Synthesis and Characterization of Injectable Poly (N

Biomacromolecules 2003, 4, 1214-1223

1214

Synthesis and Characterization of Injectable Poly(N-isopropylacrylamide-co-acrylic acid) Hydrogels with Proteolytically Degradable Cross-Links Soyeon Kim and Kevin E. Healy* Departments of Bioengineering and Materials Science and Engineering, University of California at Berkeley, Berkeley, California 94720 Received February 11, 2003; Revised Manuscript Received June 5, 2003

Hydrogels composed of N-isopropylacrylamide (NIPAAm) and acrylic acid (AAc) were prepared by redox polymerization with peptide cross-linkers to create an artificial extracellular matrix (ECM) amenable for testing hypotheses regarding cell proliferation and migration in three dimensions. Peptide degradable crosslinkers were synthesized by the acrylation of the amine groups of glutamine and lysine residues within peptide sequences potentially cleavable by matrix metalloproteinases synthesized by mammalian cells (e.g., osteoblasts). With the peptide cross-linker, loosely cross-linked poly(N-isopropylacrylamide-co-acrylic acid) [P(NIPAAm-co-AAc)] hydrogels were prepared, and their phase transition behavior, lower critical solution temperature (LCST), water content, and enzymatic degradation properties were investigated. The peptidecross-linked P(NIPAAm-co-AAc) hydrogels were pliable and fluidlike at room temperature and could be injected through a small-diameter aperture. The LCST of peptide-cross-linked hydrogel was influenced by the monomer ratio of NIPAAm/AAc but not by cross-linking density within the polymer network. A peptidecross-linked hydrogel with a 97/3 molar ratio of NIPAAm/AAc exhibited a LCST of ∼34.5 °C. Swelling was influenced by NIPAAm/AAc monomer ratio, cross-linking density, and swelling media; however, all hydrogels maintained more than 90% water even at 37 °C. In enzymatic degradation studies, breakdown of the peptide-cross-linked P(NIPAAm-co-AAc) hydrogels was dependent on both the concentration of collagenase and the cross-linking density. These results suggest that peptide-cross-linked P(NIPAAm-coAAc) hydrogels can be tailored to create environmentally-responsive artificial extracellular matrixes that are degraded by proteases. Introduction Three-dimensional (3D) polymer scaffolds that mimic native extracellular matrixes (ECMs) have been developed to control tissue structure and regulate the function of cells for tissue engineering applications.1-14 Biodegradable extracellular matrixes provide the opportunity of creating completely natural new tissues because they can be remodeled by the regenerating tissue and do not leave behind a residual scaffold that impedes functional adaptation of the new tissue.11,15-20 Therefore, a wide variety of biodegradable polymer scaffolds composed of synthetic or naturally derived materials have been developed.1,3-5 Although naturally derived materials, such as fibrin, alginate, and collagen, may closely simulate the native cellular milieu, they can carry the risk of disease transmission and require difficult purification procedures. Furthermore, the most widely used synthetic aliphatic polyesters in tissue engineering, such as polyglycolic acid (PGA), polylactic acid (PLA), and their copolymers (PLGA), degrade by random hydrolysis and do not offer * To whom correspondence should be addressed. Mailing address: Kevin E. Healy, Ph.D., University of California at Berkeley, Departments of Bioengineering and, Materials Science and Engineering, 370 HMMB #1760, Berkeley, CA 94720-1760. Phone: (510) 643-3559. Facsimile: (510) 6435792. E-mail: [email protected].

temporal specificity in degradation during cellular remodeling and tissue regeneration.3-5,17 During fetal development, inflammation, arthritis, cancer, wound healing, and tissue regeneration, the proteolytic remodeling process of the ECM is mediated by a number of enzymes secreted or locally activated by migrating cells.11,17-19 Thus, greater control over material degradation, cell ingrowth, and tissue regeneration can be achieved with bioadaptable matrixes that are designed to respond to the presence of the cells and molecules that they synthesize. To recapitulate the mechanism of ECM degradation during wound healing, West and Hubbell19 reported the first work using resorbable scaffolds with protease-susceptible crosslinks. They developed a new class of telechelic biodegradable block copolymers that when synthesized into a photocrosslinked hydrogel were specifically degraded by either plasmin or crude collagenase. Their work established the feasibility of protease degradation of oligopeptide-crosslinked hydrogels. Subsequent studies have demonstrated that proteolytic degradation of artificial ECMs, spatially and temporally in concert with cellular outgrowth, can duplicate some of the fundamental aspects of wound healing and that these artificial ECMs may promote more natural tissue regeneration than their hydrolytically degradable counterparts.11,20

10.1021/bm0340467 CCC: $25.00 © 2003 American Chemical Society Published on Web 08/08/2003

Injectable P(NIPAAm-co-AAc) Hydrogels

Matrix metalloproteinases (MMPs) have been implicated as important proteases in the remodeling of the ECM, and their substrates are prime candidates for use as cross-linkers in proteolytically degradable artificial ECMs. MMPs are a structurally and functionally related family of zinc-dependent endopeptidases that cleave either one or several ECM proteins.21-28 Specifically, we are interested in peptide crosslinkers that mimic the domain on type II collagen that is cleaved by matrix metalloproteinases-13 (MMP-13, collagenase-3). The choice of peptides that model type II collagen is based on the concept that during endochondral bone formation a cartilage matrix template rich in type II collagen is remodeled and replaced by bone and that osteoblasts synthesize MMP-13, a type II collagen specific MMP.25 The substrate specificity of MMP-13 has been demonstrated by its ability to degrade type II collagen six times more effectively than either collagen type I or collagen type III and its nearly 50-fold stronger collagenolytic activity than either MMP-1 or MMP-8.26,27 In addition, the expression of MMP-13 appears to be limited to hypertrophic chondrocytes, periosteal cells, and osteoblasts during human fetal bone development.24 However, MMP-13 has also been detected in human breast carcinoma tissue28 and in osteoarthritic cartilage and chondrocytes.26 The limited expression of MMP-13 and its specificity for type II collagen make oligopeptides based on the cleavage site of MMP-13 ideal candidates for peptide cross-linkers in hydrogels for bone tissue engineering. Recently, tissue engineering approaches using injectable, in situ gel-forming polymeric hydrogel systems have been reported.29-40 From a clinical perspective, injectable scaffolds are extremely desirable because they permit minimally invasive delivery of the construct and thereby reduce surgical risks. The ideal injectable polymer scaffold for tissue engineering would satisfy the following criteria: it should be pliable enough at room temperature to allow for the incorporation of cells and macromolecules, be amenable to minimally invasive implantation procedures, demonstrate in situ stabilization upon placement into the body, improving the mechanical integrity of the matrix to better support tissue or organ formation, be minimally toxic, and be capable of interacting with the host tissue and biological environment on a molecular level.29,31 Injectable scaffolds have been designed using materials such as alginate,32,33 pluronics [i.e., block copolymer of poly(ethylene oxide) and poly(propylene oxide)],34 hyaluronic acid,35 poly(ethylene oxide),36 chitosan,29 and poly(N-isopropylacrylamide-co-acrylic acid) [P(NIPAAm-co-AAc)].37-40 One strategy for developing an injectable polymer scaffold is to employ the principles of lower critical phase separation. It is well-known that poly(N-isopropylacrylamide) (PNIPAAm) in aqueous solution has a lower critical solution temperature (LCST) at around 32 °C.41,42 PNIPAAm and cross-linked hydrogels of PNIPAAm form hydrogen bonds with aqueous solutions. Below the LCST, PNIPAAm is soluble in water and hydrogels swell. However, above the LCST, the entropic contribution to the free energy dominates the enthalpic contribution from hydrogen bonds, causing the polymer to precipitate and the hydrogel to expel water and shrink.41,42

