Biodegradable Thermoresponsive Hydrogels for Aqueous

May 21, 2005 - Therefore, abundant water could easily access and hydrolyze the ester bonds of oligolactate cross-links so that the hydrogels were brok...
5 downloads 11 Views 129KB Size
Download PDF
Biomacromolecules 2005, 6, 2131-2139

2131

Biodegradable Thermoresponsive Hydrogels for Aqueous Encapsulation and Controlled Release of Hydrophilic Model Drugs Xiao Huang† and Tao Lu Lowe*,†,‡,§ Departments of Surgery, Bioengineering, and Materials Science and Engineering, Pennsylvania State University, 500 University Drive, Hershey, Pennsylvania 17033 Received February 14, 2005; Revised Manuscript Received April 15, 2005

A series of hydrogels with both thermoresponsive and completely biodegradable properties was developed for aqueous encapsulation and controlled release of hydrophilic drugs in response to temperature change. The hydrogels were prepared in phosphate-buffered saline (pH 7.4) through free radical polymerization of N-isopropylacrylamide (NIPAAm) monomer and a dextran macromer containing multiple hydrolytically degradable oligolactate-2-hydroxyethyl methacrylate units (Dex-lactateHEMA). Swelling measurement results demonstrated that four gels with feeding weight ratios of NIPAAm:Dex-lactateHEMA ) 7:2, 6:3, 5:4, and 4:5 (w/w) were thermoresponsive by showing a lower critical solution temperature at approximately 32 °C. The swelling and degradation of the hydrogels strongly depended on temperature and hydrogel composition. An empirical mathematical model was established to describe the fast water absorption at the early stage and deswelling at the late stage of the hydrogels at 37 °C. Two hydrophilic model drugs, methylene blue and bovine serum albumin, were loaded into the hydrogels during the synthesis process. The molecular size of the drugs, the hydrophilicity and degradation of the hydrogels, and temperature played important roles in controlling the drug release. Introduction Hydrogels have attracted wide research interest as controlled release devices due to their tunable chemical and three-dimensional physical structure, high water content, good mechanical properties, and biocompatibility.1 Bioresponsive, “intelligent” or “smart” hydrogels can regulate drug release through responding to environmental stimuli by swelling and deswelling. Various bioresponsive hydrogels have been developed for drug delivery based on thermoresponsive polymer poly(N-isopropylacrylamide) (PNIPAAm) due to its unique volume phase transition at a lower critical solution temperature (LCST) in water around 32 °C, which is close to body temperature.2-13 They swell and collapse significantly in an aqueous environment at temperatures below and above the LCST, respectively. Several investigations, including our own, have shown that the phase behavior and mechanical properties of PNIPAAm hydrogels can be modified by the addition of more hydrophobic or hydrophilic monomers for desired drug delivery.3,6,7,9,13-19 PNIPAAmbased polymers may allow aqueous loading of protein drugs, protecting the drug from a hostile environment,20 and modulate drug release in response to temperature change.21 However, many current thermoresponsive PNIPAAm hydrogels have problems in nonbiodegradability and nonsustained drug release under physiological conditions.22,23 * To whom correspondence should be addressed. Phone: (717) 5318602. Fax: (717) 531-4464. E-mail: [email protected]. † Department of Surgery. ‡ Department of Bioengineering. § Department of Materials Science and Engineering.

Degradation of the hydrogel matrix not only circumvents removal of the empty device but also can be used to modulate the release of encapsulated drugs for a long period of time.24 Hydrogels composed of poly(lactic acid) (PLA)-based and dextran-based polymers have been extensively studied for sustained release of protein drugs in recent years, because PLA is hydrolytically degradable and hydrophobic,24-27 dextran, a natural polysaccharide, is enzymatically degradable and hydrophilic,24-29 and both polymers are biocompatible. When hydrogels consist of both PLA and dextran, they may control the sustained release of drugs by adjusting the ratios between PLA and dextran and through degradation by both hydrolysis and dextranase.24-27,30 However, many currently available biodegradable hydrogels have problems in the tradeoff of aqueous loading of protein drugs (to avoid using organic solvents for the loading, which may cause instability and denaturation of the protein drugs) and mechanical strength of the hydrogels,26,30 and lack of response to physiological changes as well. In view of these aspects, a promising strategy for designing novel hydrogel drug delivery systems is to combine the merits of both bioresponsive and biodegradable hydrogels. In the literature, PNIPAAm-based polymers have been chemically incorporated with enzymatically biodegradable poly(amino acid)12 or dextran.3,8,10,31 However, the enzymes needed to decompose these polymers are localized in limited biological systems, and hence, the application of such devices may sometimes be limited. In our previous work, we designed and synthesized a series of hydrogel systems composed of PNIPAAm, poly(L-lactic acid), and dextran in

