Synthesis and Characterization of Radiopaque Iodine-containing ...

55 downloads 0 Views 290KB Size Report
Nov 30, 2007 - The product was isolated as a fluffy yellowish powder after lyophilisation ..... (19) Davy, K. W. M.; Anseau, M. R.; Odlyha, M.; Foster, G. M. Polym.
Biomacromolecules 2008, 9, 263–268

263

Synthesis and Characterization of Radiopaque Iodine-containing Degradable PVA Hydrogels Damia Mawad,† Laura A. Poole-Warren,† Penny Martens,† Leo H. Koole,‡ Tristan L. B. Slots,‡ and Catharina S. J. van Hooy-Corstjens*,‡ Graduate School of Biomedical Engineering, University of New South Wales, Sydney, New South Wales 2052, Australia, and Centre for Biomaterials Research, University of Maastricht, PO Box 616, 6200 MD Maastricht, The Netherlands Received July 9, 2007; Revised Manuscript Received September 25, 2007

Poly (vinyl alcohol) (PVA) hydrogels are highly attractive for biomedical applications, especially for controlled release of drugs and proteins. Recently, degradable PVA hydrogels have been described, having the advantage that the material disappears over time from the implantation site. Herein, we report the synthesis of radiopaque degradable PVA, which gives a further advantage that the position of the hydrogel can precisely be determined by X-ray fluoroscopy. Radiopacity has been introduced by replacing 0.5% of the pendent alcohol groups on the PVA with 4-iodobenzoylchloride. This level of substitution rendered the polymer adequately radiopaque. The subsequent modification of 0.8% of the pendent hydroxyl groups with an ester acrylate functional group allowed for cross-linking of the macromers. The radiopaque hydrogels degraded over a time span of 140 days. Rheology data suggested that the macromer solutions were appropriate for injection.

1. Introduction Biocompatible synthetic hydrogels play a central role in biomaterials science, primarily in view of applications related to controlled local release of drugs, growth factors, or cells.1 The idea to use a polymer implant as a matrix for local delivery of a drug is approximately 40 years old. Initially, stable polymers, such as silicone rubber2 and poly(ethylene)3 were explored. In terms of hydrogels, poly(vinyl alcohol) (PVA) is an attractive material in this respect. PVA is highly hydrophilic, biocompatible, and safe; the material is widely applied in medicine and personal health care.4,5 PVA-based hydrogels have found widespread application as devices for controlled release of drugs and proteins. Traditionally, PVA hydrogels were based on nondegradable PVA polymer and prepared by either freeze/ thaw techniques or chemically cured.5,6 However, use of such types of hydrogels implies that the reservoir remains at the site of implantation after drug release is completed. In the 1970s, it was proposed to build drug release strategies on polymers that undergo structural decomposition (e.g., via hydrolysis) after implantation.7,8 In this way, the requirement for surgical removement can be circumvented, and furthermore, it gives the advantage that the release of the drug can be tuned by the degradation rate. Recently, it was shown that PVA can be chemically modified to achieve a degradable polymer network with controllable decomposition characteristics.9,11 Synthetically, the approach comprises two steps: (i) PVA is reacted with anhydride 1, leading to functionalization of the polymer with pendent acrylate groups,9,10 and (ii) the acrylate groups are polymerized, resulting in the formation of a three-dimensional network. Both photopolymerization and redox-initiated polymerization have been used.11 * Corresponding author. E-mail: [email protected]. † University of New South Wales. ‡ University of Maastricht.

