International Journal of Pharmaceutics 212 (2001) 19-28
`
`international
`journal of
`_
`pharmaceutics
`
`www.elsevier. com,"locate_-'ijpharm
`
`Novel pH-sensitive citrate cross-linked chitosan film for
`drug controlled release
`
`X.Z. Shu, KJ. Zhu *, Weihong Song
`
`Department of Polymer Science and Engineering, Zhejiang University, Hangzhou 310027, People's Republic of China
`
`Received 31 July 2000; received in revised form 8 September 2000; accepted 8 September 2000
`
`Abstract
`
`Turbidimetric titration revealed that there were electrostatic attractive interactions between citrate and chitosan in
`
`the pH region of 4.3-7.6, depending on their degree of ionization. Citrate cross-linked chitosan film was prepared
`simply by dipping chitosan film into sodium citrate solution. The swelling ratio of citratefcl-iitosan film was sensitive
`to pH, ionic strength etc. Under acidic conditions, citratefchitosan film swelled and even dissociated in the pH less
`than 3.5, and the model drugs (brilliant blue and riboflavin) incorporated in the film were released quickly (usually
`within 2 h released completely in simulated gastric fluid at 37°C) while under neutral conditions the swelling ratio of
`citratefchitosan film was less significant and the release rate of brilliant blue and riboflavin was low (less than 40%
`released in simulated intestinal fluid in 24 h). Sodium chloride weakened the electrostatic interaction between citrate
`and chitosan, and therefore facilitated the film swelling and accelerated drug release. The parameters of film
`preparation such as citrate concentration, solution pH etc. influencing the film swelling and drug release profiles were
`examined. The lower concentration and the higher pH of citrate solution resulted in a larger swelling ratio and
`quicker riboflavin release. To improve the drug controlled release properties of citrate/chitosan film, heparin, pectin
`and alginate were further coated on the film surface. Among them only the coating of alginate prolonged riboflavin
`release noticeably (for 80% of drug released the time was extended from 1.5 to 3.5 h with 0.5% w/v alginate used).
`The results indicated that the citrateichitosan film was useful
`in drug delivery such as for the site-specific drug
`controlled release in stomach. @ 2001 Elsevier Science B.V. All rights reserved.
`
`Keywords: Chitosan film; Sodium citrate; pl-I-sensitive; Drug controlled release
`
`L Introduction
`
`Chitosan with excellent biodegradable and bio_
`compatible characteristics is a naturally occurring
`polysaccharide. Due to its unique polymeric
`
`“ Corresponding author. 'l'cl.: + 86-571-7952046.
`E-mail address: lcjzhu@ipsm.zju.cdu.cn (K..l. Zhu).
`
`cationic character, its gel and film forming prop-
`erties, chitosan has been extensively examined in
`the pharmaceutical industry for its potential in the
`d°"°I°Pme“t °f drug deI""°"Y Systems (Y5-‘° ct 31-!
`19954 mum» 1993)-
`Chitosan films were usually prepared by chemi-
`cal cross-linking with glutaraldehyde etc. (Nakat-
`Suka and Andrady, 1992; Thacharodi and R30,
`
`RBP_TEVA05022451
`
`TEVA EXHIBIT 1018
`TEVA PHARMACEUTICALS USA, INC. V. MONOSOL RX, LLC
`
`

`
`acidic conditions due to the ionization of amino
`
`groups but remained in a shrunken state under
`neutral condition. Moreover, chitosan was re-
`ported to have intragastric-floating characteristics
`and prolonged retention of the dosage form in the
`stomach. By utilizing these advantages, chitosan
`films or other dosage forms have been exploited
`widely for oral sustained drug delivery in the
`stomach (Inouye et
`al.,
`1988; Chandy and
`Sharma, 1993; Patel and Amiji, 1996; Gupta and
`Ravi Kumar, 2000). To improve the pH-sensitive
`performance, blended chitosan films have usually
`been prepared. For example, polyether oxide/chi-
`tosan film was reported to have excellent pH-sen-
`sitivity (Yao et al., 1993; Angelova et al., 1995;
`Patel and Amiji, 1996).
`However, the chemical cross-linking agents pos-
`sibly induce toxicity and other undesirable effects.
`To overcome these disadvantages,
`recently re-
`versible physical cross-linking by electrostatic in-
`teraction was applied in the preparation of
`chitosan film (Illum, 1998). Polyanions were usu-
`ally used as a component to prepare these films.
`For example, Yao et al.
`(1996)
`reported the
`preparation of pectin/chitosan fihns by dissolving
`this polyelectrolyte complex in formic acid and
`then evaporating the solvent. Chu et al. (1995)
`also prepared xanthan/chitosan complex film by
`the solvent evaporation method in the existence of
`concentrated sodium chloride (ca. 0.5 M) and
`then treatment at a high temperature.