Biomacromolecules, Vol. 4, No. 5, 2003 1215

These types of phase transitions can be exploited to develop materials that are injectable liquids at room temperature, and form viscoelastic solids at physiological temperature. In previous studies, we have investigated injectable thermoresponsive hydrogels composed of N-isopropylacrylamide (NIPAAm) and acrylic acid (AAc) [P(NIPAAm-coAAc)] for applications in tissue engineering.37-40 We developed these hydrogels to address the lack of engineering design rules used in the fabrication of artificial ECMs. To address this problem, we embarked on a long-term project to create artificial ECMs that are environmentally responsive and independently tunable with respect to mechanical properties, biological ligands, tissue adhesion, and protease degradation. We have demonstrated that the physical properties of P(NIPAAm-co-AAc) hydrogels such as complex shear modulus (G*) can be adjusted using semiinterpenetrating polymer networks (semi-IPNs) consisting of P(NIPAAmco-AAc) hydrogels and linear P(AAc) chains and that peptide-modified P(NIPAAm-co-AAc) hydrogels supported cell spreading, proliferation, and phenotypic expression in vitro.37-40 Therefore, the objective of this study was to develop injectable P(NIPAAm-co-AAc) hydrogels that degrade preferentially in the presence of cells secreting specific proteases, for example, MMP-13 in the case of osteoblasts. To synthesize the proteolytically degradable peptide crosslinker, we used the peptide sequence containing the MMP13-labile sequence -PQGLA-, Gln-Pro-Gln-Gly-Leu-AlaLys-NH2. We synthesized the peptide cross-linker with reactive acrylate groups and prepared protease-degradable P(NIPAAm-co-AAc) hydrogel networks with the peptide cross-linker via radical addition polymerization. To evaluate the feasibility as injectable thermoresponsive artificial ECMs, the LCST and water content of the peptide-cross-linked P(NIPAAm-co-AAc) hydrogels were characterized. Finally, their enzymatic degradation behavior in the presence of protease as a function of time was also investigated. Experimental Section Materials. N-Isopropylacrylamide (NIPAAm), acrylic acid (AAc), N,N′-methylenebis(acrylamide) (BIS; Chemzymes ultrapure grade), and N,N,N′,N′-tetramethylethylenediamine (TEMED; Chemzymes ultrapure grade) were purchased from Polysciences, Inc. (Warrington, PA). Ammonium peroxydisulfate (APS; electrophoresis grade) and 2,4,6-trinitrobenzene sulfonic acid (TNBS, 5% (w/v) aqueous solution) were obtained from Fisher Scientific, Fisher Chemicals (Fair Lawn, NJ), and Sigma (St. Louis, MO), respectively. Acryloyl chloride, triethylamine, dimethylacetamide (DMAc), and deuterium oxide (D2O) were obtained from Aldrich (Milwaukee, WI). The peptide sequence containing the MMP-13-labile sequence, Gln-Pro-Gln-Gly-Leu-Ala-LysNH2 (-QPQGLAK-), was synthesized by American Peptide Company (Sunnyvale, CA). Dulbecco’s phosphate-buffered saline (PBS, without calcium chloride, without magnesium chloride, pH ) 7.2 ( 0.1), Dulbecco’s Modified Eagle Medium (DMEM, high glucose, with L-glutamine, with pyridoxine hydrochloride, without sodium pyruvate), and heat-inactivated fetal bovine serum (FBS) were purchased

1216

Biomacromolecules, Vol. 4, No. 5, 2003

Scheme 1. Synthesis of a peptide crosslinker containing a matrix metalloproteinase-13 (MMP-13, collagenase-3) degradable sequence

from GIBCO BRL (Grand Island, NY). Dialyzers (SpectraPor, molecular weight cut off ) 500 Da) were purchased from Spectrum Laboratories, Inc. (Rancho Dominguez, CA). Crude collagenase (from Clostridium histolyticum; EC number 3.4.24.3), elastase (pancreatic solution from porcine pancreas; EC number 3.4.21.36), and sodium azide were obtained from Sigma (St. Louis, MO). All other chemicals used were reagent grade and used as purchased without further purification. Synthesis of Peptide Cross-Linker. To synthesize the peptide cross-linker that is cleaved by matrix metalloproteinases (e.g., MMP-13) synthesized by osteoblasts, we used the peptide sequence Gln-Pro-Gln-Gly-Leu-Ala-Lys-NH2 (QPQGLAK-NH2), which matched residues 904-908 of human type II collagen26 with the addition of a glutamine residue to promote solubility and a lysine residue to provide an amine functional group for modification. The QPQGLAKNH2 was characterized by mass spectrometry, and the purity was determined by reversed-phase high-performance liquid chromatography (RP-HPLC). As shown in Scheme 1, a peptide cross-linker with bifunctional acryl groups was synthesized through the reaction between the amine groups of glutamine and lysine residues in the peptide sequence and acryloyl chloride.19,43 This produces an amide linkage between the peptide and the acrylic group. Specifically, 30 mg of peptide was dissolved in 1.2 mL of dimethylacetamide (DMAc) with triethylamine, and acryloyl chloride solution was added dropwise to the solution with stirring. The reaction temperature was kept at 0-5 °C. After all of the acryloyl chloride was added, the reaction continued with stirring for 4 h at 0-5 °C and 20 h at room temperature. Subsequently, triethylamine hydrochloride salts were removed by filtration. The resulting solution was dialyzed against ultrapure water (UPW; 18 MΩ cm) for 48 h with periodic bath changes to remove unreacted compounds. The final dialysis product was lyophilized overnight using a freeze-dryer (VIRTIS, Gardiner, NY) attached to a vacuum pump (Edwards, RV12). Synthesis of Peptide-Cross-Linked P(NIPAAm-co-AAc) Hydrogels. The loosely cross-linked P(NIPAAm-co-AAc) hydrogels were prepared with peptide cross-linker by redox polymerization in aqueous media. The method used to synthesize the peptide-cross-linked P(NIPAAm-co-AAc) hydrogels was similar to that published previously (Scheme 2).37-40 The hydrogels were prepared by varying the molar

