resveratrol loaded zinc-pectinate

8/2/2019 Resveratrol Loaded Zinc-pectinate

1/14

Research Article

Formulation and Optimization of Zinc-Pectinate Beads for the ControlledDelivery of Resveratrol

Surajit Das,1,3 Ka-Yun Ng,1,2,3 and Paul C. Ho1,3

Received 2 November 2009; accepted 9 April 2010; published online 4 May 2010

Abstract. Preventive and therapeutic efficacies of resveratrol on several lower gastrointestinal (GI)

diseases (e.g., colorectal cancer, colitis) are well documented. To overcome the problems due to its rapid

absorption and metabolism at the upper GI tract, a delayed release formulation of resveratrol was

designed to treat these lower GI diseases. The current study aimed to develop a delayed release

formulation of resveratrol as multiparticulate pectinate beads by varying different formulation

parameters. Zinc-pectinate (Zn-pectinate) beads exhibited better delayed drug release pattern than

calcium-pectinate (Ca-pectinate) beads. The effects of the formulation parameters were investigated on

shape, size, Zn content, moisture content, drug encapsulation efficiency, swellingerosion, and resveratrol

retention pattern of the formulated beads. Upon optimization of the formulation parameters in relative

to the drug release profiles, the optimized beads were further subjected to morphological, chemical

interaction, enzymatic degradation, and stability studies. Almost all prepared beads were spherical with

1 mm diameter and efficiently encapsulated resveratrol. The formulation parameters revealed great

influence on resveratrol retention and swellingerosion behavior. In most of the cases, the drug release

data more appropriately fitted with zero-order equation. This study demonstrates that the optimized Zn-

pectinate beads can encapsulate very high amount of resveratrol and can be used as delayed release

formulation of resveratrol.

KEY WORDS: bead; delayed release; formulation parameters; resveratrol; Zn-pectinate.

INTRODUCTION

Trans-3,5,4-trihydroxystilbene, commonly known asresveratrol, displays various pharmacological activities and iswidely known for its cardioprotective, chemopreventive,antioxidant, anti-inflammatory, analgesic, neuroprotective,and anti-aging activities (1,2). Although in vitro pharmaco-logical studies in several disease models exhibited positiveeffects (3), the very short half-life (814 min; 1,4) andextremely low oral bioavailability of resveratrol have raisedconcerns regarding its systemic action (1,5). At the same time,researchers have demonstrated the efficacy of resveratrol inlower gastrointestinal (GI) diseases such as colorectal cancer

and colitis (69). Our hypothesis for the current study was

that the development of a delayed release drug deliverysystem that prevents drug release in the upper GI tract andallows its release at the colonic region could provide apotential approach to treat these lower GI diseases.

There are various colon-specific drug delivery systems(time, pressure, pH, and colonic bacteria controlled; 10,11).The system that is stable in contact with upper GI bacterialenzymes but specifically degraded in the presence ofcolonic bacterial enzymes is suggested as a more realisticapproach due to their less interindividual variability thanthe other three approaches (10,12). Several biopolymers(pectin, chitosan, guar gum, etc.) have shown this peculiarproperty (11).

Because of the wide variety of types and flexibility in

use, pectins appeared to be of great practical interest ascarriers for colon-specific drug delivery. Pectins are hetero-geneous polysaccharide (partially methoxylated poly (14)-D-galacturonic acids with some 1,2-linked L-rhamnosegroups), present in the cell wall of most plants which areconsumed as part of the human diet and used as a foodadditive. It is resistant to enzymes present in the stomach andintestine but can be almost completely degraded by thecolonic bacterial enzymes (11). Unfortunately, their solubilityand swellability in aqueous fluid cause premature drugrelease before the delivery system reaching to the colon.However, this can be controlled through the choice of pectintype or the presence of additives. Usually, low methoxy

1 Department of Pharmacy, Faculty of Science, National University of

Singapore, Block S4, 18 Science Drive 4, Singapore 117543,

Republic of Singapore.2 Present Address: National Institutes of Health, Bethesda, Maryland

20892, USA.3 To whom correspondence should be addressed. (e-mail: surajitdas

[email protected]; [email protected]; [email protected])

ABBREVIATIONS: Ca, calcium; CaCl2, calcium chloride; Ca-

pectinate, calcium-pectinate; Ca2+, calcium cation; EE, encapsulation

efficiency; ER, elongation ratio; FD, freeze drying; FTIR, Fourier

transform infra-red; GI tract, gastrointestinal tract; MC, moisture

content; PR, pectinresveratrol; RT, room temperature; SER,

swellingerosion ratio; T25, 25% drug retention; T50, 50% drug

retention; T75, 75% drug retention; Zn(CH3COO)2, zinc acetate; Zn,

zinc; Zn-pectinate, zinc-pectinate; Zn2+, zinc cation.

AAPS PharmSciTech, Vol. 11, No. 2, June 2010 (# 2010)DOI: 10.1208/s12249-010-9435-7

729 1530-9932/10/0200-0729/0# 2010 American Association of Pharmaceutical Scientists


2/14

pectins (LM pectins) that have more free carboxylic groupcan be cross-linked with divalent cations (e.g., calcium, zinc)to produce a more water insoluble pectinate gel which has thepotential to be an effective vehicle for drug delivery (12).Furthermore, amidated pectins are more prone to form arigid gel structure with divalent cations than nonamidatedpectin (13,14). Therefore, amidated LM pectins allow theformation of a more compact Ca- or Zn-pectinate network

than nonamidated and/or high methoxy (HM) pectins.Additionally, multiparticulate systems proved to be better

than single unit dosage forms due to their reproducible andpredictable GI transit time, more reliable drug release profile,andless local irritation (15). Further, multiple unit dosage formsof enzyme specificity quickly spread out upon their arrival tothe colon, and due to an enhanced surface area being exposedto the enzymes, rapid drug release occurs at the colon (16).