`On the other hand, the use of low molecular
`weight ions to prepare an ionic cross-linking poly-
`meric matrix was found to be very simple and
`mild, and the cross-linking process was accom-
`plished just by dipping the polymer films into
`cross-linking ion solution (Al-Musa et al., 1999).
`For
`instance Remufifin-Lopez and Bodmeier
`(1997) prepared tripolyphosphate cross-linked chi-
`tosan
`film by dipping
`chitosan
`film into
`tripolyphosphate aqueous solution.
`However, up to now, no other anion cross-
`Iinked chitosan film is reported in the literature.
`In our previous experiments, we found that there
`was electrostatic interaction between sodium cit-
`
`rate and chitosan, and citrate cross-linked chi-
`tosan beads or microspheres were prepared using
`
`2000a,b). In this paper, we aim to prepare citrate
`cross-linked chitosan film and investigate the pH-
`sensitive performances of citrate/chitosan film.
`The preliminary results of citrate/chitosan film as
`pH-dependent drug controlled release matrix are
`also reported.
`
`2. Materials and methods
`
`2.1. Materials
`
`Chitosan was obtained from Tianbao Chitosan
`
`Co. Ltd (China), and refined twice by dissolving
`in dilute HAc solution and precipitating from
`dilute ammonia, the degree of deacetylation was
`86%, Mv was 460 000. Pectin (USP XXII) and
`sodium alginate (low viscosity) were obtained
`from Sigma (USA). Heparin (MW 11 000, from
`porcine intestinal mucosa) was a gift from Ji-
`uyuan Gene Co. Ltd (China). Riboflavin (Mw
`376.37), theophylline (Mw 180.17) and 5-fluorou-
`racil (5-FU, Mw 130.08) were all purchased from
`Aldrich (USA). Coomassie brilliant blue R250
`(Mw 825) was purchased from Fluka A.G.
`(Switzerland) and used after sieving (less than 50
`um). Sodium citrate (analytical grade) and other
`reagents were all commercially available and used
`as received.
`
`2.2. Turbidimetric titration
`
`The interactions of sodium citrate and chitosan
`
`were investigated by turbidimetric titration ac-
`cording to the reported method (Park et al., 1992;
`Mattison et al., 1995). A solution of 0.2 gll
`sodium citrate and 0.2 g/l chitosan was prepared
`at pH 1.0. Titrant (0.01 - 0.2 M NaOH) was deliv-
`ered with a microburette into the solution with
`
`gentle stirring at 20 i 0.2°C, and the pH was
`monitored by a digital pH meter with a precision
`of 1- 0.01. Changes in turbidity were monitored at
`420 nm with an UV—vis spectrophotometer and
`reported as l00—%T which is linearly propor-
`tional to the true turbidity for T> 0.9. The time
`interval between turbidity measurements was ea. 4
`mm.
`
`TEVA EXHIBIT 1018
`TEVA PHARMACEUTICALS USA, INC. V. MONOSOL RX, LLC
`
`RBP_TEVA05022452
`
`TEVA EXHIBIT 1018
`TEVA PHARMACEUTICALS USA, INC. V. MONOSOL RX, LLC
`
`

`
`}(.Z. Slur er al.
`
`International Journal of Plrannaceurics 212 (2001) 19-28
`
`coating of heparin and pectin only retarded drug
`release slightly. The significant difference of cit-
`ratejchjtosan film swelling and model drug release
`profiles in SIP and SGF indicates that these films
`may be useful for site-specific drug delivery in the
`stomach.
`
`Acknowledgements
`
`This work was supported by National Natural
`Science Foundation of China.
`
`References
`
`i999. Evaluation of
`Al-Musa, S., Fara, D.A., Badwan, A.A.,
`parameters involved in preparation and release of drug
`loaded in cross-linked matrices of alginate. J. Control.
`Release 57, 223-232.
`Angelova, N., Manolova, N., Rashkov, I., Maaimova, V.,
`Bogdanova, S., Domard, A., I995. Preparation and prop-
`erties of modified ehjtosan films for drug release. J. Bioact.
`Cornpat. Polym. 10, 285-298.
`Arguelles-Mona], W., Garciga, M., Peniche-Covas, C., 1990.
`Study of the stoiehiometric polyelectrolyte complex. be-
`tween chitosan and carboxylmethyl cellulose. Polym. Bull.
`23, 307-313.
`Chandy, T., Shanna, C.P., 1993. Chitcsan matrix for oral
`sustained delivery of ampicillin. Biomaterials I4, 939-944.
`Chu, C.-I-L, Sakiyama, T., Yano, T., 1995. pH-sensitive
`swelling of a polyelectrolyte complex gel prepared from
`xanthan and chitosan. Biosei. Biotech. Biochem. 59, 7l7—
`7l9.
`Dubois, M., Gilles, K..A., Hamilton, J.K.. Rebers, P.A.,
`Smith, F., 1956. Colorimetric method for determination of
`sugars and related substances. Anal. Chem. 28, 350-356.
