gas), and the concentration of the caecal contents was adjusted to 4.0, 8.0 and 12.0% (w/v) in the PBS. Finely grounded sterculia gum powder 100 mg was added into 10 mL of caecal PBS and incubated at 37 ∘ C under anaerobic condition. e pH of caecal PBS was measured at 2 h interval up to 8 h using a pH meter.

Preparation of Swellable Core
Tablets. e detail composition of the core tablet is presented in Table 1. e core tablets of AZA having an average weight of 240 ± 5 mg were prepared by direct compression using a single stroke tablet punching machine �tted with 8 mm round standard concave punches. Sterculia gum was used as binder cum hydrophilic matrix former, anhydrous lactose as diluents, citric acid as pH regulating excipient, and magnesium stearate as lubricant [15,16].

Preparation of Chitosan-Eudragit RLPO Coating Dispersions.
In the initial trial, a coating solution of eudragit RLPO (10% w/v) in propan-2-ol: acetone (60 : 40) containing 15% w/w or 25% w/w concentration of chitosan was used to apply a semipermeable coat on the core tablet. PEG 400 (25% of total coating materials) was added to improve the physicomechanical property of eudragit RLPO �lm. e coating conditions were as follows: stainless steel pan, 200 mm diameter, four baffled, rate of rotation of the coating pan; 40 rpm, nozzle diameter of spray gun; 1 mm, spray rate; 5 mL/min, spray pressure; 2 Bar, drying temperature; 40 ∘ C [17]. Aer coating, the tablets were dried for 8 hours at 35-40 ∘ C in order to remove the residual solvent.

Experimental Design for Coating Formulations.
A full 3 2 factorial design was used for optimization of coating solutions [18,19]. e concentration of chitosan was selected by using central composite design (CCD) under design expert soware (version 8.0). e studied factors (independent variables) were concentration of pore former, chitosan ( 1 ), and weight gain in coating thickness, eudragit RLPO ( 2 ). e dependent variables selected for the study include lag time for drug release up to 2% in SCF ( 1 ) and percent drug release in 12 hours ( 2 ) and 18 hours ( 3 ).

Physical Evaluation of the Coated Tablets under Factorial
Design. e thickness, hardness, drug content uniformity and weight uniformity were determined in a similar manner as stated for conventional oral tablets in the accredited pharmacopoeia.  [20,21]. -glucosidase was added to degrade chitosan, in colonic environmental conditions. Aliquots of dissolution �uid were analyzed at speci�ed time intervals to determine the release of AZA by UV-visible spectrophotometer at wavelength of 281 nm.

Statistical Analysis of Data and Coating Optimization.
e response values (lag time in hour, % drug release in 12 hour and 18 hour resp.) of coated tablets based on 3 2 factorial design were subjected to analysis by response surface reduced quadratic model with the help of Design Expert soware (Version 8.0). Statistical validity of the polynomial was established on the basis of ANOVA provision in the design expert soware, and signi�cant terms ( ) were chosen for �nal equations [18,22]. Response surface plots and 3D contour plots were constructed using the output �les generated.

Enteric Coating of Chitosan-Eudragit RLPO Coated
Tablets with Eudragit L100. Eudragit L100, which dissolves above pH 6.0, was selected for enteric coating [23]. e optimized chitosan-eudragit RLPO coated tablets were further over coated with enteric coating using 10% w/v of eudragit L100 in 95% ethanol. e total weight gain of eudragit L100 coating was 10% w/w. Eudragit L100 was dissolved in 95% ethanol under high stirring condition until a clear solution was obtained. Triethyl citrate (TEC), 10% w/w of total dry polymer was added as plasticizer and talc (1.5% w/w of dry polymer) as a glidant. e coating conditions were same employed under semipermeable coating.

Kinetic Evaluation of Drug Release Data and Stability
Studies. Dissolution data of the optimized formulation was �tted to various mathematical models in order to describe the mechanism of drug release [24,25]. e corelation coefficient ( 2 ) was taken as the criteria for choosing the most appropriate model. e selected formulations were tested for a period of 8 weeks at different storage conditions of 25 ∘ C and 40 ∘ C with 60% RH and 75% RH, to evaluate their drug content, hardness, and in vitro dissolution rate [26].