10.1021/bm050116t CCC: $30.25 © 2005 American Chemical Society Published on Web 05/21/2005

2132

Biomacromolecules, Vol. 6, No. 4, 2005

dimethylformamide (DMF), and the hydrogels were thermoresponsive and partially hydrolytically degradable.32 To synthesize thermoresponsive and completely hydrolytically degradable hydrogels in aqueous solutions to incorporate a very high amount of protein drugs into the hydrogels, enhance the stability of the loaded drugs, and avoid future surgical removal of the hydrogels, in this study we have copolymerized NIPAAm monomer with a dextran macromer containing multiple hydrolytically degradable oligolactate2-hydroxyethyl methacrylate units (Dex-lactateHEMA) at different feeding weight ratios in phosphate-buffered saline (PBS) (pH 7.4) solutions. The dextran macromer was synthesized on the basis of the method developed by Hennink and co-workers.33,34 We have characterized the chemical structures of the synthesized hydrogels by the FTIR technique. We have also studied the following properties of the hydrogels: thermoresponsive and swelling properties at 25 and 37 °C, below and above the LCST, respectively, by weight measurements, and degradation properties at 37 °C by weight loss measurements and the FTIR technique. We have loaded two hydrophilic model drugs of very different molar masses, methylene blue (Mw ) 320 g‚mol-1) and bovine serum albumin (BSA; Mw ) 67000 g‚mol-1), into the hydrogels during the hydrogel synthesis process in PBS (pH 7.4) solutions. We have investigated the in vitro release kinetics of these two drugs at both 25 and 37 °C. Although not addressed in this paper, the enzymatic degradation of the designed hydrogels by dextranase could be expected to bring more versatility for the materials as drug delivery devices. This work provides insight for designing multifunctional biomaterials for organic-solvent-free encapsulation of hydrophilic drugs with high loading efficiency and sustained release of protein drugs at physiological temperature. Experimental Section Materials. Dextran (Mw ) 15000-20000 g‚mol-1) and methylene blue (Mw ) 320 g‚mol-1) were purchased from Polysciences, Inc., Warrington, PA, and Acros, Inc., Morris Plains, NJ, respectively. The following materials were obtained from Sigma-Aldrich, Inc., St. Louis, MO: NIPAAm, 2-hydroxethyl methacrylate (HEMA), 4-(N,N-diethylamino)pyridine (DMAP), N,N′-carbonyldiimidazole (CDI), L-lactide, stannous octoate (SnOct2), tetrahydrofuran (THF), dimethyl sulfoxide (DMSO), N,N,N′,N′-tetramethylethylenediamine (TEMED), potassium peroxydisulfate (KPS), BSA (Mw ) 67000 g‚mol-1), and Bradford reagent. All the chemicals were used as received. Deionized distilled water was used in all the experiments. Synthesis of Dex-lactateHEMA Macromers. DexlactateHEMA was synthesized according to the method developed by Hennink and co-workers33,34 with slight modification. Briefly, at first, L-lactide (4.32 g, 30 mmol) and HEMA (3.90 g, 30 mmol) were reacted in the presence of nitrogen and catalyst SnOct2 (121.5 mg, 1 mol % with respect to HEMA) at 110 °C for 1 h. HEMA-lactate product was collected by dissolving the cooled reaction mixture in THF, precipitating in ice-cold water, dissolving in ethyl acetate, drying over MgSO4, and concentrating under reduced

