Full Paper

729

Summary: The present work is focused on investigating the behavior of controlled drug release poly(N-isopropylacrylamide) (PNIPA) hydrogels in the presence of b-cyclodextrin (b-CD). For this purpose, three types of NIPA hydrogels with b-CD moieties were synthesized with different architectures according to our previous studies. An anti-cancer drug (chlorambucil, CLB), which can form an inclusion complex with b-CD, was selected for loading and in vitro release studies. The drug was loaded into hydrogels via a swelling method. DSC was used to study the interactions between the CLB molecules and the polymers. The results indicate that the CLB-polymer interactions are at the molecular level. Loading CLB into these polymers can result in an evident decrease in the glass transition temperature (Tg), and the variation of Tg (DTg) depends on the structures of the polymers and their b-CD content. The controlled release experiments show that the presence of b-CD can markedly enhance CLB release from shrunken PNIPA hydrogels and increase the ratio of CLB released in total drug loading content.

Release prole of CLB from hydrogels 1a-c and 4 at pH 1.4 and 7.4, at 37 8C.

Release of Chlorambucil from Poly(N-isopropylacrylamide) Hydrogels with b-Cyclodextrin Moieties

Yu-Yang Liu,* Xiao-Dong Fan, Hui Hu, Zhong-Hua Tang
Department of Applied Chemistry, School of Science, Northwestern Polytechnic University, Xian, 710072, Peoples Republic of China E-mail: yyliu666@yahoo.com.cn

Received: March 29, 2004; Revised: June 2, 2004; Accepted: June 8, 2004; DOI: 10.1002/mabi.200400037 Keywords: controlled release; b-cyclodextrin; drug delivery systems; hydrogels; N-isopropylacrylamide

Introduction
The synthesis and characterization of polymeric controlledrelease drug delivery systems has attracted much attention from many polymer scientists during recent years because of their potential applications.[110] Among these controlled drug release systems, PNIPA has been widely investigated as a carrier, owing to its thermosensitivity and this leads to a temperature-modulated drug release.[410] PNIPA homopolymer exhibits a lower critical solution temperature (LCST) of around 32.0 8C in aqueous solution, and its hydrogel swells and shrinks in water below and above this temperature. To further expand the applications of PNIPA, chemical modications have been carried out by incorporating another functional component into the PNIPA chain structure. For example, Dong et al.[7] synthesized a
Macromol. Biosci. 2004, 4, 729736

reversible hydrogel by g-irradiation of solutions of Nisopropylacrylamide (NIPA) and bisvinyl-terminated polydimethylsiloxane. The heterogeneous hydrogel exhibits the same LCST as the homopolymer of NIPA. The zero-order release of progesterone from the hydrogel was observed. Okano et al.[8] synthesized a poly[(N-isopropylacrylamide)-co-(butyl methacrylate)] (P(NIPA-co-BMA)) hydrogel, which could form a shrunken layer on the hydrogel surface (a skin layer). The drug release was governed by diffusion below its LCST, while above its LCST, drug release was completely stopped by the skin layer. Chung et al.[9,10] prepared another on-off drug release system using P(NIPA-b-BMA) as the carrier for adriamycin. The block copolymer can form micelles with a core-shell structure in aqueous solution below the LCST of PNIPA. At temperatures below the LCST of PNIPA, drug release is at a
2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