Kim and Healy Scheme 2. Preparation of the peptide-crosslinked P(NIPAAm-co-AAc) hydrogel

ratio of NIPAAm/AAc and the amount of peptide cross-linker in the feed. The total monomer amount of NIPAAm and AAc in the feed was always 5% w/v, and NIPAAm/AAc molar ratios of 96/4, 96.5/3.5, and 97/3 were used. In addition, control P(NIPAAm) and P(NIPAAm-co-AAc) hydrogels cross-linked with N,N′-methylenebis(acrylamide) (BIS) were synthesized by a similar method to compare the peptidecross-linked hydrogels.37-40 The polymerization formulations of BIS-cross-linked P(NIPAAm), BIS-cross-linked P(NIPAAm-co-AAc), and peptide-cross-linked P(NIPAAmco-AAc) hydrogels are described in Table 1. Dry nitrogen gas was bubbled through a mixture of NIPAAm and AAc and either peptide cross-linker or BIS in phosphate-buffered saline (PBS) in a flask for 15 min to remove dissolved oxygen. Following the nitrogen gas purge, 0.8 wt % (based on total monomer) of ammonium peroxysulfate (AP) and 8% v/w (based on total monomer) of N,N,N′,N′-tetramethylethylenediamine (TEMED) was added as the initiator and accelerator, respectively. The mixture was stirred vigorously for 15 s and allowed to polymerize at room temperature for 24 h under regular fluorescent lighting in a glass vial. Following the polymerization, the peptide-cross-linked P(NIPAAm-co-AAc) hydrogel was washed three times for 15-20 min each in excess ultrapure water to remove unreacted compounds. Characterization of Peptide Cross-Linker. 1H Nuclear magnetic resonance (NMR) spectroscopy was used to identify the synthesis of the peptide cross-linker. The acrylation of amine groups in the peptide sequence was analyzed with a 300 MHz 1H NMR spectrometer (Bruker AMX-300); the spectra were recorded in D2O at room temperature. In addition, the extent of functionalization of the amine groups in the peptide was assessed by end group analysis using 2,4,6-trinitrobenzenesulfonate (TNBS). TNBS reacts with primary amines or hydrazide groups to form a highly chromogenic derivative.44-46 Thus, the unreacted amine



Table 1. Description of P(NIPAAm) and P(NIPAAm-co-AAc) Hydrogel Samples hydrogel samplesa

cross-linker

feed molar ratio (mol %)

NIPAAm/AAc monomer ratio in feedc

1

BIS-P(NIPAAm)

N,N′-methylenebis(acrylamide)

0.145

100.0/0.0

2 3 4

BIS-P(NIPAAm-co-AAc)96/4 BIS-P(NIPAAm-co-AAc)96.5/3.5 BIS-P(NIPAAm-co-AAc)97/3

N,N′-methylenebis(acrylamide) N,N′-methylenebis(acrylamide) N,N′-methylenebis(acrylamide)

0.145 0.145 0.145

96.0/4.0 96.5/3.5 97.0/3.0

5 6 7

peptide-P(NIPAAm-co-AAc)96/4d peptide-P(NIPAAm-co-AAc)96.5/3.5d peptide-P(NIPAAm-co-AAc)97/3d

peptide cross-linkerb peptide cross-linkerb peptide cross-linkerb

0.174 0.174 0.174

96.0/4.0 96.5/3.5 97.0/3.0

8 9

peptide-P(NIPAAm-co-AAc)97/3-MCe peptide-P(NIPAAm-co-AAc)97/3-HCf

peptide cross-linkerb peptide cross-linkerb

0.287 0.401

97.0/3.0 97.0/3.0

a All hydrogels were synthesized in PBS as a reaction solvent. b The substitution degree of acryl group in the peptide cross-linker is 74.33%. c The total amount of NIPAAm and AAc in feed was always 5% w/v of reaction media (PBS). d Peptide-cross-linked hydrogel with low cross-linking density; 0.174% feed molar ratio of peptide cross-linker. e Peptide-cross-linked hydrogel with medium cross-linking density; 0.287% feed molar ratio of peptide cross-linker. f Peptide-cross-linked hydrogel with high cross-linking density; 0.401% feed molar ratio of peptide cross-linker.

groups remaining within the peptide sequence react with TNBS and can be quantified by measuring the absorbance of the solution. Freeze-dried peptide cross-linker samples were dissolved in 0.1 M sodium bicarbonate solution (pH 8.5), and then a 0.01% solution of TNBS (in 0.1 M sodium bicarbonate) was added to each sample solution. After incubation at 37 °C for 2 h, 1 N HCl was added to each sample. The absorbance of the solutions was read at 335 nm with an UV-vis spectrophotometer (SpectraMax PLUS384, Molecular Devices). The substitution degree (SD) of acryl group on the amine groups in the peptide sequence was determined from a standard curve prepared using the unmodified peptide. This experiment was carried out in triplicate. LCST Measurements. The phase transition of the hydrogel samples was measured using an UV-vis spectrophotometer (SpectraMax PLUS384, Molecular Devices). The transmittance of visible light (λ ) 500 nm; path length ) 1 cm) through the hydrogel was recorded as a function of temperature. The temperature of the cuvette chamber was regulated through an electric heater, fan, and temperature sensor. Temperature was manually ramped at rates of ca. 0.1-0.25 °C/min for all runs. The LCST of hydrogel samples was determined as the abscissa of the inflection point of the transmittance vs temperature curves. Water Content Studies. Peptide-cross-linked P(NIPAAmco-AAc), BIS-cross-linked P(NIPAAm), and BIS-crosslinked P(NIPAAm-co-AAc) hydrogel samples were freezedried overnight. The freeze-dried hydrogel samples were weighed upon removal from the freeze-dryer and immersed in excess ultrapure water (UPW), Dulbecco’s phosphatebuffered saline (PBS), or cell-culture medium [i.e., Dulbecco’s modified Eagle medium (DMEM) with 10% (v/v) fetal bovine serum (FBS)] for 24 h at room temperature and 37 °C. The water content was calculated on the basis of the weight difference of the hydrogel samples before and after swelling (water content ) (Ws - Wd)/Ws × 100, where Ws is the weight of the swollen gel and Wd is the weight of the dry gel). Rheology Studies Depending on Temperature. Dynamic shear oscillation measurements were performed to characterize the viscoelastic properties of peptide-cross-linked P(NIPAAm-co-AAc) hydrogels. These rheological measure-