`Gupta, K.C., Ravi Kumar, M.N.V., 2000. Drug release behav-
`iors of beads and microgranules of chitosan. Biornaterials
`21, lll5—-llI9.
`Illum, 1...,
`I998. Chitosan and its use as a pharmaceutical
`excipicnt. Pharm. Res. 15, I326--l33I.
`Ikeda, S., Kumagai, H., Sakiyama, T., Chu, C.-I-1., Nalta-
`mura, K.,
`1995. Method for analyzing pH-sensitive
`swelling of ampholcric hydrogels
`- application to a
`polyelectrolyte complex gel prepared from xanthan and
`chitosan. Biosci. Biotech. Biochem. 59, 1422-1427.
`Inouye, K., Machida, Y., Sannan, T., Nagai, T., I988. Buoy-
`ant sustained release tablets based on chitosan. Drug Des.
`Del. 2, I65--175.
`
`I976. Polyeleetrolyte complexes of
`Kihuchi, Y., Noda, A.,
`heparin with chitosan. J. Appl. Polym. Sci. 20, 2561-2563.
`Macleod, G.S., Collett, J.H., Fell, J.T., I999. The potential use
`of mixed films of pectin. chitosan and HPMC for bimodal
`drug release. J. Control. Release 58, 303-3l0.
`Mattison, K.W., Brittain, l.J., Dubin, P.L., 1995. Protein-
`polyelcctrolyte phase boundaries. Biotechnol. Prog. 1|,
`632-637.
`
`Nakatsulra, S., Andrady, A.L., 1992. Permeability of vitamin
`B-12 in chitosan membranes. Effect of cross-linking and
`blending with poly(vinyl alcohol) on permeability. J. Appl.
`Polym. Sci. 44, 17-28.
`I992.
`Park, J.M., Muhoberac, B.B., Dubin, P.L., Xia, 1.,
`Effects of protein charge heterogeneity in protein-polyeIec-
`lrolyte eomplexation. Macromolecules 25, 290-295.
`Patel, V.R., Amiji, M.M., I996. Preparation of characteriza-
`tion of freeze-dried chitosan—poly(ethylcne oxide} hy-
`drogels for site-specific antibiotic delivery in the stomach.
`Pharrn. Res. 13, 588-593.
`Remufiin-Lopez, C., Bodmeier, R., I997. Mechanical, water
`uptake and permeability properties of cross-linked chitesan
`glutamate and alginate films. J. Control. Release 44, 215-
`225.
`
`Shu, X.Z., Zhu, K.J., 2000a. A novel approach to prepare
`tripolyphosphatefehitosan complex beads for controlled
`release drug delivery. Int. J. Pharm. 201, 51-58.
`Shu. X.Z., Zhu, K.J., 2000b. Chitosan_-‘gelatin microspheres
`prepared by modified emulsification and ionotropic gela-
`tion. J. Microencapsulation, in press.
`Thacharodi, D., Rao. K.P.. 1993. Propranolol hydrochloride
`release behavior of cross-linked chitosan membranes. J.
`Chem. Tech. Biotechnol. 58, 177-181.
`Thu, B., Bruheim, P., Espevilt, T., Smidsrod, 0., Soon-Shiong,
`P., Skjdk-Brtek, G., 1997. Alginate polycation mierocap-
`sules
`II. Some functional properties. Biomaterials 17,
`l069—l079.
`
`Yalpani, M., Hall, L.D., 1934. Some chemical and analytical
`aspects of polysaccharide modification. 3. Fonnation of
`branched-chain,
`soluble
`chitosan derivatives. Macro-
`molecules l7, 272-279.
`Yao, K.D., Pcng, T., Goosen, M.F.A., Min, J.M., He, Y.‘{.,
`1993. pH-sensitivity of hydrogel based on complex-forming
`chitosan: polyether interpenetrating polymer network. J.
`Appl. Polym. Sci. 48, 343 -354.
`Yao, K.D., Peng, T., Yin, Y.J., Xu, M.X., I995. Microcap-
`sulesfmierosphcres
`related
`to
`chitosan.
`J.M.S.-REV.
`Maerornol. Chem. Phys. C35, 155—I80.
`Yao, K.D., Liu, J., Cheng, G.X., Lu, X.D., Tu, H.L., I996.
`Swelling behavior of pectin/chitosan complex lilms. J.
`Appl. Polym. Sci. 60, 279 -283.
`Yoshihisa, T., Yoshioka, 1., Segi, N., lkeda, K., l99l. Acid-in-
`dueed and calcium-induced gelation of alginic acid: bead
`formation and pH-dependent swelling. Chem. Phann. Bull.
`39, I072 --I074.