DOI: 10.1002/mabi.200400037

730

Y.-Y. Liu, X.-D. Fan, H. Hu, Z.-H. Tang

minimum, with a value of less than 10%. However, upon increasing the temperature above the LCST, an accelerated release of adriamycin is obtained. It is well-known that the incorporation of cyclodextrins (CDs) into polymeric drug release systems can inuence the mechanisms by which the drug is released.[1117] A CD molecule has a hydrophobic internal cavity, which enables it to include guest drug molecules.[1121] Therefore, no covalent bonds exist between CD and its guest, and the formation and dissociation of an inclusion complex are actually dynamic processes. A drug molecule included in the cavity of CD could be easily dissociated upon dilution, or replaced by a more suitable guest.[19,21] Evidently, the incorporation of CDs into polymeric drug delivery systems could change the drug-polymer interaction and the drug state in the polymer matrix. As a result, the mechanisms of drug release may be modied.[11] For instance, Guo and Cooklock incorporated CDs into buccal patches composed of poly(acrylic acid), poly(isobutylene) and poly(isoprene) for enhancing buprenorphine release[13] and Sreenivasan introduced b-CD into poly(vinyl alcohol) (PVA) for retarding the release of salicyclic acid.[14] Therefore it is expected that the incorporation of CDs into PNIPA may also modify its controlled drug release behavior. Unfortunately, to date, there are few studies being conducted in this direction. Our aim is to investigate the potential of CDs to control drug release from PNIPA. For this purpose, we have synthesized and characterized three hydrogels[2224] and two linear polymers[25] containing NIPA and b-CD moieties with different architectures. In this paper, these hydrogels were loaded with an anti-cancer drug, chlorambucil (CLB), which is capable of forming complexes with b-CD.[26] The polymer-drug interactions and the release behavior of CLB from the hydrogels at pH 1.4 and 7.4 were studied in vitro.

Hydrogel Preparation Preparation of NIPA/MAH-b-CD Hydrogels The NIPA/MAH-b-CD hydrogels were synthesized by the copolymerization of NIPA with vinyl b-CD monomer (MAHb-CD) as described in our previous report.[22] Briey, MAH-bCD was synthesized by the reaction of 5.68 g of b-CD (0.005 mol) with 4.90 g of MAH (0.05 mol) in 30 ml of dimethylformamide at 80 8C for 10 h. As a result, the MAH-bCD monomer with b-CD[COCH CHCOOH]5 composition was obtained, and its structure was characterized by IR, 13C NMR and elemental analysis. Then, the NIPA/MAH-b-CD hydrogels were copolymerized via NIPA and MAH-b-CD with different feed compositions in aqueous solution at 20 8C using an APS and SBS redox system as the initiator. The hydrogel obtained was cut into thin disks of 12 mm diameter. The samples were immersed in distilled water, which was changed every 12 h for 6 d in order to remove the unreacted monomer. Later, they were dried under ambient conditions for 2 d and at 80 8C for 48 h. The compositions of the hydrogels synthesized were determined by elemental analysis, and are listed in Table 1. Preparation of NIPA/MAH-b-CD-EPI Hydrogels The NIPA/MAH-b-CD-EPI hydrogels were prepared by the copolymerization of NIPA and b-CD macromonomer (MAHb-CD-EPI) as reported in our previous studies.[23] Briey, the MAH-b-CD-EPI macromonomer (COOH content: 2.01 mmol g1) was synthesized by the reaction of MAH (3.0 g) with the b-CD-EPI resin (7.5 g) which was obtained by the reaction of b-CD (21.0 g) with EPI (17.1 g) in alkaline solution. The b-CD contents of b-CD-EPI and MAH-b-CDEPI were determined by a colorimetric method[27,28] and were 45.1% and 33.7%, respectively. Finally, the NIPA/MAH-bCD-EPI hydrogels were synthesized from NIPA and MAH-bCD-EPI in aqueous solution at 20 8C using an APS and SBS redox system as the initiator. The hydrogel obtained was cut into thin disks of 12 mm diameter. The purication of NIPA/ MAH-b-CD-EPI hydrogels was similar to that of the NIPA/ MAH-b-CD hydrogels. Their compositions are listed in Table 1. Preparation of PNIPA/b-CD IPN The PNIPA/b-CD interpenetrating polymer network (IPN) was prepared via dispersing NIPA monomer containing N,Nmethylene-bis(acrylamide)(crosslinker) into a b-CD-based polymer network (b-CD-EPI-PVA), which was polymerized and crosslinked in situ, according to the method described in our previous publication.[24] For this purpose, b-CD-EPI-PVA exhibiting a high content of b-CD with a larger swelling ratio was rst prepared by crosslinking b-CD with EPI in the presence of a small amount of PVA. Briey, b-CD was rst dissolved in a PVA solution in a beaker, and then sodium hydroxide solution (40 wt.-%) was added. After the mixture was kept at 5060 8C for 30 min, the EPI needed (12:1 EPI:bCD molar composition) was added dropwise under vigorous stirring. The stirring was stopped prior to gelation and the system was sealed and kept at the same temperature for 24 h. The b-CD content in b-CD-EPI-PVA was determined by a
2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Experimental Part
Materials and Methods NIPA was purchased from Acros, UK (99% purity). CLB was obtained from Fluka Chemika. b-CD was acquired from Northwestern Geological Institute, China and was puried twice by recrystallization from water before use. PVA (degree of polymerization 1 750 50) was purchased from Tianda Chemical Plant, Tanjin, China. All other reagents, including maleic anhydride (MAH), ammonium persulfate (APS), sodium bisulte (SBS), epichlorohydrin (EPI) and acrylic acid (AAc) were analytical grade and were made in China. They were used as received without further purication. Elemental analyses were carried out on a Vario EL III Instrument (Germany). NMR measurements were conducted on a Varian INOVA-400 spectrometer (USA) at room temperature using D2O as solvent. UV-visible spectra were recorded on a spectrophotometer (UV-1200 model).
Macromol. Biosci. 2004, 4, 729736 www.mbs-journal.de