Huang and Lowe

pressure. The yield of the above reaction was 23%. Next, CDI (1.76 g, 10.8 mmol) was dissolved in THF (100 mL) in a nitrogen atmosphere and reacted with the above HEMA-lactate (3.74 g, 10.8 mmol) at room temperature for 16 h. The resulting HEMA-lactate-imidazolyl carbamate (HEMA-lactateCI) was obtained by solvent evaporation under reduced pressure, and the reaction yield was 100%. Finally, dextran (10.0 g, 61.8 mmol of glucopyranose residues) was dissolved in DMSO (90 mL) in a nitrogen atmosphere. After dissolution of DMAP (2.0 g), HEMAlactateCI (3.78 g, 8.60 mmol) was added, and the mixture was stirred at room temperature for 4 days. The DexlactateHEMA product was obtained by precipitating the reaction mixture in a large excess volume of cold dry 2-propanol, washed several times with 2-propanol, and dried in a vacuum for at least 24 h. The yield of the final reaction was 87%. Synthesis of Thermoresponsive and Biodegradable Hydrogels. NIPAAm and Dex-lactateHEMA (total amount 450 mg) at different weight ratios, NIPAAm:Dex-lactateHEMA ) 7:2, 6:3, 5:4, and 4:5 (w/w), were dissolved in 2.28 mL of PBS (pH 7.4) solvent under nitrogen. After dissolution, KPS in PBS (50 mg/mL, 270 µL) and TEMED in PBS (20% (v/v), 150 µL) were added as initiator and accelerator, respectively. The mixture was mixed thoroughly and quickly injected into the space between two glass plates wrapped by Teflon films (thickness 1.6 mm), and the gelation was allowed to proceed at room temperature for 1 h. Diskshaped gel samples (∼8 mm in diameter) were then cut off, washed in a large volume of a 50:50 (v/v) ethanol/water mixture for 30 min, and dried in the air till no further weight loss occurred. Hydrogels were denoted as gels 7/2, 6/3, 5/4, and 4/5 corresponding to their initial NIPAAm:DexlactateHEMA feeding ratios of 7:2, 6:3, 5:4, and 4:5 (w/w), respectively. The dried 7/2 gel appeared to be transparent, while the other dried gels were semitransparent, possibly because the hydrogels increased in heterogeneity with increasing Dex-lactateHEMA component (decreasing the NIPAAM component).19,35 The synthesis yields of the gels 7/2, 6/3, 5/4, and 4/5 were 95%, 99%, 97%, and 92%, respectively. Chemical Structure Characterization. To confirm the success of the syntheses of the Dex-lactateHEMA macromer, the chemical shifts of the dextran, lactateHEMA, and Dex-lactateHEMA which matched their nuclear spins in a magnetic field were measured using a proton nuclear magnetic resonance (1H NMR, Bruker DPX-300, Ettlingen, Germany) spectrophotometer. For lactateHEMA, deuterated chloroform (CDCl3, 99.9%, Aldrich) was used as solvent, and the chloroform-d1 resonance was set at 7.27 ppm. For dextran and Dex-lactateHEMA macromer, DMSO-d6 (99.9%, Aldrich) was used as solvent, and the central DMSO line was set at 2.50 ppm. Samples for the 1H NMR measurements were prepared by dissolving approximately 10 mg of each material in 1 mL of corresponding solvent. To confirm the successful syntheses of the hydrogels, we characterized the chemical structures of the synthesized hydrogels through studying their infrared absorption bands which match their natural vibrational modes using an

Biomacromolecules, Vol. 6, No. 4, 2005 2133

Hydrogels for Encapsulation and Controlled Release

Figure 1. Schematic chemical structure of Dex-lactateHEMA (a) and NIPAAm-co-Dex-lactateHEMA (b) hydrogels.

attenuated total reflection (Pike Technologies, Madison, WI) Fourier transform infrared (ATR-FTIR) spectroscope (Thermo Nicolet Avetar 370, Madison, WI). The dry gels 7/2, 6/3, 5/4, and 4/5 grounded into pieces were compressed onto the ZnSe crystal, and FTIR spectra were recorded in the wavenumber range of 4000-650 cm-1. Physical Properties of Hydrogels. 1. Thermoresponsive Properties. Weighed dry hydrogel samples were immersed in PBS (pH 7.4) at 10 °C. The temperature was raised every 1 h, ranging from 10 to 60 °C. Before each temperature adjustment, the swollen gel samples were removed from the solvent, the surface water was carefully absorbed by filter paper, and the gel samples were weighed. The swelling ratio q was defined as q ) (Wt - W0)/W0