Various reports suggest that multiparticulate Ca- or Zn-pectinate bead system may be a viable carrier for site-specificdrug delivery to the colon (15,1721). Indeed, we haverecently exploited Ca-pectinate beads as colon-specific carrierof resveratrol (22), and to our knowledge, our study was the

only one that explored the delivery of resveratrol specifi

callyto the colonic region. However, it has been observed that zincis a better cross-linking agent for pectin chains than calciumand it forms stronger pectinate network (1921,23). Hence, inthe present study, we attempted to compare the resveratrolrelease pattern from the beads prepared with both cross-linking agents (i.e., calcium and zinc). Our present studyindicated Zn-pectinate beads as better formulation than Ca-pectinate beads, in terms of superior delayed drug releasepattern. Subsequently, we aimed to optimize the formulationparameters (cross-linking solution pH, cross-linking agentconcentration, cross-linking time, drying condition, pectinconcentration, and pectin to resveratrol ratio) of resveratrol-loaded Zn-pectinate beads after studying their effect on

different physicochemical properties and resveratrol retentionpattern of the formulated beads.

MATERIALS AND METHODS

Chemicals

GENU pectin LM-104 AS-FS (degree of esterification =28% and degree of amidation = 20%) was a generous gift fromCPKelco (Denmark). Resveratrol (fine crystalline powder with99.12% purity) and calcium chloride dihydrate were purchasedfrom Shaanxi Sciphar Biotechnology Co., Ltd. (Xian, China)and Merck (Darmstadt, Germany), respectively. Sodium

hydroxide, sodium phosphate monobasic, zinc acetate dehy-drate, and Pectinex Ultra SP-L (pectinase/pectinolytic enzymefrom Aspergillus aculeatus, activity>9,500 PG ml1) wereobtained from Sigma (St. Louis, Missouri, USA). HPLC grademethanol and Fourier transform infra-red (FTIR) gradepotassium bromide were bought from Tedia Company,Fairfield, Ohio, USA and Aldrich, Germany, respectively.MiliQ water (18.2 Mcm at 25C from Millipore Direct-Q

ultra-pure water system (Billerica, Massachusetts, USA)) wasused throughout the study. Disodium hydrogen phosphateanhydrous and monobasic potassium phosphate werepurchased from Fluka (Steinheim, Germany). All materialswere used as received.

Formulation of Resveratrol-Loaded Pectinate Beads

Ionotropic gelation method was used to prepare thebeads (15,22,24,25). Briefly, resveratrol was homogeneouslydispersed in the aqueous solution of pectin with the help of ahomogenizer. The dispersion was then sonicated on a bathsonicator to remove the air bubbles. The pectin resveratrolmixture (6 ml) was added dropwise (1 ml min1) through a

25G needle (blunt end) from 5 cm above into gently agitatedcross-linking solution (100 ml) at room temperature (RT).Upon contact with cross-linking solution, the pectin dropsinstantly formed pectinate beads. Mild agitation of thesolution was continued for further cross-linking of thebeads. The beads were isolated from the solution afterdifferent time intervals, subsequently washed with distilledwater, and then dried under various conditions. The differentformulation parameters used to prepare the beads are listedin Table I. All formulations were prepared in triplicate.

Morphology

JEOL scanning electron microscopy (JSM-5200, Japan)was used for the morphological examination (bead surface) ofthe resveratrol-free and resveratrol-loaded beads. The exci-tation voltage was set at 20 kV. Pectinate beads were fixed onan aluminum stub and coated with platinum (30 s) to athickness of 2 nm under vacuum with the JEOL auto finecoater (JFC-1600, Japan). The micrographs were recorded atdifferent magnifications.

Shape and Size

The lengths and breadths of randomly selected 50 beads

from each batch were measured by precalibrated imageanalysis program (LAS EZ, version 1.4.0) after images weretaken through a digital camera (LEICA EC3, Switzerland)connected with an optical microscope (LEICA DM IL,Switzerland). The shape of the beads was represented byelongation ratio (ER), which is the quotient of length tobreadth of the beads (24). ER=1 means perfect sphericalshape, while ER>1 indicates more deviation from sphericalshape. The size of each bead was calculated as the average oflength and breadth (24).

FTIR Spectroscopy

Infra-red (IR) spectra of pure resveratrol, resveratrol-free pectin bead, and resveratrol-loaded pectin bead werecompared to examine chemical interactions between resver-atrol and bead components. Before FTIR study, pectin beadswere crushed and dried in a vacuum desiccator for 3 days toremove the moisture completely. Each sample (2 mg) wasmixed with 198 mg FTIR grade anhydrous potassiumbromide and ground into a fine powder using mortar andpestle. The mixture was then compressed into a disc. IRspectrum of each disc was recorded over a wave numberregion of 4004,000 cm1 at a resolution of 4 cm1 using theFTIR spectrometer (PerkinElmer Spectrum 100 Series,Norwalk, Connecticut, USA).