Release of Chlorambucil from Poly(N-isopropylacrylamide) Hydrogels with b-Cyclodextrin Moieties

731

Table 1.

Characteristics of PNIPA hydrogels with b-CD moieties. Hydrogel code NIPA wt.-% b-CD wt.-% NIPA/COOH mol/mol pH 1.4 ESR at 33 8C g/g pH 7.4 8.8 15.6 20.3 4.0 7.8 11.3 4.0 8.6

Hydrogel types

NIPA/MAH-b-CD NIPA/MAH-b-CD-EPI IPN NIPA/AAc PNIPA

a)

1a 1b 1c 2a 2b 2c 3 4 5

90.9 82.6 78.9 82.3 75.6 68.4 57.6

6.4 12.1 14.7 6.0 8.2 10.6 32.3

96.6/3.4 93.2/6.8 91.5/8.5 95.3/4.7 93.2/6.8 90.5/9.5 97.0/3.0a)

0.5 0.5 0.6 0.7 0.9 1.1 4.0 0.5

Molar feed composition of NIPA to AAc was 97:3.

colorimetric method[27,28] and was found to be 76.1 wt.-%. For comparison, a pure PNIPA and NIPA/AAc hydrogel was synthesized using N,N-methylenebis(acrylamide) as a crosslinker. Swelling Measurements The swelling ratios (SR) of the hydrogels were measured after they were swollen to a desired equilibrium state. They were then carefully removed from the solution, wiped with a lter paper to remove any free water on the surface, and then weighed. The SR (g/g) was calculated using the following equation: SR we wd =wd 1

the old solution was replaced by a fresh buffer solution of pH 7.4. The concentrations of CLB released were analyzed by spectrophotometry at 211.5 nm (pH 1.4) and 257.0 nm (pH 7.4), respectively. All the resulting solutions were kept at 37 8C for 48 h prior to measurements. All release measurements were carried out in triplicate for each hydrogel and the average values were plotted. DSC Analysis DSC (MDSC 2910, TA Instruments) measurements were used to determine the glass transition temperature (Tg) of CLBloaded or unloaded polymers. The scan rate was 20 K min1 within the temperature range 20200 8C. Every sample was subjected to two scans. The Tg was estimated from the trace of the second run and obtained at the midpoint of the special heat increment. All DSC measurements were carried out in duplicate for each sample.

where wd and we are the weights of the dry and wet samples, respectively. When a hydrogel reaches its swelling equilibrium state under a xed condition, its SR is called ESR. All ESR measurements were made in triplicate for each sample. CLB Loading into the Hydrogels Loading CLB into the hydrogel disks was performed in a CLB methanol solution of 0.6 wt.-%. After the hydrogels were swollen in the methanol solution of CLB for 24 h, they were carefully taken out and washed with methanol to remove any free CLB on the surface. They were then dried under ambient conditions for 1 d, and at room temperature for 4 d in a vacuum oven. The CLB loading capacity of each hydrogel was calculated using the following equation: Loading capacity % 100weight of drug in the polymer= weight of the dry polymer 2