ments are carried out with a rheometer (MCR300, Paar Physica, Germany) using a parallel plate (50 mm diameter) configuration and plate-to-plate distance of 1.0 mm in oscillatory mode. The hydrogel samples were loaded onto the lower plate, the upper fixture was lowered, and a humidity chamber was placed around the sample to prevent dehydration during data collection. Linear viscoelastic data were collected in a constant strain mode (20%) over the frequency range of 0.001-10 Hz at 22 and at 37 °C. The temperature of the lower plate was maintained with a recirculating water bath. Preliminary experiments demonstrated that the rheological properties were independent of the applied strain in this range. Protease-Dependent Degradation Behavior of PeptideCross-Linked P(NIPAAm-co-AAc) Hydrogels. In the degradation of type II collagen, it is known that MMP-13 initially cleaves the peptide bond at Gly906-Leu907 of type II collagen and subsequently cleaves one of two resulting fragments at sites corresponding to Gly909-Gln910 and Gly912Ile913.26 Therefore, the cleavage of the peptide cross-linker, which matched residues 904-908 of human type II collagen,26 was expected between the Gly and Leu residues. As shown in Scheme 3, P(NIPAAm-co-AAc) hydrogels formed with the degradable peptide cross-linker degrade into P(NIPAAm-co-AAc) chains with peptide residues at the cross-links. At this point, further degradation of the P(NIPAAm-co-AAc) chains is unknown. To assess the enzymatic degradation of the peptide-crosslinked P(NIPAAm-co-AAc) hydrogels with MMP-13 (collagenase-3)-labile domains, the kinetics of mass loss of the hydrogel was monitored as a function of exposure to different enzymes. As the hydrogel degrades, polymer chains freed by enzyme cleavage presumably migrate out of the hydrogel and dissolve in solution, decreasing the weight of the remaining cross-linked network. The degradation behavior of the hydrogel depending on the enzyme type, the concentration of enzyme solution, and the cross-linking density within hydrogels was investigated. Forceps were used to carefully grasp a small sample (0.1-0.2 g) from the hydrogel, and scissors were used to cut the sample from the bulk hydrogel. The peptide-cross-linked hydrogel samples were sterilized in PBS with 0.2 mg/mL of sodium azide and 1 mM CaCl2 for 24 h at 37 °C. After 24 h sterilization, the

1218


Kim and Healy

Scheme 3. Schematic representation of enzymatic degradation of peptide-crosslinked P(NIPAAm-co-AAc) hydrogels by MMP-13 (collagenase-3)

initial swollen weight of each sample was determined, and the hydrogel samples then were placed into 1 mL of PBS that contained 0.2 mg/mL of sodium azide, 1 mM CaCl2, and crude collagenase (0, 0.025, 0.0050 and 0.0025 mg/mL) or elastase (0.025 mg/mL). The hydrogel samples were incubated at 37 °C, and at various times during the course of the experiment, the hydrogels were removed from the PBS or degradation media and their swollen weight was measured. The percent mass loss of hydrogel was tracked over time based on the initial swollen weight (weight change (%) ) (Wti - Wt)/Wti × 100, where Wti is the initial weight of swollen gel and Wt is the weight of gel at different time intervals in PBS or degradation media).11,19,47-50 Media were replaced every 3 days. Statistical Analyses. Analysis of variance (ANOVA) and Newman-Keuls post hoc analyses were performed using StatSoft Statistica 5.0 (Tulsa, CA). Statistical significance was determined at a value of p < 0.05. Results and Discussion Synthesis of Peptide Cross-Linker. Peptide cross-linkers were synthesized by the acrylation of two amine groups in the peptide sequence Gln-Pro-Gln-Gly-Leu-Ala-Lys-NH2 (QPQGLAK-NH2) as shown in Scheme 1. Figure 1 shows the proton NMR spectra of the original peptide sequence (QPQGLAK-NH2), and the peptide cross-linker in D2O. As seen in Figure 1b, the NMR spectrum of peptide cross-linker showed the proton peaks of acryl group (5.61, 6.06, and 6.17 ppm) and the newly formed amide bond (3.01 ppm). Acryloyl chlorides reacted with primary and secondary amines to give amides by the addition-elimination mechanism beginning with attack of the nucleophilic amine nitrogen at the carbonyl carbon. Subsequently, the liberated acid reacted to form a salt with an additional equivalent of amine such as tertiary amine.43 The presence of a tertiary

amine (e.g., triethylamine) does not interfere with amide formation by another amine, because a tertiary amine itself cannot form an amide.43 The degree of substitution was significantly influenced by the feed molar ratio of acryl group in acryloyl chloride and amine groups in the peptide. Figure 2 exhibits the effect of molar feed ratio of acryloyl chloride and peptide on the degree of substitution of the acryl group determined by amine group analysis using TNBS. As the molar ratio of acryloyl chloride increased, the degree of substitution of acryl group increased. Specifically, the substitution degree of acryl group was more than 74.33% when the molar ratio of [acryl]/ [amine] groups in feed was ∼3.2; however, [acryl]/[amine] ratios greater than 3.2 did not significantly increase the degree of substitution. LCST Characterization. The LCST observations of the peptide-cross-linked hydrogels are similar to the BIS-crosslinked gels published previously.37-40 Furthermore, all hydrogels were injectable through a 2 mm aperture without demonstrating appreciable macroscopic fracture at room temperature. Figure 3 shows the transmittance (percent) of visible light (λ ) 500 nm) through P(NIPAAm) and peptidecross-linked P(NIPAAm-co-AAc) hydrogels as a function of temperature. Each line represents a single experiment with one hydrogel sample. The P(NIPAAm) hydrogel showed a sharp phase transition at ∼32 °C, while the peptide-crosslinked P(NIPAAm-co-AAc) hydrogel exhibited an increase in LCST and a much broader phase transition. Increasing the amount of AAc in peptide-cross-linked P(NIPAAm-coAAc) hydrogel gradually increased the LCST.37,41 The hydrophilic monomer AAc strongly influenced changes in the hydrophilic/hydrophobic nature of the polymer,41,51-54 where the incorporation of more hydrophilic monomer to P(NIPAAm) hydrogels increases the LCST value because the monomer hinders the dehydration of the polymer chains and acts to expand the collapsed structure.37,38


Figure 1.

1H


NMR spectra of (a) unmodified peptide sequence (QPQGLAK-NH2) and (b) peptide cross-linker in D2O.