(1)

in which Wt and W0 are the weights of the swollen and dry gels, respectively. 2. Swelling Properties. Dry hydrogel samples were immersed in PBS (pH 7.4) at 37 °C. Weights of the wet hydrogel samples were measured at selected time points. Equation 1 was used to calculate the swelling ratio q. PBS (pH 7.4) solvent was replaced every day. 3. Degradation Properties. The degradation of the synthesized hydrogels should be dependent on both environmental pH and the presence of enzyme dextranase due to the existence of poly(L-lactide) and dextran segments, respectively. In this paper we only focused on the degradation study under normal physiological conditions by using PBS (pH 7.4) as the degradation medium. Dry hydrogel samples were immersed in PBS (pH 7.4) at 25 and 37 °C, respectively. At predetermined time points, samples were removed from the solvents, dried in the air overnight, and weighed to determine weight loss. After weighing, the samples were placed back in PBS (pH 7.4) for continuous degradation. PBS solvent was replaced every day. The FTIR spectra of the degrading hydrogels were also recorded by ATR-FTIR spectroscopy at selected time points. The wet hydrogels were compressed onto the ZnSe crystal, and three positions of each hydrogel sample were chosen randomly for the FTIR measurements. The FTIR spectra were processed with a background subtraction of the PBS (pH 7.4) solvent.

Drug Loading and in Vitro Release. Hydrophilic model drugs methylene blue (Mw ) 320 g‚mol-1) and protein BSA (Mw ) 67000 g‚mol-1) were loaded into the hydrogels during the hydrogel synthesis. In detail, methylene blue or BSA (22.5 mg, 5 wt % with respect to the polymerization precursors) was dissolved in NIPAAm/Dex-lactateHEMA in PBS (pH 7.4) solutions before the addition of initiator KPS and accelerator TEMED (see the above text for the detailed hydrogel synthesis). After synthesis, the gels were dried directly in the air without washing. In vitro release studies were conducted by immersing dried methylene blueor BSA-loaded gels in 3 mL of PBS (pH 7.4) solvent at 25 and 37 °C, respectively. At selected time intervals during one month, 1 mL release solutions were collected and replaced with fresh PBS (pH 7.4) solvent. Methylene blue and BSA amounts were quantified by a UV/vis spectrophotometer (PerkinElmer Lamda 25, Shelton, CT) at a wavelength of 668 nm and Bradford protein assay, respectively. Results and Discussion Synthesis of Dex-lactateHEMA Macromers. The chemical structures of Dex-lactateHEMA macromer and PNIPAAm-co-Dex-lactateHEMA hydrogels are sketched in Figure 1.24 1H NMR spectra presented in Figure 2 confirmed the successful synthesis of lactateHEMA and Dex-lactateHEMA macromer. As shown in Figure 2B,C, the characteristic chemical shift peaks of HEMA residues at δ ≈ 6.0 (a, CH2dC gel 4/5. The reason was that both PNIPAAm and dextran

Huang and Lowe

Figure 8. Swelling ratios of gels 7/2 (b), 6/3 (2), 5/4 (9), and 4/5 ([) in PBS (pH 7.4) solvent at 37 °C as a function of time: symbols, experimental data; dashed lines, fitting curves based on eq 5.

moieties in the hydrogels were very hydrophilic at 25 °C and the cross-linking density (hydrophobic oligolactate) of the hydrogels increased in the order gel 7/2 < gel 6/3 < gel 5/4 < gel 4/5. From the time beyond 2 h to the time when the hydrogels started disintegration, all four hydrogels continuously absorbed water and swelled to higher and higher levels. The mechanical strength of the hydrated samples became weak, and thus, the handling of the samples became difficult after 2 h. Therefore, the swelling ratio measurements were stopped at 2 h. The swelling ratios of the four gels 7/2, 6/3, 5/4, and 4/5 in PBS (pH 7.4) solutions against time at 37 °C are depicted in Figure 8, with the time axis on the logarithmic scale. All four hydrogels had much lower swelling ratios at 37 °C than at 25 °C on the same time scale because the PNIPAAm moiety was hydrophobic at 37 °C. The swelling ratios of all four hydrogels at 37 °C increased quickly, reached maximum values within the first 4 h, and increased in the order gel 7/2 < gel 6/3 < gel 5/4 < gel 4/5 with increasing DexlactateHEMA component due to the hydrophilicity of dextran chains. From 4 h to 1 day, the swelling ratios of all four gels 7/2, 6/3, 5/4, and 4/5 started to decrease dramatically and reached relatively stable levels at around 3, 7, 12, and >20 days, respectively. The swelling ratios decreased in the order gel 7/2 ≈ gel 6/3 , gel 5/4 < gel 4/5 between 1 and 4 days, and oppositely started to increase in the order gel 7/2 > gel 6/3 g gel 5/4 g gel 4/5 between 6 and 18 days. In the previous section, we have discussed that, as the degradation of the hydrogels proceeded, the freed hydrophilic dextran segments quickly entered into the surrounding buffer solutions, not only causing a decrease in hydrogel masses, but also leaving behind hydrogel networks that became more and more hydrophobic. As a result, water was expelled from inside the gels, leading to gel deswelling. The reduction of both water and hydrogel masses may contribute to the decrease of swelling ratios after the maximum values are reached. Since the decreases of polymer masses due to degradation within 8 days were only ∼15-50% of the original dry hydrogel weights (Figure 5), hydrogel deswelling was the dominant factor that caused swelling ratios to decrease at the later stage. It is easy to understand that hydrogels with higher amounts of the Dex-lactateHEMA