Results and Discussion

Drug Loading into Hydrogels
To investigate the potential of b-CD to control drug release from PNIPA hydrogels, we synthesized and characterized three novel hydrogels by the incorporation of b-CD moieties into PNIPA with different architectures.[2224] These hydrogels were NIPA/MAH-b-CD and NIPA/MAHb-CD-EPI copolymers, and PNIPA/b-CD IPN, respectively. The NIPA/MAH-b-CD and the NIPA/MAH-b-CD-EPI hydrogels can respond to changes in pH and temperature,[22,23] while the IPN can keep its volume transition temperature (Tv) at PNIPAs LCST.[24] The compositions and the ESR values of these hydrogels are shown in Table 1, and their structures and stimuli-responsiveness were reported previously.[2224] Here, using the hydrogels as carriers for CLB, which can form an inclusion complex with b-CD,[26] the inuence of b-CD on the controlled drug
2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

CLB Release Studies At 37 8C, CLB-loaded disks were immersed in 5.0 ml of a pH 1.4 buffer solution with an ionic strength of 0.1 mol l1. In a special interval, the whole medium was removed and replaced by 5.0 ml of fresh buffer solution. When a samples cumulative release time in the solution of pH 1.4 reached 24 h,
Macromol. Biosci. 2004, 4, 729736 www.mbs-journal.de

732

Y.-Y. Liu, X.-D. Fan, H. Hu, Z.-H. Tang

Table 2. Characteristics of CLB-loaded PNIPA hydrogels with b-CD moieties. Hydrogel code Loading capacitya) % 1a 1b 1c 2a 2b 2c 3 4 5
a)

CLB/b-CD mol/mol 7.3/1.0 2.6/1.0 2.3/1.0 7.7/1.0 4.4/1.0 3.6/1.0 0.9/1.0

DTg DTg/Loading capacity 8C 17.1 10.6 6.3 17.5 17.2 20.2 20.2 27.7 11.1 1.4 1.2 0.7 1.8 1.8 1.7 2.7 2.3 2.2 Figure 1. DSC thermogram of CLB.

12.4 8.6 9.0 12.3 9.6 10.3 7.5 11.9 5.1

Refers to CLB weight content (mg) in 100 mg of dried gel.

release behavior was investigated. Owing to the hydrolysis of CLB in aqueous solution,[29] loading CLB into the polymer matrix was conducted in methanol solution, and the loading capacity of a hydrogel is presented as the content of CLB (mg) in 100 mg of dried gel. The CLBs loading capacity in different hydrogels and the corresponding CLB/b-CD molar ratios are shown in Table 2. In addition, it was found that the UV absorptions of CLB in aqueous solution at 211.5 nm (pH 1.4) and 257.0 nm (pH 7.4) decreased with the amount of time at 37 8C, but remained stable after 48 h. Hence, all solutions of CLB were kept at 37 8C for 48 h prior to the UV measurements.

DSC Study on CLB-Loaded Hydrogels

A drug-loaded polymer is a multicomponent system. The interactions among these components, especially between drug molecules and polymer chains, may exert an important inuence on the drugs release behavior. However, to date, little attention has been paid to this area. In this research, we have used DSC to study the interactions between CLB and polymeric chains by the determination of the Tg alternation of polymer matrices. Figure 1 and 2 are DSC thermograms of pure CLB, and PNIPA-based hydrogels and their loaded samples, respectively, which were obtained under the same conditions. As shown in Figure 1, the DSC thermogram of CLB shows a characteristic endothermic peak at 69.6 8C, corresponding to its melting temperature, whereas the CLB entrapped in polymer matrices shows no phenomena in the same temperature region, and leads to an evident decrease in Tg of polymer matrices. This result demonstrates that there is no pure CLB phase in the loaded samples, and that CLB behaves as a typical plasticizer. The plasticization effect of the drug was also observed by Yuksel et al.[30] According to a study by Dubernet,[31] when drug molecules and a polymer develop strong interactions, they lead to the polymer matrix being plasticized, and as a result there is no fusion peak of the drug on the DSC curve. When drugMacromol. Biosci. 2004, 4, 729736 www.mbs-journal.de