In addition, the transmittance vs temperature plots of BISand peptide-cross-linked P(NIPAAm-co-AAc) hydrogels with the same monomer molar ratio (NIPAAm/AAc ) 96.5/3.5) are presented in Figure 4. Although the feed molar ratio of NIPAAm/AAc was the same, the LCST of peptidecross-linked P(NIPAAm-co-AAc) hydrogels was higher than that of BIS-cross-linked hydrogels. This behavior is attributed to the incorporation of amino acids with uncharged polar side chain (e.g., proline and glutamine) or positively charged hydrophilic amino acids (e.g., lysine) within the cross-linker that could elevate the LCST similarly to hydrophilic comonomers with NIPAAm. According to the hydropathy index of Eisenberg that combines hydrophobicity and hydrophilicity of amino acid side chains,55,56 the peptide sequence QPQGLAK-NH2 used for the peptide cross-linker has a negative value (-1.61). The magnitude of the value may be considered roughly in kilocalories per mole for transfer from a hydrophobic to hydrophilic phase, and more negative values

Figure 2. Substitution degree of acryl groups in the peptide crosslinker determined by amine group analysis with TNBS.

1220


Kim and Healy

capacity would increase the LCST of P(NIPAAm-co-AAc) hydrogel as observed. Specifically, the peptide-cross-linked P(NIPAAm-co-AAc) hydrogels with a NIPAAm/AAc molar ratio of 97/3 had a LCST at ∼34.5 °C as shown in Figure 3. The LCST and phase transition behavior were almost the same with BIS-cross-linked P(NIPAAm-co-AAc) hydrogels with a NIPAAm/AAc molar ratio of 96/4 as those reported previously.37 Although the type of cross-linker did influence the LCST, the density of the peptide cross-linking did not influence the LCST of hydrogels in the range of 0.1740.401 mol % (data not shown). This result was somewhat expected, given previous observations.57-60 McPhee et al. 57 studied the influence of BIS cross-linker concentration on the transition temperature in the case of PNIPAAm latexes and concluded that the measured LCSTs were insensitive to the cross-linker in the range of investigated concentrations (0-15% w/w). Huglin et al. 58 also found that the LCST was unaffected by cross-linker concentration (0.25-1 mol %) for P(NIPAAm) and P(NIPAAm-co-AAc) hydrogels with 1-10 mol % of AAc. Water Content of P(NIPAAm) and P(NIPAAm-coAAc) Hydrogel Depending on Temperature. Table 2 shows the water content of BIS-cross-linked P(NIPAAm), BIS-cross-linked P(NIPAAm-co-AAc), and peptide-crosslinked P(NIPAAm-co-AAc) hydrogels in different swelling media at room temperature and 37 °C. At room temperature, all of the hydrogel samples exhibited water contents of >90% in UPW, PBS, and DMEM (containing 10% FBS). The water content of BIS-cross-linked P(NIPAAm) hydrogels at 37 °C significantly decreased until about 56% regardless of swelling medium (UPW, 56.7% ( 1.74%; PBS, 56.4% ( 2.23%; DMEM, 56.7% ( 1.87%), consistent with our previous observations.37 With the addition of AAc to the hydrogel, water contents of >90% were achieved, even when the temperature was above LCST. The water content of BIS-cross-linked P(NIPAAm) hydrogels did not show a significant difference among swelling media (p > 0.05), whereas the BIS-cross-linked and peptidecross-linked P(NIPAAm-co-AAc) hydrogels exhibited mediadependent behavior (p < 0.05) at room temperature and 37 °C. The water contents of both peptide-cross-linked and BIS-cross-linked P(NIPAAm-co-AAc) hydrogels were lower

Figure 3. Transmittance as a function of temperature for peptidecross-linked P(NIPAAm-co-AAc) hydrogel with NIPAAm/AAc molar ratio of (b) 96.5/3.5 and (O) 97/3 and (2) BIS-cross-linked P(NIPAAm) hydrogel.

Figure 4. Transmittance as a function of temperature for (b) peptidecross-linked P(NIPAAm-co-AAc) hydrogels with NIPAAm/AAc molar ratio of 96.5/3.5, (O) BIS-cross-linked P(NIPAAm-co-AAc) hydrogels with NIPAAm/AAc molar ratio of 96.5/3.5, and (2) BIS-cross-linked P(NIPAAm) hydrogel.

indicate more hydrophilic nature.56 Thus, the cross-linking with the peptide cross-linker with relatively hydrophilic

Table 2. Water Contenta of BIS-Cross-Linked P(NIPAAm), BIS-Cross-Linked P(NIPAAm-co-AAc), and Peptide-Cross-Linked P(NIPAAm-co-AAc) Hydrogels Depending on Temperature and Swelling Media room temperature (∼ 22 °C) UPW

PBS

DMEM (10% FBS)

UPW

PBS

DMEM (10% FBS)

BIS-P(NIPAAm)

97.6 ( 0.18

97.1 ( 0.20

97.1 ( 0.19

56.7 ( 1.74

56.4 ( 2.23

56.7 ( 1.87

hydrogel samples 1

37 °C

2

BIS-P(NIPAAm-co-AAc) 97/3

99.2 ( 0.21

98.3 ( 0.10

98.2 ( 0.07

98.6 ( 0.21

92.4 ( 1.48

91.3 ( 1.13

3 4 5

peptide-P(NIPAAm-co-AAc) 97/3b peptide-P(NIPAAm-co-AAc) 96.5/3.5b peptide-P(NIPAAm-co-AAc) 96/4b

99.3 ( 0.11 99.4 ( 0.10 99.4 ( 0.24

98.4 ( 0.12 98.4 ( 0.06 98.5 ( 0.05

98.6 ( 0.31 98.2 ( 0.05 98.3 ( 0.05

98.9 ( 0.33 98.9 ( 0.19 99.2 ( 0.08

94.4 ( 0.69 96.7 ( 0.07 97.4 ( 0.04

93.9 ( 0.48 95.8 ( 0.22 97.0 ( 0.09

6 7

peptide-P(NIPAAm-co-AAc) 97/3-MCc peptide-P(NIPAAm-co-AAc) 97/3-HCd

99.2 ( 0.20 99.1 ( 0.11

98.2 ( 0.03 97.9 ( 0.03

98.5 ( 0.19 98.2 ( 0.09

98.9 ( 0.10 98.8 ( 0.15

94.6 ( 0.58 96.2 ( 0.25

93.9 ( 0.15 96.5 ( 0.48

a Water content of hydrogel ) (W - W )/W × 100, where W is the weight of swollen gel and W is the weight of dry gel. Each water content value s d s s d represents the average of four samples. b Peptide-cross-linked hydrogel with low cross-linking density; 0.174% feed molar ratio of peptide cross-linker. c Peptide-cross-linked hydrogel with medium cross-linking density; 0.287% feed molar ratio of peptide cross-linker. d Peptide-cross-linked hydrogel with high cross-linking density; 0.401% feed molar ratio of peptide cross-linker.