The resulting networks decompose through hydrolysis of ester groups in the cross-links and subsequent erosion of (modified) PVA from the hydrogel network. It has been shown that the degradation profiles of the degradable PVA networks can be tailored to suit different requirements. Variations of parameters such as the degree of substitution, the hydrophilicity of the polymer backbone, and the macromer concentration resulted in cross-linked networks with different material properties, such as swelling ratio and total degradation time.9,11 Herein, we report a further advantageous modification of these PVA-based degradable hydrogels; we introduced covalently bound iodine in these materials with the aim to render them radiopaque, i.e. X-ray visible. This was done with preservation of the essential features: (i) injectability; (ii) in situ polymerization (network formation); (iii) hydrolytic degradation; (iv) water solubility of all degradation products. X-ray visibility facilitates accurate placement of the material during injection, since the material can be monitored continuously in real time. Moreover, by examining the radiopacity at the implant site in time, information about the degradation and, hence, about the drug-release can be obtained in a noninvasive way. One particular biomedical application that could benefit from this technology would be embolic agents, especially for the treatment of inoperable solid tumors. These need to be delivered precisely at the targeted site, which requires real-time visualization through X-ray fluoroscopy. Currently, liquid embolics are usually mixed with inorganic contrast agents such as bismuth oxide, metrizamide, and tantalum before being injected.12,13 While these agents introduce radiopacity to the liquid embolics, they change the properties of the injected solution such as increasing the viscosity, which makes it harder for the liquid

10.1021/bm700754m CCC: $40.75  2008 American Chemical Society Published on Web 11/30/2007

264 Biomacromolecules, Vol. 9, No. 1, 2008

embolic to be delivered through a microcatheter. Also, simply mixing the contrast agents with the solution does not eliminate the probability of sedimentation of the contrast agent inside the catheter as it is delivered, or its slow release into the surrounding tissue, making it difficult for clinical follow up.14 In addition, the toxicity of these agents is not well assessed,15 and bismuth in particular can have adverse effects, such as neurological syndromes.16,17 Over the years, research into the use of covalently bound iodine to impart radiopacity into implant biomaterials has gained increasing interest.18–28 So far, mostly nondegradable polymers, typically based on methacrylate derivates, have been rendered radiopaque. To the best of our knowledge, there is only one (very recent) example of the use of iodine to make biodegradable polymers radiopaque.29 These workers modified poly(-caprolactone) in two consecutive synthetic steps: (i) generation of carbanions through treatment with lithium di-isopropylamide (abstraction of H+), and (ii) reaction with I2. The resulting material featured slow hydrolytic degradation (30% mass loss after 70 weeks) as well as radiopacity. Degrading, iodinecontaining hydrogels have not been described so far. Now, we report the preparation of radiopaque, degradable PVA hydrogels. Synthesis, radiopacity, rheological behavior, and degradation have all been addressed.

2. Experimental Section 2.1. Materials. All chemicals were purchased from Sigma and used without further purification, unless stated otherwise. Dimethyl sulfoxide (DMSO) was distilled from calcium hydride under reduced pressure and used directly. Triethylamine (TEA), pyridine and dichloromethane (DCM) were distilled from calcium hydride and stored over either 3 Å molecular sieves (pyridine, DCM) or potassium hydroxide pellets (TEA). The photoinitiator Irgacure 2959 (2-hydroxy-1-[4-(2-hydroxyethoxy)phenyl]-2-methyl-1-propanone) was purchased from Ciba. Products of each synthetic step (10 mg) were dissolved in deuterated water (1 mL) and run through 1H NMR using a Bruker DPX 300 system operating at 300 MHz for 1H at 28–30 °C. Processing was done using Bruker WINNMR-1D software package. The % of modification in each step was calculated by comparing the area under the integrals corresponding to the phenyl or vinyl protons to the area under the peak corresponding to the methylene protons of the PVA backbone adjacent to the hydroxyl group at δ ) 3.8–4.1 ppm. 2.2. Synthesis of Degradable Radiopaque Based-PVA Hydrogels. The synthesis consisted of three consecutive steps: Step 1: Introduction of CoValently Bound Iodine to PVA. PVA powder (6.00 g, MW ) 16 kDa, 98% hydrolyzed, Aldrich) was mixed with dry pyridine (30 mL) in a 250 mL round-bottom flask, and the mixture was stirred for two hours. The pyridine was subsequently removed using a rotary evaporator. The residue was dissolved in dry DMSO (100 mL) under gentle heating (60 °C) and under exclusion of moisture. The solution was stirred magnetically until all PVA dissolved, and then cooled to 45 °C. A solution of 4-iodobenzoyl chloride (3.00 g, 11.5 mmol) in DCM (6 mL) was then added dropwise. TEA was added in excess (approximately 6 mL). The mixture was allowed to cool to room temperature, and stirring was continued for 2 h. The contents of the flask were poured into an 80/20 solvent mixture of diethylether and acetone (approximately 1 L). The precipitate was collected, washed with water, and dried under vacuum. The yield of the reaction was 4.5 g. Step 2: Introduction of the Hydrolyzable Ester Group. The product of step 1 (4.5 g) was dissolved in dry DMSO (20.5 mL) in a 250 mL round-bottom flask and reacted with anhydride (1) (3,8,11,16-tetraoxo4,7,12,15-tetraoxanonadec-1-en-19-oic anhydride;10 14.1 g, 20.0 mmol) following the protocol described in ref 9. After the reaction, the contents of the flask were poured in acetone (1 L). The precipitate was collected