polymer interactions are weak, even if the drug remains molecularly dispersed in the polymer matrix due to the high viscosity of the medium, the polymer can still conserve its Tg value and when the drug is physically dispersed in a polymer matrix in the form of crystals, it can present its melting behavior. Thus, according to the DSC results in Figure 2 (data shown in Table 2), we can reasonably conclude that strong molecular interactions exist between CLB molecules and polymer chains indeed, resulting in samples 1a-c being plasticized. To further understand CLBs plasticization effect, we have used the ratio of DTg/loading capacity (DTg/LC, shown in Table 2) to evaluate the role of CLB due to the difference in the loading capacity of CLB for different hydrogels (see Table 2). Obviously, the higher the value of DTg/LC, the stronger CLBs plasticization effect can be observed. As can be seen from Table 2, the ratios of DTg/LC for the loaded 1a-c samples are 1.4, 1.2 and 0.7, respectively, whereas the ratios of DTg/LC for reference samples 4 and 5 without b-CD are 2.3 and 2.2, respectively. This result indicates that the plasticization effect of CLB increases with increasing NIPA content. As is well known, to achieve a good plasticization effect there must be strong interactions between the polymer and the plasticizer.[3032] This suggests that a good compatibility and a strong molecular interaction exists between the CLB molecules and the PNIPA chains. Consequently, the CLB molecules are embedded among the macromolecular chains, and hence weaken the secondary bonds between polymer chains, depressing the Tg of the polymer matrix.[30,31] It was also observed that the ratio of DTg/LC of the loaded 1a-c samples decreased with increasing b-CD content. This may be attributed to the inclusion complex of b-CD with CLB. Because of the inclusion effect, some CLB molecules may be trapped in the cavities of b-CDs, and cannot plasticize polymer matrices efciently. Huang et al.[26] investigated the inclusion of parent b-CD with CLB, and they demonstrated that b-CD and CLB can form a 1:1 inclusion complex. Therefore,
2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Release of Chlorambucil from Poly(N-isopropylacrylamide) Hydrogels with b-Cyclodextrin Moieties

733

Figure 2. DSC thermograms of hydrogels studied and their CLB-loaded samples: (a) Hydrogels 1a-c; (b) Hydrogels 2a-c; (c) Hydrogel 3; (d) Reference hydrogels 4 and 5.

we can conclude that there are strong molecular interactions between CLB and the NIPA moiety and the CLB/b-CD inclusion complex in the loaded samples of series 1. Hydrogels 2a-c were prepared based on hydrogels 1a-c. They showed a higher swelling ratio at room temperature, excellent mechanical strength during a rapid deswelling process, and better responding sensitivity when subjected to external stimuli.[23] Concerning the loaded 2a-c samples, as shown in Table 2, their ratios of DTg/LC (corresponding to 1.8, 1.8 and 1.7, respectively) are dependent on NIPA content, but are lower than those of the loaded reference samples of 4 and 5. This may be caused by the lower b-CD content of 2a-c compared with 1a-c. Interestingly, although hydrogel 3 possesses the highest b-CD concentration (CLB/ b-CD molar ratio <1) of the hydrogels studied, the CLB molecules can plasticize the PNIPA network more efciently (DTg/LC 2.7). This is because the b-CD polymer exhibits no evident Tg in this temperature region and the Tg of the IPN is from the PNIPA network only.[24] This result
Macromol. Biosci. 2004, 4, 729736 www.mbs-journal.de

indicates that each polymer network of the IPN can retain its individual interactions with the CLB molecules.