in the PBS and DMEM (containing 10% FBS) cell-culture media than in the UPW (p < 0.05). This tendency was greater above the LCST as delineated in Table 2. The effect of the media on the swelling behavior can be attributed to the shielding of COO- repulsion, which prevents collapse of the gel, by the interactions between COO- groups in AAc and the ions present in the PBS and DMEM cell-culture media. In addition, the water content of peptide-cross-linked P(NIPAAm-co-AAc) hydrogels (feed molar ratio of peptide cross-linker ) 0.174 mol %) with different NIPAAm/AAc monomer molar ratios (NIPAAm/AAc ) 97/3, 96.5/3.5, and 96/4) did not change extensively at room temperature. The water content of the loosely cross-linked P(NIPAAm-coAAc) gels was not significantly influenced by the introduction of hydrophilic monomer AAc at temperatures below the LCST where PNIPAAm has a hydrophilic and expanded structure. At 37 °C, the water content of P(NIPAAm-coAAc) gels in UPW did not change significantly with the monomer molar ratio, whereas the water content of the gels in the PBS and DMEM (10% FBS) gradually increased with the AAc content in the hydrogel (p < 0.05) (Table 2). The lack of effect in UPW suggested that counterions were necessary to counteract the repulsion between COO- groups that prevent collapsing of the gel. In addition, with increasing AAc content, the water content difference of P(NIPAAmco-AAc) gels immersed in UPW and PBS (or DMEM) gradually decreased at 37 °C. This result suggests that the amount of unshielded COO- groups increased with increasing of AAc content in both PBS and DMEM. The effect of cross-linking density on the water content for peptide-cross-linked hydrogels with the same NIPAAm/ AAc monomer molar ratio of 97/3 was investigated (Table 2). As the cross-linking density within the hydrogel increased, the water content difference was not statistically significant (p > 0.05). At 37 °C, however, the difference in water content between in UPW and PBS (or DMEM) due to the shielding of COO- repulsion decreased with increasing peptide cross-linking density. For the peptide-cross-linked P(NIPAAAm-co-AAc) hydrogels with higher cross-linking density (feed molar ratio of cross-linker ) 0.401 mol %), a higher water content was observed, compared to hydrogels with lower cross-linking density (cross-linker feed molar ratio ) 0.174 and 0.287 mol %) in PBS and DMEM (p < 0.05). This result indicated that the effect of interactions between COO- groups in AAc in the hydrogel and the ions or highly charged proteins present in the PBS and DMEM (10% FBS) cell-culture media was reduced by increase of the crosslinking density. Rheological Properties of Peptide-Cross-Linked P(NIPAAm-co-AAc) Hydrogels. The viscoelastic properties of hydrogels were characterized by the complex shear modulus, G* [G*(t,ω) ) G′(t,ω) + iG′′(t,ω), where ω is the angular frequency and t is the time], which consists of the storage modulus, G′, and the loss modulus, G′′. For an oscillatory shear deformation of a gel, G′ is proportional to the stored energy, which vanishes for a whole cycle. G′′ is proportional to the dissipated energy during a cycle of deformation.61,62


Figure 5. Complex modulus (G*) of the peptide-cross-linked P(NIPAAm-co-AAc) hydrogel at 22 and 37 °C as a function of frequency at 20% strain.

Figure 5 shows the plot of complex modulus (G*) of peptide-cross-linked P(NIPAAm-co-AAc) hydrogel (PeptideP(NIPAAm-co-AAc)97/3-HC) as a function of frequency and temperature. The rheology data for the peptide-cross-linked hydrogels, depending on temperature, were similar to data previously reported for BIS-cross-linked P(NIPAAm-coAAc) hydrogels.37 Peptide-cross-linked hydrogel showed extremely soft cross-linked solid characteristics at 22 °C. The complex modulus G* was very low and approached a constant value as the frequency decreased, as shown in Figure 5. In all examined hydrogels, G′ exceeded G′′ over the entire range of frequency used (data not shown). This behavior is consistent with the dynamic mechanical behavior expected for a cross-linked network. When temperature was increased from 22 to 37 °C, the peptide-cross-linked hydrogel exhibited a significant increase in G*, which indicated that the hydrogel became a more rigid and stiff structure from the increasing hydrophobic properties of P(NIPAAm) above the LCST. These results suggest that the peptide cross-linker can provide cross-linking structure without the loss of thermoresponsive viscoelastic properties of the hydrogel and that the in situ stabilization of the cross-linked structure at body temperature (37 °C) may better support tissue growth. Enzymatic Degradation Behavior of Peptide-CrossLinked P(NIPAAm-co-AAc) Hydrogels. Proteolysis of intact collagen chains during natural tissue formation is controlled by collagenases specific to the collagen isoform, whereas smaller peptide sequences are less specifically cleaved.19 As a first step to assess the enzyme-induced degradability of the peptide-cross-linked P(NIPAAm-coAAc) hydrogels, we used crude collagenase and elastase. Figure 6 exhibits the enzymatic degradation behavior as a function of time for peptide-P(NIPAAm-co-AAc)97/3-HC hydrogels by collagenase (0, 0.0025, 0.0050, and 0.025 mg/ mL) and elastase (0.025 mg/mL). The peptide-P(NIPAAmco-AAc)97/3-HC hydrogels showed significant weight loss in the solutions containing collagenase in a concentration dependent manner (Figure 6). While hydrogels placed in 0.025 mg/mL collagenase solution were completely degraded within 6 h, the hydrogel in 0.0025 mg/mL collagenase

1222


Figure 6. Enzymatic degradation behavior of peptide-cross-linked P(NIPAAm-co-AAc) hydrogel with high cross-linking density (peptideP(NIPAAm-co-AAc)97/3-HC; 0.401% feed molar ratio of peptide cross-linker) as a function of time in PBS (containing 0.2 mg/mL sodium azide and 1 mM CaCl2): (b) 0.0025 mg/mL of collagenase; (2) 0.0050 mg/mL of collagenase; (9) 0.025 mg/mL of collagenase; (4) 0.025 mg/ml of elastase; (O) no protease. Each point represents the average of four samples.