`
`RBP_TEVA05022453
`
`TEVA EXHIBIT 1018
`TEVA PHARMACEUTICALS USA, INC. V. MONOSOL RX, LLC
`
`

`
`riboflavin release significantly, which indicated
`that
`there were possibly other reasons but not
`electrostatic interaction resulting in the above re-
`sult. It was reported that the aqueous solubility of
`alginate under acidic conditions was very poor
`(Yoshihisa et al., 1991), though in SGF the salt-
`bonds between alginate and chitosan dissociated,
`the precipitated alginate layer was still kept on the
`surface of citrate/chitosan fihn, which may limit
`film swelling and prolong drug release.
`
`3.3.5 Model drug nature
`The model drug properties, especially solubility,
`affected their release behavior from citrate,I'chi-
`tosan film seriously. Fig. 12 shows the loading
`percent and loss percent of brilliant blue,
`ri-
`boflavin,
`theophylline and 5-FU during cross-
`
`100
`
`80
`
`60
`
`40
`
`20
`
`..s OGO
`
`:3
`at
`to
`ca
`2cuE
`tn
`.2H
`2:1
`E:1
`0
`
`Time (hours)
`
`ll. The release of riboflavin from polyanion coating
`Fig.
`citratcfchitosan film. The films were prepared with 5.0% w/v
`sodium citrate (pH 7.0) and a cross-linking time of 1
`I1, and
`then dipped into polyanlon solutions (pH 5.5) with dillcrent
`concentrations (w.-'v) for 15 min, (a) pectin, (b) heparin, and (c)
`sodium alginate.
`
`
`
`
`
`Loadingpercent('5)
`
`brilliant blue
`riboflavin
`theophylline
`B - FU
`
`sx
`
`3‘-—§T‘?
`
`0.5
`
`1.0
`
`1.5
`
`2.0
`
`2.5
`
`cross-Ilnlrlng tlme (hours)
`
`Fig. 12. The loading percent and loss percent of model drugs
`during cross-linking process, 5.0% w/v sodium citrate, pH 7.0.
`
`linking process. No obvious loss of brilliant blue
`and riboflavin occurred because they slightly dis-
`solved in water, and even taking 4 h for cross-
`linking the loading percents were both larger than
`97. But for more water-soluble and also smaller
`
`theophylline and 5-FU,
`molecular weight
`loading effieiency decreased greatly with cross-
`linking time, and the loading efficiencies were
`both less than 20% in 2.5 h.
`
`The release of theophylline and 5-FU from
`citratejchitosan film in SIF was very quick,
`most cases more than 90% drug released in 2 h,
`while under the same condition the release per-
`cents of brilliant blue and riboflavin were both
`less than 5%.
`
`4. Conclusions
`
`Novel citrate cross-linked chitosan film was
`
`prepared by dipping chitosan film into citrate
`solution. Citrate/chitosan film possessed pI-l-sensi-
`tive swelling and drug controlled release proper-
`ties. Sodium chloride weakened ionic cross-linking
`and facilitated film swelling and model drug re-
`lease. Sodium citrate solution concentration and
`
`pH during cross-linking process affected film
`swelling and drug controlled release profiles, and
`using higher concentration and lower pH of
`sodium citrate resulted in less swelling and slower
`drug release. The further coating of alginate on
`the surface of citrate/chitosan prolonged ri-
`boflavin release in SGF significantly, while the
`
`TEVA EXHIBIT 1018
`TEVA PHARMACEUTICALS USA, INC. V. MONOSOL RX, LLC
`
`RBP_TEVA05022454
`
`TEVA EXHIBIT 1018
`TEVA PHARMACEUTICALS USA, INC. V. MONOSOL RX, LLC
`
`

`
`X.Z. Shu er a!. _.-‘International Journal of Pharmaceutics 212 (2001) I9-23
`
`
`
`swellingratio
`
`
`
`
`
`Cumulativerelease(%}
`
`Time (hours)
`
`Fig. 8. The influence of sodium citrate concentration on the
`swelling of blank citrate,-'chitosan film in distilled water (a) and
`SIF (b) (cross-linking time 1.0 h, pH 7.0).
`
`3.3.4. Polyanian coating
`In low pH (1.0-3.5), citrate/ehitosan film usu-
`ally dissociated and the model drug released
`quickly (Figs. 5 and 6). For example, in SGF the
`release of riboflavin was usually completed within
`2 h. To prolong the drug release from citrate/chi-
`
`2.4
`
`2.0
`
`1.6
`
`1.2
`
`0.8
`0.0
`
`0.5
`
`1.0
`
`.
`1.5
`
`2.0
`
`2.5
`
`3.0
`
`3.5
`
`4.0
`
`Time (hours)
`
`Fig. 9. The swelling curves of blank chitosan film in sodium
`citrate solution (5.0% wfv) with different solution pHs during
`cross-linking process.