Release Studies CLB Release from Copolymeric Hydrogels

All release experiments were carried out under the conditions of pH 1.4 and 7.4, and 37 8C. Figure 3 exhibits the cumulative amounts of CLB released from the loaded samples 1a-c. As can be seen from Figure 3, the release rates of CLB are faster at pH 7.4 than at pH 1.4. At pH 7.4, there is a burst release for CLB. This is attributed to the pHdependent thermosensitivity of 1a-c. As they contain COOH groups, their thermosensitivity can be affected by the pH of the solution.[22,23,3335] At pH 1.4, samples 1a-c retain the original thermosensitivity of PNIPA. They are therefore in a shrunken state at 37 8C. However, at pH 7.4, they swell fast due to the ionization of the COOH groups, leading to fast CLB release. This release character can also be observed in other pH-dependent release systems.[3539]
2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

734

Y.-Y. Liu, X.-D. Fan, H. Hu, Z.-H. Tang

Figure 3. Release prole of CLB from hydrogels 1a-c and 4 under the conditions of pHs 1.4 and 7.4, and 37 8C.

It can be seen from Figure 3 that, at pH 1.4, the release rate of CLB from 1a-c increases with decreasing b-CD content, and the corresponding cumulative amounts of CLB released are 18.9%, 36.6% and 50.5% for 24 h, respectively. This result indicates that the CLB release from the PNIPAbased hydrogels is a function of b-CD content. To understand the role of b-CD in the controlled drug release system, a b-CD-free hydrogel 4 (P(NIPA-co-AAc)) containing 3 mol-% of AAc was synthesized. It exhibits a similar ESR to hydrogel 1a at pHs 1.4 and 7.4 (see Table 1). Surprisingly, even though 4 and 1a-c exhibit a shrunken state at pH 1.4, and have similar swelling ratios (see Table 1), their release behaviors for CLB present visible differences. It can be seen from the release curve of 4 that, although 14.2% of CLB can be released at pH 1.4 for 9.5 h, less CLB is released from this hydrogel afterwards. The CLB release from 1a, however, remains relatively stable for 24 h. This result may be attributed to the existence of the cavity of b-CD as well as its inclusion effect. As the cavities of b-CDs can act as channels for solution diffusion,[11] they lead to an increase in the diffusion rate of CLB. Owing to the lack of a b-CD component in 4, the CLB located near the surface was released rst, and the CLB located inside cannot diffuse out. This is the main reason why after 9.5 h less CLB release from sample 4 can be observed while the CLB release from 1a-c is sustained over the whole period studied (24 h) at pH 1.4. In addition, the formation of the CLB/b-CD inclusion complex may be facile to CLB release from the hydrogels. Because the formation and dissociation of an inclusion complex of CD with a guest molecule is a dynamic process, the drug molecules can be easily released from their complexes with a low binding constant by dilution.[1619] Since the b-CD/CLB inclusion complex exhibits a lower binding constant at pH 1.1 and 7.2 (0.71 103 l mol1 and 6.1 103 l mol1, respectively)[26] the incorporation of the b-CD component into NIPA hydrogels can modify the interaction of NIPA chains with CLB
Macromol. Biosci. 2004, 4, 729736 www.mbs-journal.de

molecules, leading to an easier release of CLB. These aspects can explain why the presence of b-CD can enhance the release of CLB molecules at pH 1.4 in Figure 3. It was also found that, after the loaded 1a-c samples were kept in solution for 29.5 h (at pH 1.4 for 24 h and at pH 7.4 for 5.5 h), less CLB was released. The corresponding ratios are regarded as the maximum ratios of CLB released. The maximum ratios of CLB released from 1a-c were 65.2%, 85.2% and 97.2%, respectively, indicating again that the incorporation of b-CD into NIPA hydrogels can effectively enhance the release of CLB. Figure 4 presents the release behaviors of CLB from 2a-c. Compared with hydrogels 1a-c, hydrogels 2a-c have a similar release pattern for CLB due to their similar structures.[22,23] Their maximum ratios of CLB released also increased with an increase in the b-CD content of the polymers. This result illustrates that the incorporation of b-CD into PNIPA by copolymerization can indeed promote CLB release. However, their maximum ratios of CLB released were 69.9%, 74.2% and 86.1%, respectively. The latter two values are lower than the corresponding values of 1b-c. This may be attributed to their low b-CD content. However, at pH 1.4, the CLB release from 2a-c is faster than that from the corresponding 1a-c, due to 2a-c possessing a relatively higher swelling ratio, and thus making CLB diffusion much easier (see Table 1).