Mawad et al. and left to dry overnight. The product was purified further by dissolving it in water (8 mL). The solution was treated for 6 h in an Amicon stirred ultrafiltration cell with a membrane of molecular weight cutoff of 10 kDa. The product was isolated as a fluffy yellowish powder after lyophilisation. The yield of the reaction was 3.5 g. Step 3: Network Formation. Hydrogels were formed by chemical polymerization using either UV light or redox curing. The following aqueous stock solutions of initiators were prepared: Irgacure 2959 (stock solution 10.00 mg/mL), ferrous gluconate dihydrate (stock solution 20.00 mg/mL), ascorbic acid (stock solution 73.08 mg/mL), and hydrogen peroxide (H2O2, 4.147 M). UV Curing. Macromer solution for UV curing was prepared by dissolving product of step 2 (1.0 g) in water (3.5 mL) at 80 °C. Then, Irgacure 2959 (0.5 mL of the stock solution) was added. The macromer solution was sucked in a syringe (1 mL), and then exposed to UV light for 3 min using a Green Spot UV (UV Source, Torrance, CA) (peaks at 310 and 365 nm, and intensity of 1.79 ( 0.09 W/cm2). Redox Curing. Redox cured hydrogels were prepared by mixing two solutions of same polymer concentrations but one containing a reductant component (Fe2+) and the other solution an oxidant (H2O2). To prepare the reductant solution, product of step 2 (400.0 mg) was dissolved in water (1.528 mL). Then, ferrous gluconate dihydrate (40 µL of the stock solution) and ascorbic acid (32 µL of the stock solution) were added and the solution was stirred. The oxidant solution was prepared by dissolving the same amount of polymer in water (1.568 mL), followed by the addition of H2O2 (32 µL of the stock solution). Two syringes (1 mL) containing equal volumes of each solution were connected via a dual connector. The plungers of the syringes were pushed back and forth until resistance was felt denoting gelation of the solutions. In both cases, the tip of the syringe was cut and the hydrogels were pushed out. All hydrogels were obtained as flexible cylindrical rods (diameter 4.8 mm). The rods were cut into parts of 5 mm length to be used in the degradation studies. 2.3. Radiopacity. To investigate the radiopacity of the modified PVA (step 2), 190 mg polymer was compressed into a cylinder with a diameter of 6 mm. The resulting height of the cylinder was 5.3 mm. Subsequently, the cylinder was implanted subcutaneously in a mouse cadaver, and an X-ray image was recorded with a mammography imaging system (Bennett Trex Medical), operated at 25 kV and 7.9 mAs. In an analogue experiment, the same cylinder was implanted subcutaneously in the back of a Wistar rat cadaver, together with a cylinder of unmodified PVA as control material. The X-ray image was again taken with the mammography imaging system. 2.4. Rheological Characterization. Test solutions were prepared by dissolving the acrylate/iodine modified polymer in water (60 °C) to final concentrations of 10, 15, and 20 wt %. Control solutions were prepared from unmodified PVA and PVA modified with only the acrylate group. Shear viscosity (η) was measured using Advanced Rheometrics Expanded System (ARES) rheometer equipped with parallel plate geometry (25 mm in diameter). A 0.5 mL volume of each test solution was dispensed on the bottom plate. The top plate was then brought down and pressed on the sample until a meniscus was formed on the outside of the plates. Measurements were conducted in duplicate at room temperature. 2.5. Degradation Studies. Hydrogel specimens (cylinders with diameter 4.8 mm and length 5 mm, vide supra) were weighed (mi) at fabrication and then transferred to permeable tissue embedding cassettes (1 specimen per cassette; 28 × 32 × 7 mm Aurion, Wageningen, The Netherlands). The samples were incubated in phosphate buffered saline solution (PBS, pH 7.4) at 37 °C. At different time points, three tissue cassettes were removed from the degrading medium, blotted dry, and weighed to obtain the swollen weight of the hydrogels (ms). The hydrogels were lyophilised, and then reweighed to get the dry weight of the polymer remaining in the hydrogels (md). The initial dry polymer