`
`Time (hours)
`
`Fig. 10. The swelling of blank citrate/chitosan film (a) and the
`release of riboflavin from citratefchitosan film (b) in SIF. The
`film was prepared with sodium citrate 5.0% wfv (pH 5.0, 6.0 or
`7.0) and a cross-linking time of |.0 h.
`
`tosan film in SGF, polyanions were further coated
`on the surface of citrate/chitosan film. However,
`the coating of pectin and heparin only slightly
`retarded riboflavin release in SGF (Fig. 11a and
`b). On the other hand,
`the coating of alginate
`greatly prolonged riboflavin release; the time pe-
`riod for 80% riboflavin released was extended
`
`from 1.5 to 2.4 and 3.5 h after being coated with
`0.25 and 0.50% alginate (w/v), respectively.
`From the point of polyelectrolyte interaction,
`the coating of heparin should retard drug release
`in SGF most effectively, because the interaction
`between heparin and chitosan was the strongest
`due to the highest charge density of heparin (car-
`boxylic and sulfonic groups) (Kihuchi and Noda,
`1976). As for pectin and alginate,
`the weakly
`acidic carboxyl groups protonated in SGF (pH
`1.0-1.1), and the electrostatic attractive force be-
`tween pectin (or alginate) and chitosan disap-
`peared (Macleod et al., 1999; Yao et al., 1996).
`
`RBP_TEVA05022455
`
`TEVA EXHIBIT 1018
`TEVA PHARMACEUTICALS USA, INC. V. MONOSOL RX, LLC
`
`

`
`
`
`
`
`Cumulativerelease(%)
`
`Time (hours)
`
`Fig. 5. The leaching of chitosan from blank citratefchitosan
`film in buffered solution with the same ionic strength (0.145
`M). The film was prepared with 1.0 (w/v) sodium citrate (pH
`7.0) and a cross-linking time of 1.0 h. pH [.0 (0.1 M HCI); pH
`3.5, 4.5 and 5.5 (10 mM acetic acid-sodium acetate buffer);
`pH 6.5, 7.4, 8.5 and 9.5 (10 mM phosphate-buffered solution).
`
`could be seen that in SIF the film swelled to the
`
`greatest extent at first 2 h, then the swelling ratio
`decreased gradually to a constant value. It was
`probably caused by some further interpolyelec-
`trolyte bonds being formed as suggested by Ar-
`guelles-Monal et al. (1990), Macleod et al. (1999).
`In the preparation process, sodium citrate solu-
`tion was much in excess and they mainly deter-
`mined the preparation pH value in 5.0-7.0. Fig. 9
`shows the swelling of chitosan film in sodium
`
`
`
`Cumulativerelease(%) NI
`
`9G E
`
`15
`
`20
`
`25
`
`Time (hours)
`
`Fig. 6. The release of brilliant blue from citrate/chitosan film
`in bulTcred solution with the same ionic strength (0.145 M).
`The film was prepared with 1.0% (w/v) sodium citrate (pH 7.0)
`and a cross-linking time of l.0 h. pH 1.0 (0.1 M HCl); pH 3.5.
`4.5 and 5.5 (10 mM acetic acid—sodium acetate buffer); pH
`6.5. 7.4, 8.5 and 9.5 (I0 mM phosphate-buffered solution).
`
`
`
`
`
`Cumulativerelease(7.)Sw°m"9‘am’
`
`
`
`
`
`Time (hours)
`
`Fig. 1. The swelling of blank citrate/chitosan film (a) and the
`release of riboflavin from citrate/chitosan film (b) in different
`concentration NaCl solution (w/v). The film was prepared with
`5.0% (w/v) sodium citrate, pH 7.0 and a cross-linking time of
`1.0 h.
`
`citrate solution (5.0% w/v) with different pHs
`during cross-linking process. More significant
`swelling occurred in higher pH citrate solution.
`The swelling ratio was 2.51, 2.28 and 2.19 in 4 h
`for pH 7.0, 6.0 and 5.0, respectively. At pH 5.0,
`most of the amine groups of chitosan (more than
`95%) ionized (Fig. 1), so more cross-linking struc-
`ture should be formed, which resulted in the
`
`lowest swelling ratio. However, at pH 6.0 and 7.0,
`only part of amines was ionized (ca.78% for pH
`6.0 and ca.12% for pH 7.0) and hence less cross-
`Iinking sites formed.
`Fig. 10 shows the swelling and riboflavin release
`from citrate/chitosan film prepared at pH 5.0, 6.0
`and 7.0, respectively, in SIF. In accordance with
`the above discussion, the increase of preparation
`pH resulted in the increase of swelling ratio and
`drug release slightly (Fig. 10a and b). In 24 h,
`with preparation pH 5.0, 6.0 and 7.0, the swelling
`ratio was 1.85, 1.99 and 2.03, respectively, and the
`riboflavin release was 31.1, 34.2 and 39.7%,
`
`respectively.