CLB Release from IPN Hydrogels

Hydrogel 3 is an IPN-based PNIPA hydrogel with a b-CD moiety. Its volume phase transition temperature is still sensitively kept at about 33 8C. As for an IPN system, there is no chemical bonding between the two polymer networks, and each polymer network can retain its individual properties.[5,24] At 37 8C, the ESR value for hydrogel 3 was about 4 (see Table 1). Figure 5 presents the release prole of CLB from 3. Compared to 1a-c and 2a-c, the release

Figure 4. Release prole of CLB from hydrogels 2a-c under the conditions of pHs 1.4 and 7.4, and 37 8C.
2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Release of Chlorambucil from Poly(N-isopropylacrylamide) Hydrogels with b-Cyclodextrin Moieties

735

glass transition temperature (Tg), and the variation of Tg depended on the structure of the polymer and its b-CD content. (3) The study of drug release showed that the release behaviors of CLB depend strongly on the structure of the hydrogel. In the case of the copolymers NIPA/ MAH-b-CD and NIPA/MAH-b-CD-EPI, the presence of b-CD markedly enhanced the CLB release from the shrunken PNIPA hydrogels and increased the maximum ratios of CLB released in total drug loading content; at pH 7.4, they show a burst release behavior. The release rate of CLB from IPN was faster at pH 1.4 than at pH 7.4.

Figure 5. Release prole of CLB from hydrogel 3 under the conditions of pHs 1.4 and 7.4, and 37 8C.

character of CLB from 3 shows a distinct difference. The release rate of CLB from 3 is faster at pH 1.4 than at pH 7.4. Under these conditions, the corresponding cumulative amounts of CLB released were 51.9% and 36.9% for 24 h, respectively. This may be attributed to the lower binding constant of the b-CD/CLB inclusion complex at pH 1.4. On the other hand, compared to hydrogels 1a-c and 2a-c, we could conclude that a higher swelling ratio of the IPN may not result in a higher release rate at pH 1.4. This may be caused by the unique structure of IPN, which leads to a strong interaction of the drug molecules with the NIPA moiety (see Table 1). Therefore, this result implies that the swelling ratio of a hydrogel is not always an important factor with regard to drug release. In addition, it was found that after sample 3 was kept in buffer solution of pH 1.4 for 24 h and then placed into a buffer solution with pH 7.4, the release rate of CLB showed no marked change.

Acknowledgement: This work was supported by the National Nature Science Foundation of China (No.20374040) and the Doctorate Foundation of Northwestern Polytechnic University.

Conclusions
(1) Three types of NIPA hydrogels containing b-CD moieties were synthesized with different architectures. The hydrogels were NIPA/MAH-b-CD and NIPA/ MAH-b-CD-EPI copolymers, and PNIPA/b-CD IPN, respectively. The NIPA/MAH-b-CD and the NIPA/ MAH-b-CD-EPI hydrogels can respond to changes of pH and temperature. The IPN can still retain its volume transition temperature (Tv) at PNIPAs LCST. An anticancer drug, CLB, which can form inclusion complexes with b-CD was selected for drug loading and in vitro release studies. Novel controlled drug release systems were prepared by loading CLB into three types of hydrogels via a swelling method. (2) DSC measurements indicated that the CLB-polymers interactions were at the molecular level. Loading CLB into these polymers resulted in an evident decrease in
Macromol. Biosci. 2004, 4, 729736 www.mbs-journal.de