Radiopaque Iodine-Containing Degradable PVA Hydrogels

Biomacromolecules, Vol. 9, No. 1, 2008 265

Figure 1. (A) 1H NMR of modified PVA (step 2) dissolved in D2O. The peaks at δ ) 1.4–1.8 ppm and δ ) 3.8–4.1 ppm correspond to the methylene protons and the methylene proton adjacent to the hydroxyl group of the PVA backbone. Expansions of 1H NMR: (B) a and b are broadened peaks (δ ) 7.7 and 7.9 ppm) corresponding to the aromatic protons of the phenyl ring, (C) d and c are vinyl peaks (δ ) 6.4, 6.1, and 5.8 ppm) corresponding to the protons of the acrylate end group attached to the ester.

mass (mid) was calculated by multiplying mi by the weight fraction of the polymer in the initial solutions. The % mass loss was calculated as follows:

% mass loss )

(mid - md) x 100 mid

(1)

The volumetric swelling ratio Q was calculated from4

Q)1+

Fpolymer (q - 1) Fsolvent

(2)

where Fpolymer ) 1.2619 g/cm3, Fsolvent is the density of water (1 g/cm3) and q is the mass swelling ratio calculated as

q)

ms md

(3)

The % average mass loss and the average volumetric swelling ratio were calculated from the mean of the three replicates at each time points. The swelling data was fitted to an exponential model according to30 3

Q ≈ e5

k′t

(4)

The pseudo first order degradation constant, k′, and its standard error were calculated by performing linear regression on the logarithm of Q versus time.

3. Results and discussion 3.1. Synthesis of Degradable Radiopaque Based-PVA Hydrogels. The reaction of PVA with 4-iodobenzoyl chloride as well as the attachment of the acrylate group through 1 proceeded smoothly. Purity and identity of the product were checked with 1H NMR at 300 MHz. Figure 1A shows the spectrum of PVA after modification with both types of functional groups (step 2). Peaks at δ ) 7.7 ppm and δ ) 7.9 ppm in the subspectrum shown in Figure 1B correspond to aromatic protons of a phenyl ring. Broadening of these peaks confirms that the aromatic iodine-containing ring is indeed attached to the polymer chain. Peaks at δ ) 6.4 ppm, δ ) 6.1 ppm, and δ ) 5.8 ppm (Figure 1C) correspond to the vinyl protons of the acrylate end group. According to the 1H NMR spectrum, PVA backbone is modified such that the number of iodine-carrying blocks: number of unmodified blocks is 1: 211, and the number of acrylate-containing blocks: number of unmodified blocks is 1:119, implying that 0.5 and 0.8% of hydroxyl groups have been modified by the iodine and acrylate containing groups, respectively. Cross-linking, both by redox-initiation and photoinitiation proceeded in a straightforward manner and it was verified visually when no residual liquid was observed.