solution showed complete degradation after 312 h of exposure time. The peptide-P(NIPAAm-co-AAc)97/3-HC hydrogels placed in blank solution without enzyme showed some decrease in their weight as a function of time. This weight loss as a function of time is probably due to the loosely cross-linked structure, which is required for a high water content and injectability of the hydrogel. During this test, the hydrogel was still a coherent mass (>70% of initial weight) and could be physically manipulated. In the case of the hydrogels placed in elastase solution (0.025 mg/mL), the weight change kinetics was similar to that of the hydrogel in the blank solution, thus indicating that the degradation of peptide-P(NIPAAm-co-AAc)97/3-HC hydrogels was specifically dependent on collagenase, as designed. In general, the degradation of hydrogels in solution (mass loss from the networks) is linked to several network parameters such as number of cross-links per backbone chain, number of vinyl groups on the cross-linking molecule, molecular weight of backbone, and proportion of degradable groups in the main and side chain.15,63-65 The degradation behavior of peptide-cross-linked hydrogels depended on cross-linking density. Figure 7 shows the degradation profiles of peptide-cross-linked hydrogels prepared by using different feed molar ratios of the peptide cross-linker (detailed compositions of hydrogels are given in Table 1) in 0.0025 mg/mL collagenase solution. The degradation rate gradually decreased with increasing cross-linking density. As the crosslinking density increased, additional degradable units must be broken to degrade the gel. This effect results in longer inhibition time for the mass loss with increasing cross-linking density. Conclusions Injectable P(NIPAAm-co-AAc) hydrogel scaffolds with proteolytic degradability were prepared by incorporation of an oligopeptide cross-linker that was specifically cleaved by

Kim and Healy

Figure 7. Degradation behavior as a function of time for peptidecross-linked P(NIPAAm-co-AAc) hydrogels with different cross-linking density in PBS (containing 0.2 mg/mL sodium azide and 1 mM CaCl2): (O) peptide-P(NIPAAm-co-AAc)97/3-HC (0.401 mol % peptide cross-linker) in the PBS blank solution without protease; (b) peptide-P(NIPAAm-co-AAc)97/3-HC (0.401 mol % peptide crosslinker) in the 0.0025 mg/mL collagenase solution; (2) peptideP(NIPAAm-co-AAc)97/3-MC (0.287 mol % peptide cross-linker) in the 0.0025 mg/mL collagenase solution; (9) peptide-P(NIPAAm-co-AAc)97/3 (0.174 mol % peptide cross-linker) in the 0.0025 mg/mL collagenase solution. Each point represents the average of four samples.

collagenase. The peptide-cross-linked P(NIPAAm-co-AAc) hydrogels retained some of the attractive properties of BIScross-linked hydrogels, such as injectability through a 2 mm diameter aperture at room temperature. Furthermore, the LCST of peptide-cross-linked P(NIPAAm-co-AAc) hydrogels was significantly influenced by monomer molar ratio of NIPAAm/AAc, whereas cross-linking density did not affect the LCST. The peptide-cross-linked hydrogels maintained more than 90% water content at 37 °C, above the LCST, which was controlled by the molar ratio of NIPAAm/ AAc, swelling media, and cross-linking density. Moreover, the hydrogels were specifically degraded by collagenase as a function of time at 37 °C, and the degradation rate depended on the concentration of collagenase and crosslinking density of hydrogel. Enzymatically degradable domains are a necessary feature for the design of artificial ECMs that promote regeneration of tissues in situ. Thus, the addition of peptide-cross-linked P(NIPAAm-co-AAc) hydrogels, containing enzymatically degradable domains, to the arsenal of artificial ECMs is an exciting development. Acknowledgment. This work was funded by a grant from the N.I.H. National Institute of Arthritis and Musculoskeletal and Skin Diseases, Grant AR47304. References and Notes (1) Griffith, L. G. Acta Mater. 2000, 48, 263-277. (2) Seal, B. L.; Otero, T. C.; Panitch, A. Mater. Sci. Eng. R 2001, 34, 147-230. (3) Fuchs, J. R.; Nasseri, B. A.; Vacanti, J. P. Ann. Thorac. Surg. 2001, 72, 577-591. (4) Kim, B. S.; Mooney, D. J. Trends Biotechnol. 1998, 16, 224-230. (5) Lee, K. Y.; Mooney, D. J. Chem. ReV. 2001, 101, 1869-1879. (6) LeBaron, R. G.; Athanasiou, K. A. Tissue Eng. 2000, 6, 85-103. (7) Chen, G. P.; Ushida, T.; Tateishi, T. Macromol. Biosci. 2002, 2, 6777.