`
`TEVA EXHIBIT 1018
`TEVA PHARMACEUTICALS USA, INC. V. MONOSOL RX, LLC
`
`RBP_TEVA05022456
`
`TEVA EXHIBIT 1018
`TEVA PHARMACEUTICALS USA, INC. V. MONOSOL RX, LLC
`
`

`
`X.Z. Sim et al. /International Journal of Pharmaceutics 212 (200!) 19-28
`
`Fig. 3. The morphologies of bottom surface (a) and cross-section (b) of riboflavin (ca. 20% w_."w) loaded citrate.-'chitosan film. The
`film was prepared with 5.0% w/v sodium citrate (pH 7.0) and a cross-linking time of 1 h.
`
`released for brilliant blue and less than 20% for
`
`chitosan in 24 h). The results indicated brilliant
`release
`from citrate/chitosan films was
`mainly controlled by the chitosan leaching, i.e. the
`film dissociation.
`
`3.3.2. The salt concentration in media
`
`Salt usually had a shielding effect on the elec-
`trostatic force, and hence weakened the salt-bond
`between citrate and chitosan. Fig. 7 shows the
`influence of sodium chloride concentration on the
`
`swelling of citrate/chitosan film and the release of
`riboflavin from the film. More significant swelling
`occurred in higher concentrations of sodium chlo-
`ride solution, and hence resulted in a quicker
`riboflavin release. In 24 h, with sodium chloride
`concentrations of 0, 0.45, 0.90 and 1.80% (w/v),
`the swelling ratio was 1.67, 2.07, 2.49 and 2.96,
`respectively (Fig. 7a), and the drug release was
`20.2, 25.8, 27.5 and 39.9%, respectively (Fig. 7b).
`
`3.3.3. Parameters in film preparation
`In our experiments, a cross-linking time of
`more than 0.5 h only slightly limited the swelling
`of citrate/chitosan film and prolonged riboflavin
`release. For example, with 5.0% (w/V) sodium
`citrate (pH 7.0), the equilibrium swelling ratio was
`found to be ca. 1.65 in distilled water and ca. 2.0
`
`in SIF with the cross-linking time extending from
`0.5 to 4.0 h.
`
`On the other hand, under the same conditions,
`the increase of citrate concentration resulted in a
`
`decrease of swelling ratio in distilled water (Fig.
`8a) and in SIF (Fig. 8b) significantly, which indi-
`cated that more cross-linking structure formed in
`the case of high concentration of citrate (Remu-
`iifin-Lépez and Bodmeier, 1997). In Fig. 8b,
`it
`
`
`
`Swellingratio
`
`Fig. 4. The equilibrium swelling ratio of blank citratefchitosan
`film in buffered solution with the same ionic strength (0.145
`M). The film was prepared with 1.0% (w.I'v) sodium citrate (pH
`7.0) and a cross-linking time of 1.0 h. pH 1.0 (0.1 M HCI); pH
`3.5, 4.5 and 5.5 (I0 rnM acetic acid—sodium acetate buffer);
`pH 6.5, 7.4, 8.5 and 9.5 (10 mM phosphate—bulTered solution).
`
`RBP_TEVA05022457
`
`TEVA EXHIBIT 1018
`TEVA PHARMACEUTICALS USA, INC. V. MONOSOL RX, LLC
`
`

`
`Fig. 2. The surface morphology of blank eitratefchitosan film: bottom surface (a) and upper surface (b). The film was prepared with
`5.0% w/v sodium citrate (pH 7.0) and a cross-linking time of I h.
`
`3.3. Factors influencing citrate/chitosan film
`swelling and drug controlled release properties
`
`3.3.1. Media pH
`From Fig. 1, it could be seen that the degree of
`ionization of citrate and chitosan was mainly
`controlled by the solution pH; hence citrate/chi-
`tosan films exhibited pl-I-dependent
`swelling,
`which is shown in Fig. 4. At pH 5.5 and 6.5, the
`swelling ratio was the lowest (2.45—2.50) due to
`significant electrostatic attraction between citrate
`and chitosan. The decrease of pH weakened salt-
`bonds and therefore facilitated the film swelling
`(swelling ratio 2.83 at pH 4.5). Moreover, when
`pH was less than 4.5, citrate/chitosan film swelled
`more significantly and dissociated within 24 h
`because no ionic cross-linking was observed in
`this pH region as revealed by turbidirnetrie titra-
`tion (Fig. I). On the other hand, the increase of
`pH over 6.5 should also weaken salt-bonds, and
`result in a larger swelling ratio (swelling ratio 2.95
`at pH 7.4). However, the further increase of pH to
`8.5 and 9.5 led to the decrease of swelling ratio
`greatly (2.58 and 2.46 for pH 8.5 and 9.5, respec-
`tively). It was usually reported that the swelling of
`polyeleclrolyte complex films (such as pectin/chi-
`tosan film) under weak basic conditions (pH 8-
`10) was very significant (Yao et al., 1996), mainly
`
`resulting from the dissociation of ionic cross-link-
`ing and the repelling interaction between nega-
`tively charged carboxylic groups. But
`experiments, the dissociated citrate at pH 8.5-9.5
`may diffuse out from the films freely and no
`repelling interaction between carboxylic groups
`existed inside the films, which as well as other
`
`factors such as the hydrogen-bonding between
`amine of chitosan, attributed to the shrinkage of
`the film in this pH region. Similar results were
`also observed in the case of tripolyphosphate
`cross-linked cbitosan film (unpublished results).