266 Biomacromolecules, Vol. 9, No. 1, 2008

Mawad et al. Table 1. Shear Viscosity η (mPa.s) of PVA Solutions

Figure 2. (A) X-Ray image of modified PVA (step 2) compressed to cylindrical shape (diameter 6 mm, height 5.3 mm) and implanted subcutaneously in a mouse; (B) X-ray image of modified PVA implanted subcutaneously in a rat together with a control material (unmodified PVA). (C) X-ray image in which the contrast has been altered such that soft tissue cannot be distinguished anymore.

3.2. Radiopacity. Figure 2A shows the X-ray image of the mouse cadaver with the iodine-containing PVA cylindrical implant (arrow). The implant is clearly visible; its level of contrast is comparable to that of the cervical vertebrae. This implies that most likely an implant of this size will also be visible if implanted in a larger animal model, or in a human subject. Figure 2B shows a comparable X-ray image of a rat cadaver. Two cylindrical implants were placed subcutaneously: one composed of the iodine-containing material, and one composed of unmodified PVA, serving as a radiolucent control. A clear difference in contrast between the two materials is noted, especially after some image manipulation (changing the contrast such that soft tissue cannot be distinguished anymore, Figure 2C). In this case the iodine-containing PVA cylinder can still clearly be distinguished, whereas the control material is only weakly and partially visible. These observations also lead us to expect that the new material would be retrievable if implanted in a human patient. Note that the size of the implant compares well with the lumen of an artery of intermediate calibre. Figures 2A-C reveal that the new biomaterial features reasonable X-ray contrast, although its iodine content is, in fact, quite low. Note that only approximately 1 out of every 200 PVA hydroxyl groups carries an iodine-containing group. It is by no means certain that the ratio 1: 200 is optimal. Increasing the ratio, e.g., to 1:100, or 1:50 would lead to a more clear X-ray contrast, but this would go at the expense of (i) decreased water

[polymer]

iodine and acrylate modified PVA (step 2)

acrylate modified PVA

10% 15% 20%

26.15 (0.26) 113.53 (0.28) 398.15 (3.20)

14.67 (0.73) 54.13 (0.87) 151.83 (4.72)

unmodified PVA 9.07 (0.15) 130.73 (3.05)