Injectable P(NIPAAm-co-AAc) Hydrogels (8) Lee, K. Y.; Alsberg, E.; Mooney, D. J. J Biomed. Mater. Res. 2001, 56, 228-233. (9) Lee, K. Y.; Bouhadir, K. H.; Mooney, D. J. Macromolecules 2000, 33, 97-101. (10) Mann, B. K.; West, J. L. J. Biomed. Mater. Res. 2002, 60, 86-93. (11) Mann, B. K.; Gobin, A. S.; Tsai, A. T.; Schmedlen, R. H.; West, J. L. Biomaterials 2001, 22, 3045-3051. (12) Schmedlen, R. H.; Masters, K. S.; West, J. L. Biomaterials 2002, 23, 4325-4332. (13) Vernon, B.; Kim, S. W.; Bae, Y. H. J. Biomater. Sci., Polym. Ed. 1999, 10, 183-198. (14) Vernon, B.; Kim, S. W.; Bae, Y. H. J. Biomed. Mater. Res. 2000, 51, 69-79. (15) Nuttelman, C. R.; Henry, S. M.; Anseth, K. S. Biomaterials 2002, 23, 3617-3626. (16) Agrawal, C. M.; Ray, R. B. J. Biomed. Mater. Res. 2001, 55, 141150. (17) Hubbell, J. A. Curr. Opin. Solid State Mater. Sci. 1998, 3, 246251. (18) Sakiyama-Elbert, S. E.; Hubbell, J. A. Annu. ReV. Mater. Res. 2001, 31, 183-201. (19) West, J. L.; Hubbell, J. A. Macromolecules 1999, 32, 241-244. (20) Halstenberg, S.; Panitch, A.; Rizzi, S.; Hall, H.; Hubbell, J. A. Biomacromolecules 2002, 3, 710-723. (21) Winchester, S. K.; Bloch, S. R.; Fiacco, G. J.; Partridge, N. C. J. Cell. Physiol. 1999, 181, 479-488. (22) Curran, S.; Murray, G. I. Eur. J. Cancer 2000, 36, 1621-1630. (23) Massova, I.; Kotra, L. P.; Fridman, R.; Mobashery, S. FASEB J. 1998, 12, 1075-1095. (24) Johansson, N.; Saarialho-Kere, U.; Airola, K.; Herva, R.; Nissinen, L.; Westermarck, J.; Vuorio, E.; Heino, J.; Kahari, V. M. DeV. Dyn. 1997, 208, 387-397. (25) Bilezikian, J. P., Raisz, L. G., Rodan, G. A., Eds. Principles of Bone Biology; Academic Press: San Diego, CA, 1996. (26) Mitchell, P. G.; Magna, H. A.; Reeves, L. M.; LoprestiMorrow, L. L.; Yocum, S. A.; Rosner, P. J.; Geoghegan, K. F.; Hambor, J. E. J. Clin. InVest. 1996, 97, 761-768. (27) Knauper, V.; LopezOtin, C.; Smith, B.; Knight, G.; Murphy, G. J. Biol. Chem. 1996, 271, 1544-1550. (28) Freije, J. M. P.; Diezitza, I.; Balbin, M.; Sanchez, L. M.; Blasco, R.; Tolivia, J.; Lopezotin, C. J. Biol. Chem. 1994, 269, 16766-16773. (29) Gutowska, A.; Jeong, B.; Jasionowski, M. Anat. Rec. 2001, 263, 342349. (30) Temenoff, J. S.; Mikos, A. G. Biomaterials 2000, 21, 2405-2412. (31) Peter, S. J.; Miller, M. J.; Yasko, A. W.; Yaszemski, M. J.; Mikos, A. G. J. Biomed. Mater. Res. 1998, 43, 422-427. (32) Atala, A.; Cima, L. G.; Kim, W.; Paige, K. T.; Vacanti, J. P.; Retik, A. B.; Vacanti, C. A. J. Urol. 1993, 150, 745-747. (33) Marler, J. J.; Guha, A.; Rowley, J.; Koka, R.; Mooney, D.; Upton, J.; Vacanti, J. P. Plast. Reconstr. Surg. 2000, 105, 2049-2058. (34) Cao, Y. L.; Rodriguez, A.; Vacanti, M.; Ibarra, C.; Arevalo, C.; Vacanti, C. A. J. Biomater. Sci., Polym. Ed. 1998, 9, 475-487. (35) Duranti, F.; Salti, G.; Bovani, B.; Calandra, M.; Rosati, M. L. Dermatol. Surg. 1998, 24, 1317-1325. (36) Sims, C. D.; Butler, P. E. M.; Casanova, R.; Lee, B. T.; Randolph, M. A.; Lee, W. P. A.; Vacanti, C. A.; Yaremchuk, M. J. Plast. Reconstr. Surg. 1996, 98, 843-850.

Biomacromolecules, Vol. 4, No. 5, 2003 1223 (37) Stile, R. A.; Burghardt, W. R.; Healy, K. E. Macromolecules 1999, 32, 7370-7379. (38) Stile, R. A.; Healy, K. E. Biomacromolecules 2001, 2, 185-194. (39) Stile, R. A.; Healy, K. E. Biomacromolecules 2002, 3, 591-600. (40) Stile, R. A.; Burghardt, W. R.; Healy, K. E. Biomacromolecules, submitted for publication, 2002. (41) Schild, H. G. Prog. Polym. Sci.ence 1992, 17, 163-249. (42) Hoffman, A. S. MRS Bull. 1991, 16, 42-46. (43) Loudon, G. M. Organic Chemistry, 3rd ed.; Benjamin/Cummings Publishing Company, Inc.: Redwood City, CA, 1995. (44) Hermanson, G. T. Bioconjugate Techniques; Academic Press Inc.: San Diego, CA, 1996. (45) Bubnis, W. A.; Ofner, C. M. Anal. Biochem. 1992, 207, 129-133. (46) Morcol, T.; Subramanian, A.; Velander, W. H. J. Immunol. Methods 1997, 203, 45-53. (47) Yoshida, T.; Aoyagi, T.; Kokufuta, E.; Okano, T. J. Polym. Sci., Part A: Polym. Chem. 2003, 41, 779-787. (48) Kumashiro, Y.; Lee, T. K.; Ooya, T.; Yui, N. Macromol. Rapid Commun. 2002, 23, 407-410. (49) Kurisawa, M.; Matsuo, Y.; Yui, N. Macromol. Chem. Phys. 1998, 199, 705-709. (50) de Jong, S. J.; van Eerdenbrugh, B.; van Nostrum, C. F.; Kettenesvan de Bosch, J. J.; Hennink, W. E. J. Controlled Release 2001, 71, 261-275. (51) Yoshida, R.; Sakai, K.; Okano, T.; Sakurai, Y. J. Biomater. Sci., Polym. Ed. 1994, 6, 585-598. (52) Inoue, T.; Chen, G. H.; Nakamae, K.; Hoffman, A. S. Polym. Gels Networks 1997, 5, 561-575. (53) Ebara, M.; Aoyagi, T.; Sakai, K.; Okano, T. Macromolecules 2000, 33, 8312-8316. (54) Zhang, J.; Peppas, N. A. Macromolecules 2000, 33, 102-107. (55) Eisenberg, D. Annu. ReV. Biochem. 1984, 53, 595-623. (56) Ratner, B. D., Hoffman, A. S., Schoen, F. J., Lemons, J. E., Eds. Biomaterials Science; An Introduction to Materials in Medicine; Academic Press: San Diego, CA, 1997. (57) McPhee, W.; Tam, K. C.; Pelton, R. J. Colloid Interface Sci. 1993, 156, 24-30. (58) Huglin, M. B.; Liu, Y.; Velada, J. L. Polymer 1997, 38, 5785-5791. (59) Duracher, D.; Elaissari, A.; Pichot, C. J. Polym. Sci., Part A: Polym. Chem. 1999, 37, 1823-1837. (60) Hua, J. D.; Liu, Y. F.; Hu, J.; Wang, Q. Q.; Gong, Z. B.; Guo, X. Z. J. Appl. Polym. Sci. 1999, 74, 2457-2461. (61) Borchard, W.; Lechtenfeld, M. Mater. Res. InnoVations 2001, 4, 381387. (62) Kavanagh, G. M.; Ross-Murphy, S. B. Prog. Polym. Sci. 1998, 23, 533-562. (63) Metters, A. T.; Bowman, C. N.; Anseth, K. S. J. Phys. Chem. B 2000, 104, 7043-7049. (64) Martens, P.; Metters, A. T.; Anseth, K. S.; Bowman, C. N. J. Phys. Chem. B 2001, 105, 5131-5138. (65) Anseth, K. S.; Metters, A. T.; Bryant, S. J.; Martens, P. J.; Elisseeff, J. H.; Bowman, C. N. J. Controlled Release 2002, 78, 199-209.

BM0340467

Synthesis and Characterization of Injectable Poly (N

Recommend Documents