730 Das, Ng and Ho


3/14

Zinc Content

Zinc content in the beads was determined to confirm thesafety aspect of the prepared beads. The amount of zincretained in the beads was determined as follows. The beads(20 mg) were dissolved in 19 ml of 50 mM sodiumphosphate buffer (pH 7.4) containing 1% ethylenediaminete-traacetic acid. To acidify the solution, 1 ml of concentratedhydrochloric acid was added to it and mixed well. The solutionobtained was then centrifuged at 10,000g for 10 min. Theamount of zinc present in the supernatants was determined byatomic absorption spectroscopy (AAS, PerkinElmer AAnalyst100 model, Norwalk, Connecticut, USA).

Moisture Content

The moisture content (MC) was determined by gravi-metric method: 50 dry and fully dry beads were randomlyselected from each batch and weighed on an analyticalbalance with readability of 0.00001 g (Metler Toledo, Switzer-land; 22). The mean MC was estimated by the followingequation:

MC % WD WWD

100% 1

where, WD is the weight of the dry beads and Wis the weightof fully dry beads (dried at 60C until no further weightchange was observed).

Resveratrol Encapsulation Efficiency

During formulation of the resveratrol-loaded pectinatebeads (as described in the above section), resveratrol contentin the beads was determined using a spectrophotometer(UVVisible Spectrophotometer UV-1601, Shimadzu, Japan)at a wavelength of 320 nm (22). Encapsulation efficiency(EE) was determined by analyzing spectrophotometricallythe total amounts of resveratrol lost to the cross-linking andwashing solutions during formulation. EE of the beads werethen calculated by the following equation:

EE % QT QIQT

100% 2

where, QT is the amount of resveratrol initially added in 6 ml

pectin gel mixture, and QI is the total amount of resveratrollost in the cross-linking and washing solutions.

In Vitro Drug Release Study

The in vitro release study of the resveratrol-loadedpectinate beads was carried out according to the proceduresdeveloped in our previous study (22). Because of themarginal solubility of resveratrol in simulated intestinal fluid(SIF) and simulated gastric fluid (SGF), the release ofresveratrol (in SIF/SGF) is not observed even in the presenceof a high amount of surfactant (SIF with 0.2% Tween 80exhibited resveratrol solubility of 0.720.13 g ml1). Thus,

Table I. Formulation Design

Variables Values

Constants

Zinc acetate

concentration (% w/v)

Cross-linking

solution pH

Cross-linking

time (h) Drying

Pectin concentration

(% w/v) PectinResveratrol

Cross-linking agent Zn 5 1.5 0.5 RT 5 3:1

Ca 5a 1.5 0.5 RT 5 3:1

Zinc acetateconcentration (% w/v)

1 1.5 0.5 RT 5 3:12.5 1.5 0.5 RT 5 3:1

5 1.5 0.5 RT 5 3:1

10 1.5 0.5 RT 5 3:1

Cross-linking pH 1.5 5 0.5 RT 5 3:1

6.3 5 0.5 RT 5 3:1

Cross-linking time (h) 0.25 5 1.5 RT 5 3:1

0.5 5 1.5 RT 5 3:1

2 5 1.5 RT 5 3:1

24 5 1.5 RT 5 3:1

Drying RT 5 1.5 0.5 5 3:1

37C 5 1.5 0.5 5 3:1

50C 5 1.5 0.5 5 3:1

FD 5 1.5 0.5 5 3:1

Pectin concentration

(% w/v)

3 5 1.5 0.5 RT 3:1

4 5 1.5 0.5 RT 3:15 5 1.5 0.5 RT 3:1

6 5 1.5 0.5 RT 3:1

PectinResveratrol 1:1 5 1.5 0.5 RT 5 2:1 5 1.5 0.5 RT 5 3:1 5 1.5 0.5 RT 5 4:1 5 1.5 0.5 RT 5

Blank 5 1.5 0.5 RT 5

a Calcium chloride was used instead of zinc acetate

731Formulation and Optimization of Zinc-Pectinate Beads


4/14

the standard dissolution methods are not feasible in this caseas the drug is released in the dissolution media as particulateform rather than dissolved form due to the poor aqueoussolubility of resveratrol, which impede proper quantificationof the released drug.

Hence, we devised an alternative method in which weestimated the amount of resveratrol remaining in the intactbeads at each time point instead of measuring the drug

released in the media (SIF/SGF). This method precludesmeasurement of drug released (either in dissolved orparticulate form) in the release media. Briefly, resveratrol-loaded beads (25 mg) were suspended in 10 ml releasingmedium (preheated at 370.2C) in screw cap glass testtubes. Separate tubes were used for each time point. Thetubes were kept on a shaking water bath (100 rpm) at 37 0.2C. Intact beads were separated from the specified tubes atpredetermined time intervals. The isolated intact beads werethen dissolved (pectinate matrix)/dispersed (resveratrol par-ticles) in 5 ml of 50 mM PBS (pH 7.4) and fully dissolved(resveratrol) by the addition of 10 ml methanol. The mixturewas mixed well on a magnetic stirrer and centrifuged at

10,000g for 10 min. The supernatant was collected anddiluted with methanolwater (1:1) to the calibration range(0.110 g ml1). The resveratrol content in the supernatant(i.e., resveratrol remaining in the beads) was determined by aUVVisible Spectrophotometer (UV-1601, Shimadzu) at320 nm. Resveratrol-free beads were used as a control.

SIF (pH 6.8) was used as release media throughout thebead optimization process. Further, release study of theoptimized beads was performed in SIF (pH 6.8) containing300 PG Pectinex Ultra SP-L (simulated colonic fluid) andSGF (pH 1.2) to investigate the effect of colonic and gastricfluids on resveratrol release from the bead.