Biomacromolecules, Vol. 6, No. 4, 2005 2137

Hydrogels for Encapsulation and Controlled Release Table 1. Fitting Parameters in Equation 5 for the Swelling Ratio Kinetics of Hydrogels at 37 °C hydrogel sample

n

m

K1

K2

R2

gel 4/5 gel 5/4 gel 6/3 gel 7/2

1.236 0.9686 0.6837 1.742

0.4326 0.5082 0.7576 0.04721

0.001114 0.00549 0.03219 0.0007231

0.1784 0.1778 0.2642 0.7019

0.994 0.988 0.911 0.640

moiety had faster initial swelling and reached higher maximum swelling ratios due to their higher hydrophilicity. Previously, we also discussed that the mass loss rates of the hydrogels decreased with time. When the mass losses entered into slower stages, changes in hydrogel hydrophilicity also became less dramatic, resulting in lower deswelling rates and eventually minimal changes of swelling ratios with time. On the basis of the above discussion, to further understand and interpret the mechanisms behind the biphase swelling ratio data depicted in Figure 8, we developed an empirical mathematical model. The increasing phase of the swelling ratio curves at the early stage before 4 h in Figure 8 might be fit by the power law, which has been suggested as a simple model to describe solvent uptake and swelling of nonbiodegradable hydrogels:38,39 q1 ) k1tn

(2)

where q1 is the swelling ratio, t is time, and k1 and n are constants. When the exponential n ) 0.5, solvent diffusion follows Fickian diffusion, when n ) 1.0, the diffusion is case II diffusion, and when 0.5 < n < 1.0, the diffusion is a combination of Fickian and case II diffusion and is usually called anomalous diffusion.38 On the other hand, the decreasing phase of the swelling ratio kinetics might be fit by an exponential decay function:40 q2 ) k2 exp(-mt)

(3)

where q2 is the swelling ratio and k2 and m are constants. Here we combined eqs 2 and 3 to establish a new equation for the description of the whole range of swelling ratio change of the degradable hydrogel systems designed in this study: 1 1 1 ) + q q1 q2

(4)

or q)

K1t

-n

1 + K2 exp(mt)

(5)

where q is the swelling ratio and K1, K2, n, and m are all constants. When the value of t is small, K1t-n is the dominant term in the denominator, so that eq 5 is similar to eq 2; on the other hand, when t takes a large value, K2 exp(mt) makes a major contribution to the denominator, and eq 5 may be restored to the form of eq 3. Fitting curves are plotted together with the experimental data in Figure 8, and the values of the parameters are tabulated in Table 1. The results suggested that the developed

Figure 9. Fractional release of methylene blue from gel 5/4 in PBS (pH 7.4) solvent at 25 (O) and 37 (b) °C.