solubility of the precursor materials, as well as the degradation products; (ii), slower degradation, since the material becomes more hydrophobic; (iii), higher viscosity, which may hamper the injection or may require thicker injection catheter tubes. Most likely, the optimum composition of these materials will have to be carefully engineered on the basis of specific requirements for each intended application. 3.3. Rheological Characterization of Modified PVA. Viscosity of aqueous solutions of PVA carrying acrylate and iodine is of interest, in view of their injectability. Table 1 compiles the shear viscosity (η), measured for 10, 15, and 20% solutions of this material. The viscosity shows a steep increase as a function of the concentration. This result is expected due to increase in hydrodynamic interactions that occur in more concentrated solutions causing resistance to solute molecules and hence an increase in viscosity. For comparison, Table 1 also lists the shear viscosity of solutions of unmodified PVA and PVA that merely carries acrylate groups. It is clear that the incorporation of acrylate groups has a small influence on the viscosity of the PVA solutions, whereas the incorporation of iodine leads to a marked increase in viscosity. This is explained by enhanced hydrophobic interchain interactions of the iodinated aromatic groups. Literature data on injectable hydrogels show that aqueous solutions with a viscosity of approximately 700 mPa.s behave as viscous liquids that flow easily and are injectable through a 20 gauge needle.31 Based on these data it can be reasoned that the solutions containing modified PVA as described in this work, with the highest viscosity of 400 mPa.s for the 20% radiopaquePVA solution, should be easily injectable. 3.4. Degradation of Modified PVA. Degradation studies of radiopaque degradable PVA hydrogels are compared to a similar study conducted on nonradiopaque degradable PVA hydrogels modified with the same number of hydrolyzable ester groups but containing no iodine.11 Figure 3 presents the data as obtained for the redox-cured systems. The swelling profile (Figure 3A) shows two characteristics. The subgraph in Figure 3A shows the first characteristic region from day 0 to day 1 during which the hydrogel starts swelling immediately to reach thermodynamic equilibrium with its environment. Once the hydrogels reach equilibrium, an exponential increase in their swelling ratio is observed. The degradation follows pseudo first-order hydrolysis kinetics as the exponential fit shows. The values of the degradation rate constant k′ calculated according to eq 4 are reported in Table 2. It can be seen that the radiopaque hydrogel has a lower k′ value than the nonradiopaque hydrogel. Figure 3B shows the mass loss profiles, from which two important features can be extracted. The first feature is the mass loss determined at day 1 and accounted for as sol fraction (i.e., the fraction of polymer chains that are not connected to the network). Table 3 represents the sol fractions determined for radiopaque and nonradiopaque degradable hydrogels. It is clear that the radiopaque degradable network displays a significantly lower sol-fraction than the comparable nonradiopaque degradable network. The second feature is the rate of mass loss and the overall time required for complete degradation (see Table 3). These data show that the radiopaque degradable network

Radiopaque Iodine-Containing Degradable PVA Hydrogels

Biomacromolecules, Vol. 9, No. 1, 2008 267 Table 3. Sol Fraction and Degradation Time of Radiopaque and Nonradiopaque Degradable Uv and Redox Cured Hydrogelsa polymer radiopaque degradable PVA nonradiopaque degradable PVA radiopaque degradable PVA nonradiopaque degradable PVA a

curing method

sol fraction

total degradation time (days)

UV

12.12 (2.85)

146 (6)

UV

23.27 (2.04)

73 (1)

redox

28.12 (2.92)

136 (7)

redox

45.65 (8.53)

49 (2)

Mean (standard deviation) (n ) 3).

Figure 3. Degradation profiles of redox cured radiopaque degradable hydrogels. (A) volumetric swelling ratio (Q) plotted in a logarithmic scale for a 20% hydrogel (2). For comparison, the results for the nonradiopaque degradable PVA hydrogel11 have also been presented (4). Hydrogels reached equilibrium after one day incubation (a). The encircled point highlights the last Q measured before the network fully degraded and was not included in the exponential fit. (B) Average mass loss for the radiopaque (2) and nonradiopaque (4)11 hydrogel. Error bars represent standard deviations calculated from three measures at each time point. Table 2. Degradation Rate Constant k′ (day-1) for 20 wt% Radiopaque and Nonradiopaque Degradable PVA Hydrogels polymer radiopaque degradable PVA nonradiopaque degradable PVA radiopaque degradable PVA nonradiopaque degradable PVA

curing method redox redox UV UV

k′ (day-1) 0.018 (0.002) 0.046 (0.005) 0.020 (0.001) 0.046 (0.009)

degrades over a significantly longer time (136.0 ( 7.2 days) than the nonradiopaque network (49.0 ( 1.7 days). Degradation studies for UV-cured samples (Figure 4, Table 2,3) show that, in agreement with the results obtained earlier on nonradiopaque hydrogels,11 the kinetics of UV-cured samples strongly resemble those of the redox-cured samples. In other words, the type of network-curing hardly influences the degradation profiles. The swelling and mass loss results indicate that the attachment of the hydrophobic aromatic ring causes a decrease in the hydrophilicity of the network. A more hydrophobic hydrogel system should have a lower uptake of water, which will ultimately result in lower swelling values and a longer degradation time. In addition, hydrophobic chains in solution have the

Figure 4. Degradation profiles of UV-cured radiopaque degradable hydrogels. (A) volumetric swelling ratio (Q) plotted in a logarithmic scale for a 20% hydrogel (2). For comparison, the results for the nonradiopaque degradable PVA hydrogel11 have also been presented (4). Hydrogels reached equilibrium after one day incubation (a). (B) Average mass loss for the radiopaque (2) and nonradiopaque (4)11 hydrogel. Error bars represent standard deviations calculated from three measures at each time point.

tendency to form interchain interactions bringing the macromolecules together. More interchain interactions in solution can result in a more efficient cross-linking (i.e., more cross-links, less cyclization), which explains the lower sol fractions that were observed. Finally, we performed a preliminary analysis of the cytotoxicity of the degradation products. The cell growth inhibition assay, as described previously,11 led to cell-growth inhibition ratios of 29% for the degradation products of the UV-cured iodine-containing PVA hydrogel, and 26% for the redox-cured iodine-containing PVA hydrogel. These values correspond very well with the data that were found earlier for the non-iodinecontaining counterpart materials which had inhibition of

268 Biomacromolecules, Vol. 9, No. 1, 2008

9-26%.11 This suggests that the presence of covalently bound iodine confers little additional inhibition over that for baseline degradable PVA polymers. This confirms our previous studies on toxicity of iodine-containing polymeric biomaterials where it was shown that iodine-containing methacrylic polymers are biocompatible materials, comparable to PMMA.24 The cell inhibition ratios of 26 and 29% are most likely attributable to a combination of the degradation products plus remnants of the respective radical-generating chemicals. Biocompatibility of the new iodine-containing cross-linked PVAs and their degradation products in ViVo is currently under investigation in our laboratories.

4. Conclusions The aim of this study was to introduce intrinsic radiopacity to degradable PVA polymer while maintaining other characteristics such as water solubility, injectability, degradation kinetics, and most important, the ability of the polymer to form three-dimensional networks by redox or UV initiation. Attaching the aromatic iodine group to PVA was successful and at a 0.5% of iodine substitution, water solubility of the polymer was maintained. The iodine-containing PVA polymer displayed good X-ray visibility. Although the presence of the iodine group increased the hydrophobicity of the macromer chains, this may have been a beneficial property as the hydrophobic interactions have led to more efficient cross-linking, as proven by the lower sol fraction of the radiopaque degradable hydrogels. Even though the iodine group prolonged the degradation time of the hydrogels because of the increased hydrophobicity, radiopaque hydrogels still displayed degradation profiles similar to their counterpart nonradiopaque degradable hydrogels. Acknowledgment. We acknowledge Dr. Justin Cooper-White and Helena Hadisaputra at University of Queensland for assistance with the rheology studies, and Lynn Ferris at the University of New South Wales for her assistance with the cell studies. This study was funded in part by an Australian Research Council Linkage Project Grant (LP0220056), Australian Research Council Discovery Project (DP0557863).

References and Notes (1) Peppas, N. A.; Huang, Y.; Torres-Lugo, M.; Ward, J. H.; Zhang, J. Annu. ReV. Biomed. Eng. 2004, 2, 9–29. (2) Folkman, J.; Long, D. M. J. Surg. Res. 1964, 4, 139–142. (3) Desai, S. J.; Siminelli, A. P.; Higuchi, W. I. J. Pharm. Sci. 1965, 54, 1459–1464.

Mawad et al. (4) Peppas, N. A. In Hydrogels in Medicine and Pharmacy; RC Press, Boca Raton, FL, 1987. (5) Li, J. K.; Wang, N.; Wu, X. S. J. Controlled Release 1998, 56, 117– 126. (6) Peppas, N. A. Makromol. Chem. 1975, 176, 3433–3440. (7) Mason, N.; Thies, C.; Cicero, T. J. J. Pharm. Sci. 1976, 65, 847–850. (8) Marty, J. J.; Oppenheim, R. C. Aust. J. Pharm. 1977, 6, 65–76. (9) Martens, P.; Holland, T.; Anseth, K. S. Polymer 2002, 43, 6093–6100. (10) The CAS systematic name of anhydride 1 was determined through the program ChemDraw (Cambridge Software), and found to be 3,8,11,16-tetraoxo-4,7,12,15-tetraoxanonadec-1-en-19-oic anhydride. (11) Mawad, D.; Martens, P.; Odell, R. A.; Poole-Warren, L. A. Biomaterials 2007, 28, 947–955. (12) Kazushi, K.; Shinya, M.; Shohei, T.; Kenji, S.; Ichiro, K.; Kouji, T.; Takashi, O.; Kohji, T. Neurosurgery 1994, 34 (4), 694–698. (13) Tokunaga, K.; Kinugasa, K.; Mandai, S.; Handa, A.; Hirotsune, N.; Ohmoto, T. Neurosurgery 1998, 42 (5), 1135–1142. (14) Mottu, F.; Rufenacht, D. A.; Laurent, A.; Doelker, E. Biomaterials 2002, 23, 121–131. (15) Hopkins, L. N.; Lopes, D. K. Neurosurgery 1998, 42 (5), 1142–1143. (16) Standard, C. S.; Guterman, L. R.; Hopkins, L. N. Neurosurgery 1994, 34 (4), 699–701. (17) Stoltenberg, M.; Larsen, A.; Zhao, M.; Danscher, G.; Brunk, U. T. APMIS 2002, 110, 396–402. (18) Jayakrishnan, A.; Chithambara, B. J. Appl. Polym. Sci. 1992, 44, 743– 748. (19) Davy, K. W. M.; Anseau, M. R.; Odlyha, M.; Foster, G. M. Polym. Int. 1997, 43, 143–145. (20) Davy, K. W. M.; Anseau, M. R.; Berry, C. J. Dent. 1997, 25 (6), 143–145. (21) Kruft, M. B.; Benzina, A.; Blezer, R.; Koole, L. H. Biomaterials 1996, 17, 1803–1812. (22) van Hooy-Corstjens, C. S. J.; Govaert, L. E.; Spoelstra, A. B.; Bulstra, S. K.; Wetzels, G. M. R.; Koole, L. H. Biomaterials 2004, 25, 2657– 2667. (23) Boelen, E. J.; van Hooy-Corstjens, C. S. J.; Bulstra, S. K.; van Ooij, A.; van Rhijn, L. W.; Koole, L. H. Biomaterials 2005, 26, 6674– 6683. (24) Aldenhoff, Y. B. J.; Kruft, M. B.; Pijpers, A. P.; Veen, F. H. v. d.; Bulstra, S. K.; Kuijer, R.; Koole, L. H Biomaterials 2002, 23, 881– 886. (25) Galperin, A.; Margel, S. J Polym. Sci., Part A: Polym Chem. 2006, 44, 3859. (26) James, N. R.; Jayakrishnan, A. Biomaterials 2007, 28, 3182–3187. (27) James, N. R.; Philip, J.; Jayakrishnan, A. Biomaterials 2006, 27, 160– 166. (28) Lakshmi, S.; James, N. R.; Nisha, V. S.; Jayakrishnan, A. J. Appl. Polym. Sci. 2003, 88, 2580–2584. (29) Nottelet, B.; Coudane, J.; Vert, M. Biomaterials 2006, 27, 4948–4954. (30) Metters, A. T.; Anseth, K. S.; Bowman, C. N. Polymer 2000, 4, 3993– 4004. (31) Bhattarai, N.; Ramay, H. R.; Gunn, J.; Matsen, F. A.; Zhang, M. J. Controlled Release 2005, 103, 609–624.

BM700754M