`Fig. 5 shows the chitosan leaching from the
`citrate/chitosan film in a buffered solution with
`different pHs at the same ionic strength (0.145
`M). At pH 1.0 and 3.5,
`the film dissociated
`quickly (within 5.0 h), while at pH 4.5, the leach-
`ing percent of chitosan was less than 20% in 24 h
`due to the relatively weak electrostatic attractive
`force between citrate and chitosan (Fig. 1), and at
`pH 5.5, 6.5 and 7.4 no leaching of chitosan
`occurred.
`
`The release of brilliant blue with poor water
`solubility from citrate/chitosan films also pos-
`sessed pH-sensitivity (Fig. 6), which was in accor-
`dance with the pH-dependent chitosan leaching
`(Fig. 5) except at pH 4.5 where brilliant blue
`release was faster than chitosan leaching (ca. 67%
`
`TEVA EXHIBIT 1018
`TEVA PHARMACEUTICALS USA, INC. V. MONOSOL RX, LLC
`
`RBP_TEVA05022458
`
`TEVA EXHIBIT 1018
`TEVA PHARMACEUTICALS USA, INC. V. MONOSOL RX, LLC
`
`

`
`X.Z. Shu er al. _.-‘International Journal of Pharmaceutics 212 (2001) 19-28
`
`cH,oH
`
`I-{ZN
`
`chitosan
`
`of the same dissolution medium was added back
`
`to maintain a constant volume. In some cases,
`chitosan leaching from the films was also moni-
`tored by the phenol—su1furic acid color reaction
`(Dubois et al., 1956; Thu et al., 1997).
`
`3. Results and discussion
`
`3.1. The interaction between citrate and chitosan
`
`Citrate is an anion with three carboxylic groups
`and chitosan is a polybase (Scheme 1), the charge
`densities of citrate and chitosan are mainly con-
`trolled by solution pH (Fig. 1). Due to the weak
`acid characteristic of citric acid, under neutral and
`weakly acidic conditions the decrease of the solu-
`tion pH resulted in a significant decrease of the
`degree of ionization of citrate. At low pH (less
`than ca. 4.1),
`the ionization of carboxyl groups
`was normally depressed (the degree of ionization
`
`Degreeofionization
`
`Fig. 1. The turbidity titration curves of citrate.-‘chitosan solu-
`tion at 420 nm (0.2 g.-'1 citrate and chitosan. respectively), and
`the degree of ionization curves of citrate and chitosan.
`
`was usually less than 0.3), i.e. less than one nega-
`tive charge was carried by one citrate. For chi-
`tosan (a weak polybase),
`the opposite was the
`case as the ionization of amine groups decreased
`greatly when the solution pH increased above 6.0
`(around the pit, of chitosan 6.3) (Yalpani and
`Hall, 1984), and at pH higher than 7.5 usually less
`than 10% of amine groups were ionized.
`The turbidimetric titration curve of sodium cit-
`
`rate/chitosan is also shown in Fig. 1, which is in
`accordance with the pH-dependent charge density
`of citrate and chitosan. At low pH (l.0—4.0), the
`solution was optically clear due to the low charge
`density of the citrate. The turbidity increased
`greatly and the solution began to separate into
`two phases when pH increased over 4.3. This
`could be attributed to the significant charge densi-
`ties of citrate and chitosan in this pH region.
`Further increase of solution pH over ca. 6.3 led to
`the decrease of the charge density of chitosan
`greatly, and hence the decrease of turbidity signifi-
`cantly. The lowest value of turbidity was observed
`at pH ca. 7.6, and then turbidity increased at pH
`values over 7.6, which was attributed to the poor
`solubility of chitosan in this pH region (Shu and
`Zhu, 2000b).
`
`3.2. Morphology of citrate/chitosan films
`
`The surface morphologies of citrate/chitosan
`films are shown in Fig. 2. The bottom surface of
`citratejchitosan films was very smooth (Fig. 2a)
`while the upper surface was relatively rough (Fig.
`2b), which was in accordance with the morphol-
`ogy of chitosan films before cross-linking (pictures
`not shown). Sodium citrate concentration, pH
`and cross-linking time had little effect on the
`surface morphology of citrate/chitosan films. The
`cross-section of the citrate/chitosan films was very
`integral and dense (pictures not shown).