equation fit the swelling ratio data of gels 4/5, 5/4, and 6/3 well. With a decrease of the hydrophilicity of the hydrogels, the exponential index n decreased from 1.236 for gel 4/5 to 0.9686 for gel 5/4 and to 0.6837 for gel 6/3, corresponding to so-called super case II, perfect case II, and anomalous transport, respectively.41 Oppositely, the parameter m increased in the order gel 6/3 > gel 5/4 > gel 4/5, which might indicate that water inside the hydrogel networks was expelled at a higher rate with increasing hydrogel hydrophobicity. The fitting of the swelling data of gel 7/2 (Figure 6) was unsatisfactory by using the developed model, which might be owed to the relatively too high hydrophobicity and rigidity of gel 7/2 at 37 °C. Equation 5 was established based on eq 2 for swellable hydrogels and eq 3 for shrinking hydrogels, so that when hydrogels were hydrophobic and slightly swellable, more or less like homo-PNIPAAm gels, such as gel 7/2, eq 5 was not applicable to model their swelling behavior anymore. In Vitro Release of Model Drugs from 5/4 Gels. To develop the designed hydrogels as controlled drug release devices, two hydrophilic model drugs, methylene blue and BSA, were loaded into the hydrogels during the synthesis process in PBS (pH 7.4) solutions. Methylene blue was chosen because its molar mass is small (Mw ) 320 g‚mol-1) and its concentration can be easily detected by UV/vis spectroscopy without interference from degraded polymer materials. BSA was employed because it is widely used as a model protein drug, its molar mass (Mw ) 67000 g‚mol-1) is much higher than that of methylene blue, and its concentration can be easily quantified by Bradford protein assay. In vitro release profiles of methylene blue in PBS (pH 7.4) solvent were studied at temperatures below (25 °C) and above (37 °C) the LCST. Figure 9 demonstrates the fractional release Mt/M∞ of methylene blue against time, where Mt and M∞ are the cumulative amount of drug released at time t and equilibrium, respectively. At 25 °C, data were collected before the hydrogels started disintegrating. The fractional release of methylene blue showed more than 80% initial burst release within the first 1 h at 37 °C, and reached equilibrium after 1 and 4 days at 37 and 25 °C, respectively. Surprisingly, the release rate was higher at 37 °C than at 25 °C, suggesting that the release was not dominantly controlled by the swelling

2138

Biomacromolecules, Vol. 6, No. 4, 2005

Huang and Lowe

from gel 5/4 at 25 °C. The release rate of BSA from gel 5/4 was higher at 25 °C than at 37 °C, which was opposite the temperature-dependent trend of the release rate of methylene blue. This might be because the diffusion of a high molar mass molecule such as BSA (Mw ) 67000 g‚mol-1) could be hindered and thus controlled by the degradation of gel 5/4 at 37 °C when gel 5/4 is relatively hydrophobic and dense at 37 °C. Conclusions

Figure 10. Fractional release of BSA from gel 5/4 in PBS (pH 7.4) solvent at 25 (O) and 37 (b) °C. The dashed line is the fitting curve generated from eq 6.

and degradation of the hydrogel. The possible reasons might be as follows: (1) Methylene blue is a small hydrophilic molecule with molar mass 320 g‚mol-1, so that its diffusion into the release medium was hardly hindered by the hydrogel network. (2) The hydrogel network was more hydrophobic at 37 °C than at 25 °C, which would be favorable for the small hydrophilic molecules to enter into the aqueous release medium. (3) The hydrogel size was bigger at 25 °C than at 37 °C, so that the concentration gradients of methylene blue inside the hydrogels were lower at 25 °C than at 37 °C.42,43 It is worth mentioning that 90% of loaded methylene blue was released from the hydrogels at equilibrium time at both 25 and 37 °C. Figure 10 demonstrates the fractional release Mt/M∞ profiles of model protein BSA. At a temperature below the LCST, 25 °C, the BSA release was completed within 4 days before the hydrogels started to disintegrate. At a temperature above the LCST, 37 °C, the release curve was featured by a moderate initial burst followed by a sustained release with diminishing rate for up to 15-20 days. The BSA release reached a plateau with around 40% of the initially loaded BSA retained inside gel 5/4 after 15 days according to the accumulative release data (data not shown). The BSA fractional release at 37 °C for up to 15 days could be fit by the well-known power equation, in the same format as eq 2:39 Mt ) ktn M∞

(6)

The above equation has been used to analyze drug release from both swellable and nonswellable polymeric delivery systems.43 Fickian diffusion (n ) 0.5) and case II transport (n ) 1) are often obtained when drugs are released from diffusion-controlled and swelling-controlled systems, respectively. A system controlled by both diffusion and swelling usually generates 0.5 < n < 1. In our system, after data fitting, we obtained k ) 0.531, n ) 0.2595, and R2 ) 0.991. The reason that n was smaller than 0.5 might be due to the increasing hydrophobicity of the hydrogel networks with time. Due to the fast swelling and degradation of gel 5/4 at 25 °C, eq 6 could not be used to fit the release data of BSA