Analysis of the Drug Release Kinetics

The drug release kinetics from the beads were inves-tigated by fitting the drug release data into zero order andsquare root of time equations. The zero order equation canbe obtained by plotting the percent drug released versus timeas follows:

Q Q0 k0t 3The square root of time equation can be obtained by plottingthe percent drug released versus square root of time asfollows:

Qt

kffiffi

tp

4

where, Q is the percentage of drug remaining at time t(h), Q0is the percentage of drug at t=0 h, k0 is the zero order releaserate constant, Qt is the percentage of drug released at time t(h), and k is the Higuchis release rate constant.

The drug release data obtained from the release studieswere fitted into the above models, and the appropriate modelwas selected after linear regression analysis. R2>0.95 was setfor linearity. The release rate constants were calculated fromthe slope of each straight line of best fit model. Timescorresponding to 25%, 50%, and 75% resveratrol retention inthe beads (T25, T50, and T75, respectively) during releasestudy were determined.

SwellingErosion Behavior

The experimental conditions for swellingerosion studywere similar to the release study. The beads were removed atthe predetermined time point from the specified tubes,blotted with filter paper to remove the excess water, andweighed. Swellingerosion behavior of the optimized beadswere also checked in SGF (pH 1.2) and in SIF (pH 6.8)

containing 300 PG Pectinex Ultra SP-L. The swellingerosion ratio (SER) was calculated using the followingequation:

SER % WT W0W0

100% 5

where, WT and W0 represent the weight of the beads at thegiven time point and the initial weight of the dry beads,respectively. Positive SER and upward trend denote overallswelling, while negative SER and downward trend demon-strate the erosion and/or drug release of beads. SER =100%denotes complete dissolution of beads.

Enzymatic Degradation

The beads prepared with the optimized formulationparameters were incubated in SIF (pH 6.8) with or without300 PG pectinase enzyme at 370.2C. After 3 h incubation,the beads were separated from the media and lyophilized.The appearance of the enzyme-treated and nonenzyme-treated beads were compared by scanning electron micro-scopy (JSM-5200) using the same parameters to those of themorphological study.

Stability

About 100 mg of the beads prepared with the optimizedformulation parameters were stored at three different con-ditions: cold (4C), RT, and accelerated temperature (40C).Resveratrol content of the samples was analyzed at prede-termined time intervals (0, 3, 7, 15, 30, 90, and 180 days).

Statistical Analysis

Graph-Pad Prism Version 2.00 (San Diego, California,USA) was used for statistical analysis. All experiments wereperformed in triplicate and data were expressed as mean standard deviation (SD). One-way ANOVA with the posthoc Tukey test and two-tail unpaired t test were performedfor all statistical analysis. Statistical significance was set at p


5/14

linking reduces the solubility and induces noncovalentassociations of the carbohydrate chains in pectin. Theseinteractions in amidated LM pectin allowed the formation ofa compact pectinate network, in which insoluble resveratrolparticles were entrapped. Nevertheless, the beads were easilyprepared without any sophisticated instruments.

Morphology

All beads were spherical in shape (ER


6/14

components when formulated into Zn-pectinate beads(Fig. 2). All characteristic peaks of resveratrol (olefinic bandat 966 cm1, CO stretching at 1383, CC olefinic stretching at

1585, and C

C aromatic double-bond stretching at 1607 cm

1

)are present in the resveratrol-loaded Zn-pectinate bead.

Zinc Content

The zinc contents per mg of the beads are depicted inTable II. The results suggest the potential impact offormulation parameters on the zinc retention in the beads.Zinc content increased with cross-linking solution pH, zincacetate concentration, cross-linking time, and pectin toresveratrol ratio, while content decreased with pectin con-centration. Lower amount of zinc is expected to bind withpectin chains due to suppression of ionization of the

carboxylic group of the pectin chains at low cross-linkingsolution pH, which might repress the zinc entrapment.Augmentation of zinc content with increasing zinc acetateconcentration and cross-linking time was due to availability ofmore zinc and more time for the cross-linking, respectively.One previous report evidenced that only small amount ofcalcium is required for Ca-pectinate network formation withthe majority of calcium remaining as free calcium chloride in

the Ca-pectinate beads (27). Similar to Ca-pectinate beads,we anticipate that majority of zinc remained as free zincacetate in Zn-pectinate beads. At lower pectin to resveratrolratio, less pectin is available for cross-linking which couldprobably explain lower zinc retention. During bead prepara-tion, when dilute pectin droplets come in contact with zincacetate solution, zinc acetate can easily penetrate into thepectin droplets. This might be the reasons for higher zinccontent at lower pectin concentration. Zinc content was foundin the order of RT< 37C 95.9%) wasobserved in all cases (Table II). Poor aqueous solubility ofresveratrol was responsible for its high encapsulation in thebeads. However, Zn-pectinate beads revealed lower resvera-trol encapsulation and loading than Ca-pectinate beads. Thefinding is similar to observation in other studies on pectinatebeads (19,21). This is probably due to formation of strongerzinc-pectinate network during bead formation, which led to

Fig. 1. Scanning electron micrographs of the surface of a resveratrol-

free Zn-pectinate bead ( 500), b resveratrol-loaded Zn-pectinate bead

( 350), and c resveratrol-loaded Ca-pectinate bead ( 500). Magnifica-

tions corresponding to each figure are presented in parentheses

734 Das, Ng and Ho


7/14

ejection of resveratrol towards the outer part of the beadsand subsequent drug leakage in the cross-linking solution.

Figure 3a (resveratrol retention profile) and Table III(T75, T50, and T25) demonstrate that Zn-pectinate beads wereable to retain higher amount of drug than Ca-pectinate beadsin SIF. This observation is in the same line with others (1921,23). Consequently, huge decrease of SER was observed for

Zn-pectinate than Ca-pectinate beads (Fig. 3b). Ca-pectinatebeads were hydrated and swelled rapidly in contact with SIF,leading to subsequent drug release due to their high molecularporosity. In contrast, Zn-pectinate beads swelled very little.Wellner etal. compared different divalentcations for gel formingability with pectin and found zinc was a stronger cross-linkerthan calcium (29). It was also reported that zinc poses higherbinding ability with a higher affinity than calcium (30), which canform pectinate bead with higher gel strength. Thus, we decidedto use zinc acetate as the cross-linking agent for furtheroptimization of the formulation parameters.

Fig. 2. FTIR spectra of a blank Zn-pectinate bead, b resveratrol-loaded Zn-pectinate bead, and c pure resveratrol

Fig. 3. Effect of cross-linking agents (Zn(CH3COO)2: zinc acetate,

CaCl2: calcium chloride) on a retention of resveratrol in the beads

and b SER of the beads. The beads were incubated in SIF. Data

represents mean SD. *p


8/14

Fig. 5. Effect of zinc acetate concentration on a retention of

resveratrol in the beads and b SER of the beads. The beads were

incubated in SIF. Data represents mean SD. *p


9/14

Cross-linking Solution pH

Diminution of pH of the cross-linking solution producedbeads with smaller size (Table II). A probable reason is that

at reduced pH, nonionic interaction occurs along with ionicinteraction between pectin chains, which leads to small andcompact beads. Weight of beads slightly increased withincreasing cross-linking solution pH (Table II). Moreover,MC of the beads was also found to increase with increasingpH of cross-linking solution (Table II). Nevertheless, insig-nificant increase in EE was observed in the present study withdecreased cross-linking solution pH (Table II).

Drug retention profile shows that resveratrol retentionwas significantly higher in case of beads prepared at cross-linking solution pH 1.5 (Fig. 4a). Table III shows that therelease parameters (T75, T50, and T25) were significantlylower in case of beads prepared at cross-linking solution withunmodified pH (pH 6.3). Similar effect of cross-linking

solution pH on pectinate beads was reported before (21).Our previous study with Ca-pectinate beads also evidencedsimilar effect of cross-linking solution pH (22). As shown inFig. 4b, SER values were higher in case of beads prepared atpH 6.3 up to 4 h and then erosion dominated over swelling.This might be because of the formation of a stronger Zn-pectinate network at acidic cross-linking solution (31). Inaddition to the ionic interaction between carboxylic groups ofpectin and zinc cation, several nonionic interactions (e.g.,hydrophobic interaction, hydrogen bonding) are involve inamidated pectin at pH below 3 (lower than pKa value ofpectin), leading to the formation of strong Zn-pectinatenetwork at acidic pH (3234). Similar to Ca-pectinate net-

work, reduction in cross-linking solution pH may lead toconformational transition from the twofold to the morecompact threefold helical conformation of the polygalacturo-nate chains of pectin in the Zn-pectinate network (22). Thesecould account for the effects of adjusting the cross-linkingsolution pH on resveratrol retention and SER of Zn-pectinatebeads. Subsequently, we set the cross-linking solution pH at1.5, while optimizing the other formulation parameters of

resveratrol-loaded Zn-pectinate beads.

Zinc Acetate Concentration

Our preliminary studies showed that zinc acetate con-centration below 1% produced irregular-shaped beads andabove 10% gave no improvement of drug retention (data notshown), concentration was then kept between 1% and 10% inthe present study. Augmentation of the beads size andweight was observed with increasing zinc acetate concen-tration in the cross-linking medium (Table II). However,some other studies reported an inverse relationship betweenthe size and concentration of cross-linking agent (17). Table II

demonstrates that MC increased with increasing zinc acetateconcentration. However, a little improvement in EE wasnoticed with decreasing zinc acetate concentration (Table II).

Figure 5a demonstrates that the retention of resveratrolin the beads depends on the zinc acetate concentration. Therelease of resveratrol was significantly faster at 1% and 2.5%zinc acetate than 5%, while 5% and 10% zinc acetate showedalmost similar drug retention pattern. In addition, the releaseparameters (T75, T50, and T25) support the above phenomena

Fig. 9. Effect of pectin to resveratrol ratio on a retention of

resveratrol in the beads and b SER of the beads. The beads were

incubated in SIF. Data represents mean SD. *p


10/14

(Table III). Our previous work with Ca-pectinate beadsshowed increased retention of drug with increasing calciumchloride concentration up to a certain value (22). Anotherearlier work also exhibited slower drug release at higher zincacetate concentration (20). In general, SER decreased withincreasing concentration of zinc acetate (Fig. 5b). Very highSERs were detected when beads were prepared at 1% zincacetate. It demonstrates an early phase of swelling, followed

by rapid erosion of the beads. These differences in drugretention and SER could be explained by difference in degreeof cross-linking, which could effect the swelling rate andsubsequent penetration of solvent into the beads. It wasreported that the calcium cation leads to dimer formationwith pectin chain at low calcium concentration, whereas athigh calcium concentration, enhancement of the number and/or strength of cross-links between pectin and calcium ions aswell as aggregation of initial dimmers take place (13,27). Weanticipate similar phenomena with Zn-pectinate beads, whichled to a higher degree of cross-linking with increasingconcentration of zinc acetate. In summary, higher drugretention and lower SER were observed due to higher gel

strength at higher zinc acetate concentrations. Based on thisfinding, 5% zinc acetate was chosen as the optimal concen-tration for subsequent optimization processes.

Cross-Linking Time

The longer cross-linking time resulted larger and heavierbeads with higher MC and lower EE (Table II).

The release profiles of resveratrol from the beadsprepared with various cross-linking times are shown inFig. 6a. It is obvious from the figure that the cross-linkingtime has great influence on the release of resveratrol from theZn-pectinate beads. Retention of resveratrol was higher whenbeads were prepared at longer cross-linking time (0.524 h);whereas, retention of resveratrol was the lowest at theshortest cross-linking time (0.25 h). However, insignificantdifferences in drug retention were observed among the beadsprepared at cross-linking time of 0.5, 2, and 24 h, althoughmore drug retention was noticed in all cases than beadsprepared at cross-linking time of 0.25 h (Table III). El-Gibalyhas also demonstrated reduced drug release with increasing

Table III. Release Kinetics and Parameters of the Beads (Data Represents Mean SD ( N= 3))

Variables Values SQRT (R2) Zero order (R2)

Time (h)i

T75 T50 T25

Cross-linking agent Zn 0.980 0.997 3.480.21a 6.370.20a 9.260.27a

Ca 0.972j 0.923 2.180.17 4.270.15 7.040.08

Cross-linking pH 1.5 0.980 0.997 3.480.21b 6.370.20b 9.260.27b

6.3 0.989j

0.942 1.620.02 3.160.02 5.230.03Zinc acetate concentration (%) 1 0.992j 0.982 1.230.23c 3.300.24c 6.380.18c

2.5 0.986 0.995 2.820.36 5.290.34d 7.750.31d

5 0.980 0.997 3.480.21 6.370.20 9.260.27

10 0.976 0.998 3.460.42 6.310.47 9.160.52

Cross-linking time (h) 0.25 0.989 0.990 1.960.41e 4.800.27e 7.640.18e

0.5 0.980 0.997 3.480.21 6.370.20 9.260.27

2 0.964 0.995 3.510.33 6.520.22 9.530.36

24 0.974 0.988 3.410.28 6.410.23 9.410.19

Drying RT 0.980 0.997 3.480.21 6.370.20 9.260.27

37C 0.979 0.998 3.470.13 6.330.03 9.200.18

50C 0.975 0.992 3.530.11 6.410.19 9.280.27

FD 0.913k 0.821k Pectin concentration (%) 3 0.988 0.992 2.820.42f 5.70 0.32 8.58 0.22

4 0.986 0.987 3.000.19 5.950.35 8.900.52

5 0.980 0.997 3.480.21 6.370.20 9.260.276 0.971 0.998 3.510.09 6.390.18 9.270.35

Pectin-Resveratrol 1:1 0.994j 0.972 1.950.27g 3.940.27g 6.630.21g

2:1 0.979 0.996 2.610.08 5.330.11 8.050.16

3:1 0.980 0.997 3.480.21 6.370.20 9.260.27

4:1 0.982 0.994 4.100.14 7.430.21 10.760.29

Enzyme 0.994j 0.967 2.020.23h 4.200.28h 7.170.31h

ap


11/14

cross-linking time (20). Lower drug release at higher cross-linking time was also evident in case of Ca-pectinate beads inour previous study (22). The SER profile is depicted inFig. 6b. In our findings and as expected, 0.25 h cross-linkingtime showed the highest SER (swelling of the beadsaccompanied by erosion and drug release). However, almostsimilar SER profiles were noticed among 0.5, 2, and 24 hcross-linking time. Promotion of cross-linking between pectin

chains and zinc ions with respect to time might be thepossible explanation for the observed SER and resveratrolretention pattern. Additionally, the aggregation of primarydimmers probably takes place at higher cross-linking time,which improves the strength of Zn-pectinate network. In thisstudy, the optimized cross-linking time was found to be 0.5 h.

Drying Condition

Beads size and weight decreased with increasing dryingtemperature (Table II). The size of the freeze-dried beadswas almost similar to the wet beads. This might be because ofmoisture evaporation from the beads without affecting thebead structure during lyophilization. A clear correlation wasfound between MC of the beads with the drying method(Table II). Lower MC was observed with higher dryingtemperature. The lyophilized beads and those dried at 50Cexhibited minimum MC (FD


12/14

tion. Resveratrol particles probably released through thesepores during matrix degradation.

Release Kinetics

In the present study, we noticed that the drug releasekinetics of Ca-pectinate beads as well as Zn-pectinate beadsprepared with 1% zinc acetate, cross-linking solution pH of6.3, and pectin to resveratrol ratio = 1:1 together with Zn-pectinate beads in presence of enzyme showed high correla-tion with the square root of time equation (Table III). Therest of the beads more appropriately fitted with the zeroorder equation (except lyophilized beads which did not fitboth equations). However, these observations do not matchwith the release kinetics of the Ca-pectinate beads, where allrelease data more appropriately fitted with the square root of

time equation (22). One example to show the linearity ofplots for the cumulative percent resveratrol released versustime is presented in Fig. 12. Because of the marginal solubilityof resveratrol in SIF (0.110.01 g ml1; 4), most of theresveratrol particles should not be dissolved in the releasemedia that penetrated into the bead matrix. Similar to the Ca-pectinate beads, the release of resveratrol from the Zn-pectinate beads probably follows the following threeconsecutive phenomena: (a) hydration of the matrix, (b)swelling and erosion of the matrix, and (c) the release ofresveratrol from the matrix as undissolved particles ratherthan diffusion of the soluble resveratrol. Upon contact withSIF (basically a phosphate buffer), fluid penetrates into the

Zn-pectinate network and ion exchange between Zn2+ andNa+/K+ ions (components of SIF) takes place. As a result,partially soluble pectin regions are formed due todisplacement of the Zn2+ ions from Zn-pectinate network,which are more permeable (36). Pores are created in thematrix due to the release of undissolved resveratrol particles,which allow more fluid penetration in to the matrix andsubsequent release of more resveratrol particles followed byerosion of the bead matrix. When the pectinate matrix isstrong, the drug release from the beads is impeded by lessfluid penetration and less erosion of the matrix.

STABILITY

The stability profile of resveratrol in the optimized beadsindicates that resveratrol was stable inside the beads at 4C

and RT, even after 6 months (Fig. 13). Stability was 99.87%and 98.99% in case of storage at 4C and RT, respectively,whereas 93% when stored at accelerated condition (40C).These results indicate desired stability of the formulation at4C and RT.

Fig. 11. Scanning electron micrographs of a surface of resveratrol-

loaded bead after 3 h incubation in SIF containing 300 PG pectinase

( 350) and b surface of resveratrol-loaded bead after 3 h incubation

in SIF ( 350). Magnifications corresponding to each figure are

presented in brackets

Fig. 12. The release of resveratrol from Zn-pectinate beads prepared

at different cross-linking time, plotted as the cumulative percent

resveratrol released versus time. Data represents mean SD

Fig. 13. Stability of resveratrol in the optimized resveratrol-loaded

Zn-pectinate beads. Data represents mean SD. # represents p


13/14

CONCLUSIONS

Because resveratrol has shown preventive as well astherapeutic activity to the lower GI diseases (e.g., colorectalcancer, colitis) but quickly absorbed and rapidly metabolizedat the upper GI tract, a delayed release or colon-specific drugdelivery system is essential to treat these diseases. To fulfillthis aim, we have attempted to develop a multiparticulate

resveratrol-loaded pectinate bead. Our present studyrevealed that Zn-pectinate beads are better than Ca-pectinatebeads as delayed release formulation of resveratrol. Duringoptimization of the resveratrol-loaded Zn-pectinate formula-tion, formulation parameters appeared to have great influ-ence on beads properties and drug release characteristics.The release mechanism of resveratrol from most of the Zn-pectinate beads followed zero order equation. Beads wereable to encapsulate very high amount of resveratrol and thebeads prepared with optimized formulation parametersexhibited delayed release of resveratrol. The beads were alsostable at RT and 4C storage. However, despite its delayeddrug release pattern, additional modifications of the beads

such as enteric coating, use of hardening agent, and complex-ation with other polymers are necessary to achieve colon-specific delivery of resveratrol. We are currently aiming toachieve this goal.

ACKNOWLEDGMENTS

This work was partially supported through a NationalUniversity of Singapore Academic Research Funds R148-050-068-101 and R148-050-068-133.

REFERENCES

1. Baur JA, Sinclair DA. Therapeutic potential of resveratrol: thein vivo evidence. Nat Rev Drug Discov. 2006;5(6):493506.

2. Jang M, Cai L, Udeani GO, Slowing KV, Thomas CF, Beecher CWet al. Cancer chemopreventive activity of resveratrol, a naturalproduct derived from grapes. Science. 1997;275(5297):21820.

3. Delmas D, Lancon A, Colin D, Jannin B, Latruffe N. Resver-atrol as a chemopreventive agent: a promising molecule forfighting cancer. Curr Drug Targets. 2006;7(4):42342.

4. Das S, Lin HS, Ho PC, Ng KY. The impact of aqueous solubilityand dose on the pharmacokinetic profiles of resveratrol. PharmRes. 2008;25(11):2593600.

5. Baur JA, Pearson KJ, Price NL, Jamieson HA, Lerin C, Kalra Aet al. Resveratrol improves health and survival of mice on a high-calorie diet. Nature. 2006;444(7117):33742.

6. Martin AR, Villegas I, La Casa C, de la Lastra CA. Resveratrol,

a polyphenol found in grapes, suppresses oxidative damage andstimulates apoptosis during early colonic inflammation in rats.Biochem Pharmacol. 2004;67(7):1399410.

7. Martin AR, Villegas I, Sanchez-Hidalgo M, de la Lastra CA. Theeffects of resveratrol, a phytoalexin derived from red wines, onchronic inflammation induced in an experimentally inducedcolitis model. Br J Pharmacol. 2006;147(8):87385.

8. Tessitore L, Davit A, Sarotto I, Caderni G. Resveratrol depressesthe growth of colorectal aberrant crypt foci by affecting bax andp21(CIP) expression. Carcinogenesis. 2000;21(8):161922.

9. Schneider Y, Duranton B, Gosse F, Schleiffer R, Seiler N, Raul F.Resveratrol inhibits intestinal tumorigenesis and modulates host-defense-related gene expression in an animal model of humanfamilial adenomatous polyposis. Nutr Cancer. 2001;39(1):1027.

10. Basit AW. Advances in colonic drug delivery. Drugs. 2005;65(14):19912007.

11. Friend DR. New oral delivery systems for treatment ofinflammatory bowel disease. Adv Drug Deliv Rev. 2005;57(2):24765.