HPLC Assay.
In the present method, the plasma 6mercaptopurine (6-MP) rather than AZA concentration was measured because aer oral administration AZA is quickly converted into its active metabolite 6-MP. e 6-MP concentration in plasma was determined according to the HPLC method reported by Shao-Jun et al. [27]. e HPLC system consisted of a Rheodyne Isocratic pump (Model-LC-10, Shimadzu Corp., Kyoto, Japan) a model 2250 pump (Bischoff, Germany), and a UV detector (Model-SPD, Shimadzu Corp., Kyoto, Japan) set at a wavelength of 325 nm ( max ). e samples were chromatographed on a reverse phase Hypersil ODS C18 column (5 m, 2 cm × 4 6 mm i.d., ermo Electron Company, Bellefonte, North America) protected with a guard column (4 × 4 mm) packed with the same material. e mobile phase was consisting of 80 parts of 0.01 M KH 2 PO 4 and 20 parts of Acetonitrile (80 : 20, v/v, pH 4.5). It was pumped at a �ow rate of 1 mL/min for the run time of 10 min under the experimental conditions with an injection volume of 20 L [28,29]. e column was thermostated at an ambient temperature 30 ∘ ± 2 ∘ C throughout the study. e effluent was monitored with the UV-Visible detector at 325 nm. Metronidazole was used as an internal standard (IS).

In Vivo Study in
Rabbits. e pharmacokinetics of marketed tablet (MKT), enteric coated tablet (EC), and MCDDS of AZA were assessed and compared in rabbits in a randomized, two-period crossover study. e washout period between administrations was one week. Six rabbits each weighing from 1.5 to 2.0 kg were used in this study. e rabbits were fed standard laboratory chew diet with water and fasted overnight before the experiments. e animals used in the experiments received care in compliance with the "Principles of Laboratory Animal Care" and "Guide for the Care and Use of Laboratory Animals. " Experiments followed an approved protocol from Department of Pharmaceutical Sciences, Dibrugarh University Institutional Animal Ethical Committee.
e MKT, EC, and MCDDS (containing 50 mg/Kg of drug) were orally administered in rabbits. At time intervals, two milliliters of blood samples were collected from marginal ear vein into heparinized tubes and centrifuged at 5000 rpm for 15 min at 4 ∘ C to separate plasma. e plasma samples, 0.2 mL, were deproteinized with 2.0 mL of methanol and acetonitrile mixture (1 : 1, v/v), vortexed for 5 min, centrifuged at 6000 rpm for 15 min, and supernatants were collected. e supernatants were evaporated to dryness under a gentle nitrogen stream at 40 ∘ C. e residues were reconstituted in 200 L of mobile phase, and then 20 L of each solution was injected into the HPLC column for analysis of the drug in vivo. Blood sampling time points were 0, 1, 2, 3, 4, 5, 6,7,8,9,10,12,14,16,18,20,22 samples was determined using a validated HPLC procedure as described by Shao-Jun et al. [27].

Determination of Pharmacokinetic Parameters and Data
Analysis. Pharmacokinetic parameters were calculated by noncompartment analysis based on statistical moment theory using Microso Excel soware. e pharmacokinetic parameters, such as maximum plasma concentration ( max ) and time of maximum concentration ( max ), were obtained directly from the plasma concentration-time plots. e area under the plasma concentration-time curve up to the last time (t) (AUC 0− ), area under curve extrapolated to in�nity (AUC 0−∞ ) and area under the �rst moment curve extrapolated to in�nity (AUMC 0−∞ ) were calculated using the linear trapezoidal rule. e mean residence time (MRT) was calculated as AUMC/AUC. Results were expressed as mean ± standard deviation. Variations in pharmacokinetic parameters were tested using analysis of variance (ANOVA). In all the cases, a value of 0 0 was considered statistically signi�cant.

�icro�ora Degradation �tudies of �terculia Gum.
Micro�ora degradation studies of sterculia gum revealed that the pH of caecal-PBS was decreased markedly from pH 7.4 to 5.0 aer incubation for 2 h with sterculia gum. e rate of decrease of pH was depended on the concentration of caecal contents within the 8 h of incubation (Figure 1). e decrease in pH was due to the appearance of degradation products of sterculia gum such as organic acids by the bacterial enzyme present in rat caecal contents.

Formulation
Aspects of Core Tablets. e weight of each tablet was determined to be within the range of 240 ± mg in order to maintain the relatively constant volume and surface area. e core tablet (240 mg each) was prepared at average tensile strength of 4.0 Kg/cm 2 and average diameter of 8 mm and thickness 4 mm. e incorporation of citric acid in the core composition increased the hydration of large amount of the gum and expanded its volume to great extent.

Evaluation of the Chitosan/Eudragit Coated Tablets.
e weight variation was in the range of 27 ± 2 09 to 287 ± 1 98 mg and friability was less than 0.5%. Uniformity in drug content was found among different batches of the tablet, and the drug content was more than 95%.

3.�. �n�uence of Coating Formulation �ariables on Drug
Release. e core tablet was successfully coated by conventional pan coating technique with varying proportion of chitosan-eudragit RLPO provided by central composite design. e coating composition of the various formulations under 3 2 factorial designs are presented in Table 2. e results of the in vitro dissolutions studies of different batches of coated tablets indicated that increase in concentration of chitosan from 15% to 25% w/w and keeping constant weight gain in thickness of polymers at 10% w/w, the lag time (the time required for drug release up to 2% in SCF) was signi�cantly decreased from 0.�0 h to 0.25 h (FC1 FC4 FC7). e lag time was determined by separately running dissolution studies of chitosan/eudragit coated tablets in SCF for one hour at minimum time intervals. e amount of chitosan present in the eudragit coat was the key factor for such lag time. Lower amount of chitosan shows longer lag time, and higher amount shows shorter lag time.

Effects of Concentration of Chitosan on Drug Release.
To study the effect of concentration of chitosan, its concentration in the coating solution was kept at 15% w/w for the batch FC1, 20% w/w for FC4, and 25% for FC7. e result of the in vitro release pro�le from these formulations is shown in Figure 2. It is observed that concentration of chitosan has direct effect on drug release. e formulation FC7 containing highest concentration (25% w/w) of chitosan in the coating composition released more than 90% of AZA aer 18 h of the dissolution study. is might be due to the reason that an increased in the amount of chitosan (FC7 > FC4 > FC1), it became more susceptible to bacterial attack creating pores immediately resulting in shorter lag time (0.15 h) for drug release.

Effect of % Weight Gain in Coating ickness.
It was observed that increased in the level of weight gain from 10%, 12%, and 14% in the batches of FC1, FC2, and FC3 and keeping the concentration of chitosan constant at 15% w/w made chitosan particles less susceptible to bacterial attack, resulting in longer lag time and lesser percentage of drug released in 18 h owing to less accessibility of the chitosan particles across the eudragit coat by the colonic bacteria. Figure 3 shows that as the coating thickness was increased, drug release was decreased, as evidenced by the difference factor f 1 value which was lower than 15. For the calculation of f 1 and f 2 (similarity factor) values, only one data point at which more than 85% of the drug release had been released was taken into consideration. Drug release decreased. Coating materials (% w/w) Formulation code  FC1  FC2  FC3  FC4  FC5  FC6  FC7  FC8  FC9  Micronized chitosan  15  15  15  20  20  20  25  25  25  Eudragit RLPO  60  60  60  55  55  55  50  50  50  PEG 400  25  25  25  25  25  25  25  25  25  Total weight  100  100  100  100  100  100  100  100

Statistical Analysis of Dissolution
Data. ANOVA of the dependent variables indicated that the assumed regression models were signi�cant ( < ) and valid for each considered response (Table 3). e response values of the coated tablets based on factorial design generated a mathematical model, which indicated that both the level of pore former and coating thickness had signi�cant in�uence on percentage of drug release in the simulated colonic �uid at p� 7.4. e equations of the responses were found to be as follows:  (1) to obtain the predicted values of , 2 , and 3 . e observed and predicted values for the 2 response were found to be in good agreement (Table 4). e three-dimensional response surfaces plots were drawn to estimate the effects of the independent variables on each considered response (Figure 4).

Optimization of Chitosan-Eudragit RLPO
Coating. e best colonic drug delivery system based on coating with microporous eudragit RLPO containing optimum amount of chitosan would be a system that could protect drug release in the higher parts of the small intestine and deliver the drug only at the colonic region. Chitosan particles in the RLPO coat remained undigested in the intestinal �uid due to absence of bacterial enzyme, but degraded in the colonic �uid due to the presence of vast anaerobic bacteria and allowed the drug release to occur. erefore, the concentration of chitosan in the eudragit coat could be the key factor for lag time. e lag time was inversely related to the level of chitosan in the eudragit coat. e lag time in colonic environment (pH 7.4) was considered as response 1 and optimum duration for the response was considered to be 30 minutes. During this lag time, the chitosan in the eudragit coat comes in contact with the colonic bacteria formed in situ delivery pores for release of the drug. us, the percent of drug release in 12 h and 18 h was considered as response 2 and 3 with a constraint of minimum of 40% and 80% release, respectively. A suitable formulation which could meet these target responses would be able to release the maximum amount of drug in the colon despite its 2 h lag time in simulated gastric �uid (S�F, 0.1 M HCl at pH 1.2 containing 3.2 mg/mL pepsin) and 4 h lag time in simulated intestinal medium (SIF, phosphate buffer media at pH 6.8 containing 5 mg/mL pancreatin). e best formulation showing drug release corresponded to 18.96% of chitosan (pore former) and 11.3% of coating thickness of eudragit RLPO �lm provided the desired release as shown in Figure 5. e above quantity ( 1 and 2 ) of formulation was substituted in (1) to obtain the predicted responses. e validity of the optimization procedure was con�rmed by preparing a new batch of coating formulation with the concentration provided by the soware and the observed response were found to be inside the constraints and close to the predicted responses. us, the factorial design was valid for predicting the optimum formulation.
Results of in vitro dissolution study showed that the over coating with 10% w/w of enteric coating material (eudragit L100, dissolves above pH 6.0) provided the desired acid and intestinal resistance of the optimized chitosan-eudragit RLPO coated tablet. Figure 6 shows the in vitro release pro�le of optimized MCDDS in sequential phosphate buffer medium at different pH releasing more than 90% of the drug within 24 h duration.

Mechanism of Drug Release from MCDDS and Stability.
Release kinetic data revealed that the optimized MCDDS was �tted well into �rst-order model and apparent lag time was found to be 6 hour, followed by higuchi spherical matrix release. It was evident that 2 (0.9888) value was higher in �rst-order kinetic model as compared to the other release models. e reason for �rst-order kinetic release was due to the presence of enzyme degradable chitosan in the eudragit RLPO �lm which led to the formation of in situ ori�ces by bacterial enzyme and leaching out drug into the surrounding medium from the central polysaccharide core tablet containing sterculia gum. When the majority of chitosan particle in the eudragit coat was degraded by colonic bacterial enzymes, it ruptured due to swelling pressure of the gum core and a gradual increase in drug release was observed, as swelling increases greater surface area of sterculia gum available for bacterial action. From the stability study, the developed MCDDS was found to bestable, because there was no signi�cant change in the percentage drug content and hardness aer six month of stability study stored at 40 ∘ C ± 2 ∘ C/75% ± 5% RH.

HPLC Method Development.
A novel simple, precise, selective, speci�c, reproducible, and low cost routine reverse phase HPLC method was developed and validated as per ICH guidelines. ere were no such interfering peaks observed between the retention time of 6-MP and IS. A good resolution was obtained between 6-MP and IS with retention time of 7.88 minutes for 6-MP and 4.9 minutes for IS. e method was found to be linear ( 2 = 0.999) within the analytical range of 53.32 to 4975.00 ng/mL. Maximum recovery of the drug was obtained by using methanol: acetonitrile mixture (1 : 1). e results of the method validation were proved to be accurate and reproducible, and the drug was stable in rabbit plasma up to one month period at room temperature and at three freeze-thaw cycles.   Figure 7. Mean values of pharmacokinetic parameters are summarized in Table 5.

Journal of Pharmaceutics
In case of, MKT, the peak plasma concentration ( max ) of 6-MP was obtained within 1.5 h of administration, indicating the immediate absorption of AZA from the gastrointestinal tract and quick conversion into its active metabolite, 6-MP in blood. e max value of 6-MP following oral administration of MKT tablet was found to be 1430.08 ng/mL at the time maximum ( max ) of 1.5 h. e max value of 6-MP for EC tablet of AZA without containing sterculia gum was found to be 847.5 ng/mL at max of 5.0 h. From the results of in vitro release study, it was observed that the drug was released aer 2.0 h of dissolution study which was quite desirable, due to the fact that the drug would be released from the tablets aer passing the stomach region as the tablets were enteric coated. e results of in vivo studies of EC tablets showed that drug was not released in the stomach up to 2.0 h and therefore it gives max of 5.0 h. us, the in vivo �nding has good correlation with the in vitro results. A lag time of 6.0 h was observed from the MCDDS which revealed that the tablet had passed through the GIT and aer reaching the colon only the drug was released and appeared in plasma as 6-MP. erefore, the max value of 6-MP for the MCDDS could found to be 453.56 ng/mL at max of 9.0 h aer oral administration. e results of ANOVA revealed that there was signi�cant di�erence of AUC 0−∞ between the MCDDS, EC and MKT formulation ( 0 0 ). e results explained that the MKT formulation was more rapidly absorbed from the upper gastrointestinal tract of rabbit. But the EC and MCDDS were not absorbed from the upper GIT due to which they showed greater value of AUC 0−∞ as shown in the Table 5. It is evident that AUC for MCDDS was higher as compared to the reference formulation EC and MKT formulations (MCDDS EC MKT). Result suggests that the extent of absorption of AZA from the developed MCDDS was decreased from the large intestine, but increased from the upper part of the GIT as seen in case of EC and MKT formulation. From the in vivo studies, the max of MCDDS was found to be almost half of the EC tablet without containing sterculia gum. e longer max value (9.0 h) and low max value (453.56 ng/mL) of MCDDS as compared to the reference formulations had proved that the MCDDS released drug only at the colonic region of the rabbit intestine. is reveals localization of the drug in the colonic mucosa from the MCDDS and thereby, possibly reducing the risk of systemic toxicity.

Conclusions
Micro�ora degradation study revealed that sterculia gum can be used to release drug in the colonic region by utilizing the action of enterobacteria. e developed MCDDS exhibit gastric and small intestinal resistance but were susceptible to bacterial enzymatic attack and the potential of the system as a carrier for drug delivery to the colon is con�rmed. e swelling property of sterculia gum can be used to produce hydrostatic pressure inside the tablet if it is coated with semipermeable membrane and can be used to target drug to the colon. Chitosan-eudragit RLPO mixed �lm coating provided the favourable characteristics to the sterculia gum core tablets to deliver it directly into the colon. Chitosan in the mixed �lm coat was found to be degraded by enzymatic action of the micro�ora in the colon. e degradation of chitosan was the rate-limiting factor for drug release in the colon. Drug release from the MCDDS was directly proportional to the concentration of chitosan, but inversely related to the weight gain in thickness of eudragit RLPO coat. e enteric layer of eudragit L100 could protect eudragit RLPO membrane containing chitosan from formation of pore or rupture before SCF dissolution procedure. Drug release from optimized MCDDS �tted well into �rst-order kinetic model followed by higuchi spherical matrix release model. e HPLC method developed shows good resolution to evaluate the pharmacokinetic parameters of the drug. Pharmacokinetic studies revealed that the MRT value (13.81 h) was higher for MCDDS as compared to the other two reference formulations, which were 3.60 h for MKT and 6.62 h for EC tablets, respectively. Finally, in vivo evaluation of MCDDS in rabbit showed delayed max , prolonged absorption time, decreased max , and decreased absorption rate constant (Ka) indicating that drug was slowly absorbed from the colon making the drug available for local action in the colon, thereby, reducing the risk of systemic toxicity of the drug as compared to other dosage forms.