A series of novel thermoresponsive and biodegradable hydrogels were designed and prepared in aqueous PBS (pH 7.4) solvent by copolymerizing NIPAAm and Dex-lactateHEMA, a dextran methacrylate macromer containing a hydrolyzable oligolactate spacer, at different feeding weight ratios: NIPAAm:Dex-lactateHEMA ) 7:2, 6:3, 5:4, and 4:5 (w/w). All four synthesized hydrogels were temperature sensitive by showing an LCST at approximately 32 °C. Their swelling ratios increased and oppositely decreased at temperatures below and above the LCST, respectively, with decreasing Dex-lactateHEMA moiety (or increasing PNIPAAm moiety). The four hydrogels swelled less and degraded slower in PBS (pH 7.4) solutions at 37 °C (above the LCST) than at 25 °C (below the LCST), and the network hydrophobicity increased while the hydrogels were degraded at 37 °C. The dynamic weight loss amounts and rates of the hydrogels at 37 °C increased with increasing Dex-lactateHEMA moiety (or decreasing PNIPAAm moiety). The swelling ratios of the hydrogels against time at 37 °C showed biphase curves, and were fit by an empirical mathematical equation (5) combining power eq 2 and exponential eq 3 when the hydrogels were not highly hydrophobic such as gels 6/3, 5/4, and 4/5. Equation 2 was based on the main contribution of the solvent diffusion and hydrogel swelling at the early increasing swelling stage, and exponential eq 3 was based on the main contribution of the increased hydrogel hydrophobicity with time at the late decreasing swelling stage. Two hydrophilic model drugs, low molar mass methylene blue (Mw ) 320 g‚mol-1) and high molar mass BSA (Mw ) 67000 g‚mol-1), were loaded into gel 5/4 during the synthesis processes in aqueous PBS (pH 7.4) solutions. The release of methylene blue from gel 5/4 was slower at 25 °C (below the LCST) than at 37 °C (above the LCST), while the release of BSA at these two temperatures showed the opposite trend. These results suggest that the drug release kinetics strongly depend on environmental temperature, the swelling and degradation of the hydrogel, and the interactions of the loaded drugs with the hydrogel networks. BSA release data at 37 °C were fit by a traditional power equation (6), generating a diffusion exponent n < 0.5, which might be due to the increasing hydrogel hydrophobicity with time. Currently, we are carrying out further aqueous synthesis of the thermoresponsive and biodegradable hydrogels by changing the length of the oligolactate spaces, the number of dextran’s -OH pending groups substituted by crosslinkable groups, and the feeding weight ratios between NIPAAm and dextran macromer for controlled release of a

Hydrogels for Encapsulation and Controlled Release

variety of drugs in response to temperature changes. We are also investigating the biosafety of the developed hydrogels and the stability of loaded protein drugs. Acknowledgment. We are grateful to Dr. Shengsheng Liu for his generous help with the NMR analysis and the Whitaker Foundation and Penn State Surgery Feasibility Grant for financial support. References and Notes (1) Langer, R.; Peppas, N. A. AIChE J. 2003, 49, 2990-3006. (2) Zhang, J.; Peppas, N. A. J. Biomater. Sci., Polym. Ed. 2002, 13, 511-525. (3) Kurisawa, M.; Matsuo, Y.; Yui, N. Macromol. Chem. Phys. 1998, 199, 705-709. (4) Lowe, T. L.; Virtanen, J.; Tenhu, H. Polymer 1999, 40, 2595-2603. (5) Lowe, T. L.; Tenhu, H.; Tylli, H. J. Appl. Polym. Sci. 1999, 73, 1031-1039. (6) Katime, I.; Valderruten, N.; Quintana, J. R. Polym. Int. 2001, 50, 869-874. (7) Guilherme, M. R.; Silva, R.; Girotto, E. M.; Rubira, A. F.; Muniz, E. C. Polymer 2003, 44, 4213-4219. (8) Zhang, X. Z.; Wu, D. Q.; Sun, G. M.; Chu, C. C. Macromol. Biosci. 2003, 3, 87-91. (9) Kaneko, Y.; Yoshida, R.; Sakai, K.; Sakurai, Y.; Okano, T. J. Membr. Sci. 1995, 101, 13-22. (10) Kumashiro, Y.; Huh, K. M.; Ooya, T.; Yui, N. Biomacromolecules 2001, 2, 874-879. (11) Zhang, J. T.; Cheng, S. X.; Huang, S. W.; Zhuo, R. X. Macromol. Rapid Commun. 2003, 24, 447-451. (12) Yoshida, T.; Aoyagi, T.; Kokufuta, E.; Okano, T. J. Polym. Sci., Part A: Polym. Chem. 2003, 41, 779-787. (13) Brazel, C. S.; Peppas, N. A. Macromolecules 1995, 28, 8016-8020. (14) Stile, R. A.; Burghardt, W. R.; Healy, K. E. Macromolecules 1999, 32, 7370-7379. (15) Shibayama, M.; Mizutani, S.; Nomura, S. Macromolecules 1996, 29, 2019-2024. (16) Vernon, B.; Kim, S. W.; Bae, Y. H. J. Biomed. Mater. Res. 2000, 51, 69-79. (17) Vernon, B.; Gutowska, A.; Kim, S. W.; Bae, Y. H. Macromol. Symp. 1996, 109, 155-167. (18) Lowe, T. L.; Virtanen, J.; Tenhu, H. Langmuir 1999, 15, 42594265. (19) Lowe, T. L.; Benhaddou, M.; Tenhu, H. Macromol. Chem. Phys. 1999, 200, 51-57.