[1] M. N. V. R. Kumar, N. Kumar, A. J. Domb, M. Arora, Adv. Polym. Sci. 2002, 160, 54. [2] A. S. Hoffman, Adv. Drug Deliver. Rev. 2002, 43, 3. [3] P. Gupta, K. Vermani, S. Garg, Drug Discov. Today 2002, 7, 459. [4] A. Kikuchi, T. Okano, Drug Deliver. Rev. 2002, 54, 53. [5] J. Zhang, N. A. Peppas, Macromolecules 2000, 33, 102. [6] H. Feil, Y. H. Bae, J. Feijen, S. W. Kim, J. Membrane Sci. 1991, 64, 283. [7] L.C. Dong, A. S. Hoffman, J. Control. Release 1990, 13, 21. [8] T. Okano, Y. H. Bae, H. Jacobs, S. W. Kim, J. Control. Release 1990, 11, 255. [9] J. E. Chung, M. Yokoyama, M. Yamato, T. Aoyagi, Y. Sakurai, T. Okano, J. Control. Release 1999, 62, 115. [10] J. E. Chung, M. Yokoyama, T. Okano, J. Control. Release 2000, 65, 93. [11] D. C. Bibby, N. M. Davies, I. G. Tucker, Int. J. Pharm. 2000, 197, 1. [12] D. C. Bibby, N. M. Davies, I. G. Tucker, Int. J. Pharm. 1999, 187, 243. [13] J. H. Guo, K. M. Cooklock, Drug Del. Ind. Pharm. 1995, 21, 2013. [14] K. Sreenivasan, J. Appl. Polym. Sci. 1997, 65, 1829. [15] F. Quaglia, G. Varricchio, A. Miro, M. I. L. Rotonda, D. Larobina, G. Mensitieri, J. Control. Release 2001, 71, 329. [16] J. Li, X. Ni, K. W. Leong, J. Biomed. Mater. Res. 2003, 65A, 196. [17] N. Pariot, F. Edwards-Levy, M.-C. Andry, M.-C. Levy, Intern. J. Pharm. 2002, 232, 175. [18] M. Oftsson, E. Brewster, J. Pharm. Sci. 1996, 85, 1017. [19] R. A. Ajewski, V. J. Stella, J. Pharm. Sci. 1996, 85, 1142. [20] T. Rie, K. Uekama, J. Pharm. Sci. 1997, 86, 147. [21] V. J. Stella, R. A. Rajewski, Pharm. Res. 1997, 14, 556. [22] Y. Y. Liu, X. D. Fan, Polymer 2002, 43, 4997. [23] Y. Y. Liu, X. D. Fan, J. Appl. Polym. Sci. 2003, 89, 361. [24] Y. Y. Liu, X. D. Fan, Q. Zhao, J. Macrom. Sci.- Pure Appl. Chem. 2003, A40, 1095. [25] Y. Y. Liu, X. D. Fan, L. Gao, Macromol. Biosci. 2003, 3, 715.
2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

736

Y.-Y. Liu, X.-D. Fan, H. Hu, Z.-H. Tang

[26] Y. Huang, X. D. Fan, Chinese J. Appl. Chem. 2003, 20, 38. [27] M. Dubois, K. A. Gilles, J. K. Hamilton, P. A. Pebers, F. Smith, Anal. Chem. 1956, 28, 350. [28] B. L. He, X. B. Zhao, Science in China (Ser. B) 1992, 12, 1240. [29] W. R. Owen, P. J. Stewart, J. Pharm. Sci. 1979, 68, 992. [30] N. Yuksel, T. Tincer, T. Baykara, Int. J. Pharm. 1996, 140, 145. [31] C. Dubernet, Thermochim. Acta 1995, 248, 259. [32] E. Dez-Pena, I. Quijada-Garrido, P. Frutos, J. M. BarralesRienda, Macromolecules 2002, 35, 2667.

[33] E. Kokufuta, B. Wang, R. Yoshida, A. R. Khokhlov, M. Hirata, Macromolecules 1998, 31, 6878. [34] G. H. Chen, A. S. Hoffman, Nature 1995, 373, 49. [35] W. F. Lee, C. H. Shieh, J. Appl. Polym. Sci. 1999, 73, 1955. [36] L. C. Dong, A. S. Hoffman, J. Control. Release 1991, 15, 141. [37] H. He, X. Cao, L. J. Lee, J. Control. Release 2004, 95, 391. [38] V. Pillay, R. Fassihi, J. Control. Release 1999, 95, 229. [39] Y. H. Lim, D. Kim, D. S. Lee, J. Appl. Polym. Sci. 1997, 64, 2647.

Macromol. Biosci. 2004, 4, 729736

www.mbs-journal.de

2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim