- Email: [email protected]
temperature-sensitive injectable pentablock copolymer hydrogel
Journal of Controlled Release 137 (2009) 20–24
Contents lists available at ScienceDirect
Journal of Controlled Release j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j c o n r e l
Controlled release of insulin from pH/temperature-sensitive injectable pentablock copolymer hydrogel Dai Phu Huynh a,1, Guang Jin Im a, Su Young Chae b, Kang Choon Lee b, Doo Sung Lee a,⁎ a b
Department of Polymer Science and Engineering, Sungkyunkwan University, Suwon, Gyeonggi-do 440-746, Republic of Korea College of Pharmacy, Sungkyunkwan University, Republic of Korea
a r t i c l e
i n f o
Article history: Received 11 November 2008 Accepted 27 February 2009 Available online 12 March 2009 Keywords: Insulin release pH/temperature-sensitive hydrogel Pentablock copolymer β-amino ester STZ-induced diabetic rat
a b s t r a c t A pH- and temperature-sensitive hydrogel of poly(β-amino ester)-poly(ε-caprolactone)-poly (ethylene glycol)-poly(ε-caprolactone)-poly(β-amino ester) (PAE-PCL-PEG-PCL-PAE) pentablock copolymer was evaluated as a sustained injectable insulin delivery system. Insulin was readily loaded into the matrix, forming an ionically linked insulin–PAE complex. Complex mixtures containing various concentrations of insulin and copolymer were subcutaneously injected into male Sprague–Dawley rats to study the profile of insulin release in vivo. The insulin-release profile showed that insulin was maintained at a constant steadystate level for 15 days, and further demonstrated that insulin levels were controlled by the amount of insulin loaded into the copolymer and the copolymer concentration in the hydrogel. The effect of the insulin–gel complex was further investigated in the streptozotocin (STZ) diabetic rat model. After subcutaneously injecting complex mixtures into STZ-induced diabetic rats, blood glucose and plasma insulin levels were measured. The results showed that the diabetic rats could be treated for more than 1 week with a single injection of the complex mixture containing 10 mg/mL insulin in a 30 wt.% copolymer solution, suggesting that this pH/temperature-sensitive insulin–hydrogel complex system may have therapeutic potential. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Diabetes frequency has increased in recent years and disease prevalence is expected to continue to increase [1], making diabetes one of the most dangerous epidemics in the world [2]. The standard treatment for insulin-dependent diabetic patients remains periodical parenteral injections of insulin, a treatment regimen that is associated with poor control of blood glucose level and poor patient compliance [3]. Improved glycemic control and prevention or minimization of diabetic complications requires an insulin delivery strategy that mimics normal physiologic pattern of insulin secretion. Multiple insulin injections do provide a constant basal insulin level; however, they pose a heavy burden for diabetic patients that need insulin therapy. A controlled insulin delivery system that served as artificial pancreas was introduced for the treatment of diabetes in the 1980s, and many additional systems have been developed since then [4–10]. However, inadequate insulin release and stability issues, requiring single or multiple daily doses of insulin to control blood glucose level in diabetic patients, have remained critical impediments for sustained insulin delivery systems. Biodegradable polymeric systems, especially biodegradable hydrogels [11–17], have been attractive candidates for use in insulin ⁎ Corresponding author. Tel.: +82 31 290 7282; fax: +82 31 292 8790. E-mail address: [email protected] (D.S. Lee). 1 Present address: National Key Laboratory of Polymer and Composite Materials, HoChiMinh City University of Technology, HoChiMinh City, Vietnam. 0168-3659/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jconrel.2009.02.021
formulations because of their biodegradability and biocompatibility. Drug release from hydrogels can be regulated by controlling several hydrogel properties, including pore size, hydrophobicity and biodegradability, as well as by controlling certain functionalities that foster interaction between the matrix and the drug. Among the insulin delivery formulations considered, those based on injectable temperature-sensitive hydrogels [11,15,17] have shown several advantages for once-weekly treatment. The mixture of insulin and copolymer solution remains in a liquid state at low temperature and changes to a gel state when the temperature increases after being injected into the body. Insulin is then slowly released from the hydrogels by the combination of diffusion and hydrogel degradation. However, there are drawbacks that have limited the application of these systems for the treatment of diabetic patients. The first is clogging during injection: when a temperature-sensitive hydrogel is subcutaneously injected into the body via a syringe, the warmth of the body increases the viscosity of the solution inside the needle, causing a gel to form and making injection difficult [18,19]. The second is inefficient encapsulation of insulin into hydrogels. Although enhancing agents, such as zinc [15,17], have been used to improve insulin encapsulation, the amount of insulin ultimately loaded has been limited due to physical absorption by hydrophobic–hydrophobic interactions or isolation of the drug inside the voids in the hydrogel matrix. Recently, a novel pH/temperature-sensitive injectable hydrogel of PAE-PCL-PEG-PCL-PAE pentablock copolymers was developed [20,21]. This hydrogel exhibits a sol–gel transition that is responsive to both temperature and pH and, importantly, poses no clogging problems
D.P. Huynh et al. / Journal of Controlled Release 137 (2009) 20–24
during injection. At pH 3–4, nitrogen in the amino groups of PAE combines with H+ ions in the buffer solution to form positively charged nitrogen, increasing the solubility of hydrogel. When insulin is included with the copolymer at this low pH, most of the biomolecules can interact ionically with the PAE blocks, forming ionic linkages between the positive charges in the ionized PAE and negatively charged amino acids in insulin [20]. In addition, the hydrophilic/hydrophobic balance of PEG/PCL shifts to a more hydrophilic character at lower temperature, resulting in a hydrogel that stays in a sol state at lower pH and lower temperature. When pH and temperature are increased to 7.4 and 37 °C, respectively, the hydrogel forms a gel structure because the PAE block is de-ionized and becomes hydrophobic when the pH is higher than the pKb of PAE. The sol–gel phase transition of aqueous copolymers is easily controlled by adjusting the PEG molecular weight, the PCL/PEG ratios and the molecular weight of PAE [21]. The pentablock copolymers also provide various additional advantages, including ease of dissolution, easy sterilization by UV radiation, simple dose adjustment, and biocompatibility with no inflammatory reaction. In this study, we evaluated the profile of insulin release from mixtures of pH/temperature-sensitive PAE-PCL-PEG-PCL-PAE hydrogel and insulin in Sprague–Dawley (SD) rats. We also extended these investigations to include a consideration of the efficacy of insulin-loaded gel complexes in treating streptozotocin (STZ)-induced diabetic rats. 2. Materials and methods 2.1. Materials The materials used for polymer synthesis, which include poly (ethylene glycol) (PEG) (Mn = 1650), ε-caprolactone, acryloyl chloride, triethylamine, 1,4-butanediol diacrylate, 4,4′-trimethylene dipiperidine, Tween 80, Cremophor EL, and stannous octoate [Sn(Oct)2], were purchased from Sigma-Aldrich Co. (St. Louis, MO, USA). HCl 37%, NaCl, chloroform, dichloromethane, and diethyl ether were all obtained from Samchun Co. All other reagents were of analytical grade, and were used without further purification. PAE-PCL-PEG-PCL-PAE (block length, 1258–1584–1650–1584–1258) was synthesized as described previously [20]. Human insulin, STZ, plasma heparin, and phosphate buffered saline (PBS) were purchased from Sigma-Aldrich Co. 2.2. Insulin loading and release in vitro 2.2.1. Insulin loading Copolymer solutions were prepared by dissolving PAE-PCL-PEGPCL-PAE in PBS buffer containing 2–4 volume percent (vol.%) of 37% HCl and stored at 2 °C for 1–2 days. The pH of the copolymer solutions was adjusted to 3.0–4.0 by adding 5 M NaOH or 5 M HCl at 2 °C. Insulin was then loaded into the complex by mixing the drug with copolymer solutions at 2 °C for 6 h. 2.2.2. Insulin release from the gel complex A liquid gel–insulin mixture (0.5 mL) at pH 7.4 was placed in a 6-mL vial, and the solution was converted to a gel state by incubating at 37 °C for 30 min. Three milliliters of fresh release medium (2.4 wt.% Tween 80, 4 wt.% Cremophor EL in PBS buffer at pH 7.4) at 37 °C [19], added to the vial samples, provided a reservoir into which insulin could be released from the gel complex and subsequently measured. At a given time, a 1.5-mL sample of release medium was extracted from the sample vials and replaced with 1.5 mL fresh release medium. 2.2.3. Insulin measurement in vitro Insulin released into the release medium in vitro was determined by HPLC using a C18 column (250 × 4.0 mm, 5.0 µm particles) with an ACN:H2O mobile phase (28:72) containing 0.15% (v/v) trifluoroacetic acid and a flow rate of 0.5 mL/min, with UV detection at 214 nm.
21
2.3. Controlled insulin release in male SD rats In vivo controlled insulin release experiments were performed using 5–6-week-old, male SD rats (average body weight 200 g) obtained from Hanlim Experimental Animal Laboratory (Seoul, Korea). SD rats were maintained according to the National Institutes of Health (NIH) guidelines for the care and use of laboratory animals (NIH publications 85-23, revised 1985). Before experiments, the rats were acclimatized in the animal facility (12-h light/dark cycle) for 1 week and provided free access to food and water. The rats were divided into four groups containing five rats per group. Group 1 animals were administered a complex mixture containing 0.75 mg/mL insulin in 20 wt.% copolymer solution using a 26G hypodermic needle. Groups 2, 3, and 4 animals were administered complex mixtures containing copolymer solutions with concentrations of 20, 25 and 30 wt.%, respectively, and 5 mg/mL insulin. A total of 200 µL of the complex solution at pH 7.0 and 10 °C was injected subcutaneously into the back of SD rats. 2.4. In vivo gel formation and degradation The in vivo gel formation and degradation tests were performed using 5–6-week-old male SD rats (average body weight 200 g). The complex mixture containing 5 mg/mL insulin in 20 wt.% copolymer solution were injected subcutaneously into the back of SD rats. At predetermined time, the animals were sacrificed by cervical dislocation and then the injection site was carefully cut open and the images of the formed gel was taken with a camera. 2.5. Controlled insulin release in STZ-induced diabetic rats 2.5.1. Preparation of STZ-induced diabetic rats An STZ solution (300 mg/dL), prepared by dissolving STZ in chilled 0.05 M sodium citrate buffer at pH 4.5, was intraperitoneally injected into male SD rats at a final dose of 60 mg/kg STZ. Food was withheld for 3 h prior to STZ injection. Five days after treating rats with STZ, blood glucose levels were greater than 450 mg/dL. 2.5.2. Insulin treatment in STZ-induced diabetic rats The STZ-induced diabetic rats were divided into four groups (n = 5 per group) and treated with the complex mixtures. In the control group (group 1), only a 30 wt.% copolymer solution was administered. Groups 2, 3 and 4 were administered complex mixtures containing 1 mg/mL, 5 mg/mL and 10 mg/mL insulin in 30 wt.% copolymer solution, respectively. A total of 200 µL of the complex solution, adjusted to pH 7.0 with 5 M NaOH or 5 M HCl at 10 °C, was injected subcutaneously into the back of STZ-induced diabetic rats. 2.6. Blood glucose and plasma insulin level measurements Blood was taken from the tail vein, and blood glucose levels were measured using a one-touch blood glucose monitoring system (Accucheck sensor, Roche Diagnostics Inc., Switzerland). Plasma for insulin measurements was obtained by collecting 300-μL blood samples from the rat tail vein into a polypropylene tube containing 3 µL plasma heparin, and then centrifuged at 13,000 rpm for 5 min. Insulin concentration in plasma samples was analyzed using a Mercodia Insulin ELISA kit (Mercodia AB, Sweden). All samples were stored at −10 °C before analysis. 3. Results and discussion 3.1. Controlled insulin release in vitro To evaluate the ability of the pH/temperature-sensitive pentablock hydrogel of PAE-PCL-PEG-PCL-PAE to effectively deliver insulin, we performed in vitro release studies, measuring insulin released into the
22
D.P. Huynh et al. / Journal of Controlled Release 137 (2009) 20–24
surrounding medium (pH 7.4) at 37 °C. Fig. 1 shows the release profile of insulin from the pH/temperature-sensitive hydrogel. There were no significant differences in the cumulative release profiles between complex mixtures containing 5 mg/mL insulin in 20 or 30 wt.% copolymer solutions (Fig. 1A), and between the complex mixtures containing 5 or 10 mg/mL insulin in 20 wt.% copolymer solution (Fig. 1B). Insulin was continuously released from the hydrogel with constant release kinetics over 20 days, indicating that the complex gel played an important role in the long-term sustained release of insulin. This was likely due to ionic complexation between insulin and the PAE segments of the polymer, which stabilized the insulin physically and chemically [20].
3.2. Controlled insulin release in SD rats Fig. 2 shows the steady-state release of insulin from the insulin–gel complexes after a single subcutaneous injection into an SD rat. Increasing the insulin concentration loaded into a 20 wt.% copolymer solution from 0.75 mg/mL to 5 mg/mL increased steady-state insulin levels from 282 mU/L to 546 mU/L. At a fixed insulin concentration of 5 mg/mL, increasing the copolymer concentration from 20 wt.% to 25 and 30 wt.% also increased the steady-state insulin levels from 546 mU/L to 572 and 710 mU/L, respectively, and maintained insulin levels for 15 days. These results show that there was a good correlation between in vitro and in vivo insulin-release profiles, and demonstrate that the steady-state plasma insulin levels were directly proportional to the insulin concentration loaded and the copolymer concentration used. An initial burst of insulin release was detected after a single injection of the complex mixtures containing 5 mg/mL of insulin in 20
Fig. 1. In vitro release of insulin from pentablock copolymer hydrogels containing 5 mg/mL insulin in 20 or 30 wt.% copolymer solutions (A) and those containing 5 or 10 mg/mL insulin in 20 wt.% copolymer solution (B). The results are presented as mean± standard deviation (n = 4).
Fig. 2. Controlled insulin release in plasma in SD rats. The results are presented as mean± standard deviation (n = 5).
Fig. 3. Gel formation and degradation of the insulin–gel complexes in SD rats.
D.P. Huynh et al. / Journal of Controlled Release 137 (2009) 20–24
and 25 wt.% copolymer solutions (Fig. 2). At a copolymer concentration of 20 wt.%, the peak insulin level for the initial burst reached 1120 mU/L and caused the rats to undergo hypoglycemic shock. At this copolymer concentration, there might be insufficient copolymers to form electrostatic linkages with all the insulin loaded into the copolymer solution [20]. Thus, some insulin encapsulated into the complex mixture without ionic linkage could be rapidly released by diffusion at the early phase of the insulin-release period, resulting in a brief spike in blood insulin levels that caused a rapid decrease in blood glucose levels and induction of hypoglycemic shock. However, increasing the copolymer concentration to 25 wt.% decreased the peak insulin level during the initial burst to 967 mU/L without changing steady-state insulin concentrations; under these conditions, no animals underwent hypoglycemic shock. Increasing the copolymer concentration to 30 wt.% completely eliminated the initial burst of insulin release; thus, this formulation was selected for the subsequent experiment. 3.3. In vivo gel formation and degradation We performed in vivo gel formation and degradation studies with a complex mixture containing 5 mg/mL insulin in 20 wt.% copolymer solution. After a single subcutaneous injection of the insulin-loaded complex mixture into the back of SD rats, the gel that formed under the skin remained for more than 14 days (Fig. 3), which correlates with the results of in vivo insulin-release experiments. 3.4. Controlled insulin release in STZ-induced diabetic rats We extended these studies to investigate the possibility of using insulin-loaded complex mixtures to treat STZ-induced diabetic rats. Intraperitoneal injection of STZ into SD rats induced hyperglycemia and hypoinsulinemia due to damage caused to insulin-secretory cells in the pancreatic islets (Figs. 4 and 5). As shown in Fig. 4, the blood glucose levels in the control group rapidly increased from about 100 mg/dL to 470 mg/dL within 5 days after STZ injection, and then reached and maintained 590 ± 22 mg/dL for more than 15 days. Although endogenous insulin was still produced in STZ-induced diabetic rats, the levels were too low to control blood glucose. After injecting insulin-loaded complex mixtures into diabetic rats, the exogenously released insulin started to decrease the blood glucose levels, which dropped rapidly (i.e., within one day) after a single injection of the insulin-loaded complex mixtures and were main-
Fig. 4. Blood glucose levels in STZ-induced diabetic rats after treatment with different concentrations of insulin in a 30 wt.% copolymer solution. The insulin concentrations were 0 mg/mL in the control group (pentablock solution only), 1 mg/mL in group 1, 5 mg/mL in group 2, and 10 mg/mL in group 3. The results are presented as mean ±standard deviation (n = 5).
23
Fig. 5. Plasma insulin levels in STZ-induced diabetic rats after treatment with different concentrations of insulin in a 30 wt.% copolymer solution. The insulin concentrations were 0 mg/mL in the control group (pentablock solution only), 1 mg/mL in group 1, 5 mg/mL in group 2, and 10 mg/mL in group 3. The results are presented as mean ± standard deviation (n = 5).
tained at steady-state levels throughout the insulin-release period. For groups 1, 2 and 3, the steady-state blood glucose levels were 189 ± 22, 119 ± 22 and 104 ± 29 mg/dL, and the steady-state periods were 5, 7 and 10 days, respectively. Furthermore, steady-state blood glucose levels in groups 2 and 3, where the insulin concentrations were 5 and 10 mg/mL, respectively (both in 30 wt.% copolymer solution), dropped to normal levels. This was especially striking in group 3, where the steady-state blood glucose levels were maintained at normal levels for 1 week and then at near-normal levels for an additional 3 days. The plasma insulin levels correlated well with the blood glucose levels after a single injection of the insulin–gel complexes (Fig. 5). In group 3, insulin release from the complex mixture was constant until day 10. Four days after receiving an injection of pentablock solution only, the body weight of STZ-induced diabetic rats in the control group, which had reached the highest level of blood glucose (590 ± 22 mg/dL), started to decrease (Fig. 6). In contrast, body weight increased continuously up to 10 days after a single injection of complex mixture containing 10 mg/mL insulin in 30 wt.% copolymer solution (group 3), before gradually declining. This decline in body weight began earlier in groups 1 and 2 treated with 1 and 5 mg/mL insulin, respectively, than in group 3. These results suggest that the change in body weight in diabetic rats correlates with the amount of insulin loaded into the copolymer.
Fig. 6. Body weight changes in STZ-induced diabetic rats after treatment with different concentrations of insulin in a 30 wt.% copolymer solution. The insulin concentrations were 0 mg/mL in the control group (pentablock solution only), 1 mg/ml in group 1, 5 mg/mL in group 2, and 10 mg/mL in group 3. The results are presented as mean ± standard deviation (n = 5).