Biomacromolecules, Vol. 6, No. 4, 2005 2139 (20) Ramkissoon-Ganorkar, C.; Liu, F.; Baudys, M.; Kim, S. W. J Controlled Release 1999, 59, 287-298. (21) Zhang, J. T.; Huang, S. W.; Cheng, S. X.; Zhuo, R. X. J Polym Sci., Part A: Polym. Chem. 2004, 42, 1249-1254. (22) Qiu, Y.; Park, K. AdV Drug DeliVery ReV. 2001, 53, 321-339. (23) Bromberg, L. E.; Ron, E. S. AdV. Drug DeliVery ReV. 1998, 31, 197221. (24) vanDijkWolthuis, W. N. E.; Hoogeboom, J. A. M.; vanSteenbergen, M. J.; Tsang, S. K. Y.; Hennink, W. E. Macromolecules 1997, 30, 4639-4645. (25) Zhang, Y. L.; Chu, C. C. J. Biomed. Mater. Res. 2001, 54, 1-11. (26) Zhang, Y. L.; Chu, C. C. J. Biomater. Appl. 2002, 16, 305-325. (27) Zhang, Y. L.; Chu, C. C. J Biomed. Mater. Res. 2002, 59, 318-328. (28) Hennink, W. E.; Franssen, O.; vanDijkWolthuis, W. N. E.; Talsma, H. J. Controlled Release 1997, 48, 107-114. (29) Hennink, W. E.; Talsma, H.; Borchert, J. C. H.; DeSmedt, S. C.; Demeester, J. J. Controlled Release 1996, 39, 47-55. (30) Bos, G. W.; Jacobs, J. J. L.; Koten, J. W.; Van Tomme, S.; Veldhuis, T.; van Nostrum, C. F.; Den Otter, W.; Hennink, W. E. Eur. J. Pharm. Sci. 2004, 21, 561-567. (31) Kurisawa, M.; Yui, N. Macromol. Chem. Phys. 1998, 199, 26132618. (32) Huang, X.; Nayak, B. R.; Lowe, T. L. J. Polym. Sci., Part A: Polym. Chem. 2004, 42, 5054-5066. (33) Vandijkwolthuis, W. N. E.; Franssen, O.; Talsma, H.; Vansteenbergen, M. J.; Vandenbosch, J. J. K.; Hennink, W. E. Macromolecules 1995, 28, 6317-6322. (34) vanDijkWolthuis, W. N. E.; Tsang, S. K. Y.; KettenesvandenBosch, J. J.; Hennink, W. E. Polymer 1997, 38, 6235-6242. (35) Gehrke, S. H. AdV. Polym. Sci. 1993, 110, 81-144. (36) Copinet, A.; Bertrand, C.; Govindin, S.; Coma, V.; Couturier, Y. Chemosphere 2004, 55, 763-773. (37) Ho, K. L. G.; Pometto, A. L.; Hinz, P. N. J. EnViron. Polym. Degrad. 1999, 7, 83-92. (38) Baker, R. Controlled release of biologically actiVe agents; John Wiley & Sons: New York, 1987. (39) Franson, N. M.; Peppas, N. A. J. Appl. Polym. Sci. 1983, 28, 12991310. (40) Lynch, I.; Gorelov, A.; Dawson, K. A. Phys. Chem. Chem. Phys. 1999, 1, 2103-2108. (41) Peppas, N. A.; Brannonpeppas, L. J. Food Eng. 1994, 22, 189-210. (42) Cadee, J. A.; de Groot, C. J.; Jiskoot, W.; den Otter, W.; Hennink, W. E. J. Controlled Release 2002, 78, 1-13. (43) Colombo, P.; Bettini, R.; Catellani, P. L.; Santi, P.; Peppas, N. A. Eur. J. Pharm. Sci. 1999, 9, 33-40.

BM050116T