12. Liu L, Fishman ML, Kost J, Hicks KB. Pectin-based systems forcolon-specific drug delivery via oral route. Biomaterials. 2003;24(19):333343.

13. Wakerly Z, Fell J, Attwood D, Parkins D. Studies on amidatedpectins as potential carriers in colonic drug delivery. J PharmPharmacol. 1997;49(6):6225.

14. Thakur BR, Singh RK,HandaAK. Chemistryand uses of pectina review. Crit Rev Food Sci Nutr. 1997;37(1):4773.

15. Maestrelli F, Cirri M, Corti G, Mennini N, Mura P. Develop-ment of enteric-coated calcium pectinate microspheres intendedfor colonic drug delivery. Eur J Pharm Biopharm. 2008;69(2):50818.

16. Rodriguez M, Vila-Jato JL, Torres D. Design of a new multi-particulate system for potential site-specific and controlled drugdelivery to the colonic region. J Control Release. 1998;55(1):6777.

17. Sriamornsak P. Effect of calcium concentration, hardening agentand drying condition on release characteristics of oral proteinsfrom calcium pectinate gel beads. Eur J Pharm Sci. 1999;8(3):2217.

18. Maestrelli F, Zerrouk N, Cirri M, Mennini N, Mura P. Micro-spheres for colonic delivery of ketoprofen-hydroxypropyl-beta-cyclodextrin complex. Eur J Pharm Sci. 2008;34(1):111.

19. Dupuis G, Chambin O, Gnelot C, Champion D, Pourcelot Y.Colonic drug delivery: influence of cross-linking agent on pectinbeads properties and role of the shell capsule type. Drug DevInd Pharm. 2006;32(7):84755.

20. El-Gibaly I. Oral delayed-release system based on Zn-pectinategel (ZPG) microparticles as an alternative carrier to calciumpectinate beads for colonic drug delivery. Int J Pharm. 2002;232(12):199211.

21. Chambin O, Dupuis G, Champion D, Voilley A, Pourcelot Y.Colon-specific drug delivery: influence of solution reticulationproperties upon pectin beads performance. Int J Pharm. 2006;321(12):8693.

22. Das S, Ng K-Y. Resveratrol-loaded calcium-pectinate beads:effects of formulation parameters on drug release and beadcharacteristics. J Pharm Sci. 2010;99(2):84060.

23. Atyabi F, Majzoob S, Iman M, Salehi M, Dorkoosh F. In vitroevaluation and modification of pectinate gel beads containing

trimethyl chitosan, as a multi-particulate system for delivery ofwater-soluble macromolecules to colon. Carbohydr Polym.2005;61(1):3951.

24. Wong TW, Nurjaya S. Drug release property of chitosan-pectinate beads and its changes under the influence of micro-wave. Eur J Pharm Biopharm. 2008;69(1):17688.

25. Bourgeois S, Tsapis N, Honnas H, Andremont A, Shakweh M,Besnard M et al. Colonic delivery of beta-lactamases does notaffect amoxicillin pharmacokinetics in rats. J Pharm Sci. 2008;97(5):185363.

26. Grant GT, Morris ER, Rees DA. Biological interactions betweenpolysaccharides and divalent cations: the egg box model. FEBSLett. 1973;32(1):1958.

27. Bourgeois S, Gernet M, Pradeau D, Andremont A, Fattal E.Evaluation of critical formulation parameters influencing thebioactivity of beta-lactamases entrapped in pectin beads. Int J

Pharm. 2006;324(1):2

9.28. Drinker KR, Thompson PK, Marsh M. An investigation of theeffect upon rats of long continued ingestion of zinc compounds,with a special reference to the relation of zinc excretion to zincintake. Am J Physiol. 1927;80:284306.

29. Wellner N, Kacurakova M, Malovikova A, Wilson RH, BeltonPS. FT-IR study of pectate and pectinate gels formed by divalentcations. Carbohydr Res. 1998;308(12):12331.

30. Dronnet VM, Renard CMGC, Axelos MAV, Thibault JF.Characterisation and selectivity of divalent metal ions bindingby citrus and sugar-beet pectins. Carbohydr Polym. 1996;30(4):25363.

31. Pillay V, Fassihi R. In vitro release modulation from crosslinkedpellets for site-specific drug delivery to the gastrointestinal tract.I. Comparison of pH-responsive drug release and associatedkinetics. J Control Release. 1999;59(2):22942.

741Formulation and Optimization of Zinc-Pectinate Beads


14/14

32. Cardoso SM, Coimbra MA, Lopes da Silva JA. JA. Temperaturedependence of the formation and melting of pectin-Ca 2+ networks:A rheological study. Food Hydrocoll. 2003;17(6):8017.

33. Gilsenan PM, Richardson RK, Morris ER. Thermally reversibleacid-induced gelation of low-methoxy pectin. Carbohydr Polym.2000;41(4):33949.

34. Lootens D, Capel F, Durand D, Nicolai T, Boulenguer P,Langendorff V. Influence of pH, Ca concentration, temperature

and amidation on the gelation of low methoxyl pectin. FoodHydrocoll. 2003;17(3):23744.

35. Sriamornsak P, Nunthanid J. Calcium pectinate gel beads forcontrolled release drug delivery: I. Preparation and in vitrorelease studies. Int J Pharm. 1998;160(2):20712.

36. Sriamornsak P. Investigation of pectin as a carrier for oraldelivery of proteins using calcium pectinate gel beads. Int JPharm. 1998;169(2):21320.

742 Das, Ng and Ho

resveratrol loaded zinc-pectinate

Documents