`The surface and cross-section morphologies
`changed significantly due to the incorporation of
`model drugs into the citrate/chitosan film. For
`example, large pores were observed on both the
`bottom and upper surface of riboflavin loaded
`citrate.-‘chitosan films (Fig. 3a), and the cross-sec-
`tion was very rough and many deficiencies were
`observed (Fig. 3b).
`
`RBP_TEVA05022459
`
`TEVA EXHIBIT 1018
`TEVA PHARMACEUTICALS USA, INC. V. MONOSOL RX, LLC
`
`

`
`Potentiometric titration was performed accord-
`ing to the method reported by Ikeda et al. (1995)
`to evaluate the pH-dependent ionization degree of
`chitosan and citrate, respectively. Sodium citrate
`solution (100 ml; 10 mM) or 0.1% (w/v) chitosan
`solution were neutralized by adding 0.1 M HCl or
`NaOH,
`respectively, at 20 1 0.2°C with a mi-
`croburette in a nitrogen atmosphere, and the solu-
`tion pH was monitored by a digital pH meter with
`a precision of i 0.01.
`
`2.4. Preparation of cross-flaked chitosan film
`
`Chitosan films were produced by a casting/sol-
`vent evaporation technique. Chitosan solutions
`(4.0%, w/v) containing model drug (brilliant blue,
`riboflavin, theophylline or 5-FU, 1.0% w/v) were
`prepared by dissolving chitosan and the model
`drug (or dispersing) in 4.0% (w/v) acetic acid. The
`above solutions (40 ml) were sonicated,
`left to
`stand until trapped air bubbles were removed, and
`poured on a glass plate (casting area, 10 x10
`cml). The films were dried for 48 h in an oven at
`37°C, then further dried under vacuum at room
`
`temperature until constant weight. The dried films
`were cut into 2 x 2 cm’ test sections. The thick-
`ness of the dried films were determined to be ca.
`
`100 um.
`Citrate cross-linked chitosan films were pre-
`pared by soaking the chitosan films (ca. 50 mg) in
`an aqueous solution of sodium citrate (100 ml) at
`4°C. The cross-linking conditions were 1.0-- 10.0%
`(w/v) sodium citrate; solution pH, 5.0-7.0; cross-
`linking time, 0.5-4 h. The citrateichitosan films
`formed were then washed with distilled water, put
`on a glass plate and oven-dried at 37°C for 48 h,
`and then dried under vacuum at room tempera-
`ture until reaching a constant weight.
`In some cases, wet citrate/chitosan films were
`further coated with polyanions (pectin, alginate or
`heparin) by dipping the films into 100-ml polyan-
`ion aqueous solution (concentration 0.25 or 0.5%
`w/v) at pH 5.5 for 15 min.
`The model drug loss during cross-linking pro-
`cess was determined by measuring the UV— vis
`absorption, 590 nm for brilliant blue, 444 nm for
`
`for 5-FU. Model drug loading percent = (1 —the
`drug loss/the drug given) x 100%.
`
`2.5 Morphology observation
`
`The surface and cross-sectional morphologies
`of citrate/chitosan films were examined using
`scanning electron microscopy (SEM, S-S90, Hi-
`tachi). Cross-sectional samples were prepared by
`fracturing films in liquid nitrogen. Prior to obser-
`vation, samples were mounted on metal grids,
`using double-sided adhesive tape, and coated with
`gold under vacuum before observation.
`
`2.6 Swelling ratio measurement
`
`Blank citrate/chitosan films (ca. 50 mg) were
`suspended in glass bottles containing 250 ml of
`media, and incubated on a shaking water-bath at
`37°C, 50 rpm. At an appropriate time interval, the
`films were taken out, and the excess water was
`
`removed carefully with filter paper from the film
`surface,
`and then weighed immediately. The
`swelling ratio = W,/ W", where W, was the film
`weight at
`time t and W0 was the initial
`weight, was calculated. The media for the swelling
`studies were either 0.1 N HCl (pH 1.0), 10 mM
`acetic acid—sodium acetate (pH 3.5, 4.5 and 5.5),
`or 10 mM phosphate-buffered solution (pH 6.5,
`7.4, 3.5 and 9.5). The ionic strength of above
`buffered solutions was carefully adjusted to 0.145
`M by adding an appropriate amount of sodium
`chloride. Different concentrations of sodium chlo-
`
`ride solution (0, 0.45, 0.9, and 1.8% w/v), and
`enzyme-free simulated gastric fluid (SGF) and
`simulated intestinal fluid (SIF) (USP XXII) were
`also used for test.
`
`2.7. Release studies
`
`The model drug release from citrate/chitosan
`film was performed under the same conditions
`described in the swelling studies. At appropriate
`time intervals, the solutions were withdrawn and
`the content of the model drugs were determined
`by measuring UV--vis absorption at
`the wave-
`lengths desc