24
D.P. Huynh et al. / Journal of Controlled Release 137 (2009) 20–24
Sustained-release formulations of insulin prepared with a biodegradable matrix have been recently developed. Takenaga et al. [22] reported that insulin-loaded microcapsules made of poly(D,L-lactic-coglycolic acid) (PLGA) increased plasma insulin levels in STZ-induced diabetic rats, but the glucose levels of diabetic rats were not reported. In addition, drug delivery systems using such polymers involve the use of organic solvents or heat, which may denature proteins and lead to a loss of bioactivity. Choi and Kim [23] used a water soluble, temperaturesensitive biodegradable triblock copolymer of PLGA-PEG-PLGA for controlled release of insulin, and reported that steady-state blood glucose levels were maintained in a euglycemic range in diabetic rats for 1 week. The triblock copolymer, however, has several drawbacks, including clogging during injection and inefficient encapsulation of insulin into hydrogels, as noted above. Here, we used the pH/ temperature-sensitive, injectable PAE-PCL-PEG-PCL-PAE pentablock copolymer hydrogel, which, in addition to circumventing clogging problems during injection [20], maintains a constant level of blood glucose within a normal range for more than 1 week in diabetic rats. And it dose so after only a single subcutaneous injection of a complex mixture containing 10 mg/mL insulin in 30 wt.% of pentablock copolymer solution, and without a loss of body weight.
4. Conclusions The controlled insulin-release experiments in SD rats performed using complex formed from mixtures of pH/temperature-sensitive hydrogel of PAE-PCL-PEG-PCL-PAE pentablock copolymer and insulin demonstrated that the major factors that influence the insulin-release profile are the amount of insulin loaded into the copolymer and the copolymer concentration in the hydrogel. After a single subcutaneous injection into STZ-induced diabetic rats, a complex mixture containing 10 mg/mL insulin in 30 wt.% copolymer solution achieved a constant steady-state blood glucose level that was maintained in the normal range for more than 1 week without a decrease in body weight. Thus, our pH/temperature-sensitive insulin-hydrogel complex system offers advantages over other insulin delivery systems, and has the potential to improve therapeutic efficacy and diabetic patient compliance.
Acknowledgment This work was supported by Korea Research Foundation Grant KRF-2006-005-J04602.
References [1] American Diabetes Association, National Diabetes Fact Sheet, , October 5 2004 www.diabetes.org. [2] H. King, R. Aubert, W. Herman, Global burden of diabetes, 1995–2025: prevalence, numerical estimates, and projections, Diabetes Care 21 (9) (1998) 1414–1431. [3] K. Zhang, X.Y. Wu, Modulated insulin permeation across a glucose-sensitive polymeric composite membrane, J. Control. Release 80 (2002) 169–178. [4] P. Tyagi, Insulin delivery systems: present trends and the future direction, Indian J. Pharmacol. 34 (2002) 379–389. [5] K. Ishihara, M. Kobayashi, I. Shinohara, Control of insulin permeation through a polymer membrane with responsive function for glucose, Makromol. Chem., Rapid Commun. 4 (1983) 327–331. [6] C.M. Hassan, F.J. Doyle III, N.A. Peppas, Dynamic behavior of glucose-responsive poly(methacrylic acid-g-ethylene glycol), Macromolecular 30 (1997) 6166–6173. [7] T. Traitel, Y. Cohen, J. Kost, Characterization of glucose-sensitive insulin release systems in stimulated in vivo conditions, Biomaterials 21 (2000) 1679–1687. [8] K. Zhang, X.Y. Wu, L.Y. Chu, Y. Li, J.H. Zhu, H.D. Wang, Y.J. Liang, Control of pore size and permeability of a glucose-responsive gating membrane for insulin delivery, J. Control. Release 97 (2004) 43–53. [9] T. Uchiyama, Y. Kiritoshi, J. Watanabe, K. Ishihara, Degradation of phospholipid polymer hydrogel by hydrogen peroxide aiming at insulin release device, Biomaterials 24 (2003) 5183–5190. [10] J. Varshosaz, Insulin Delivery Systems for Controlling Diabetes, Recent patents on Endocrine, Metabolic & Immune Drug Discovery, vol. 1, 2007, pp. 25–40. [11] P. Calicetia, S. Salmaso, A. Lante, M. Yoshida, R. Katakai, F. Martellini, L.H.I. Mei, M. Carenza, Controlled release of biomolecules from temperature-sensitive hydrogels prepared by radiation polymerization, J. Control. Release 75 (2001) 173–181. [12] S.K. Bajpai, S.S. Saggu, Insulin release behavior of poly(methacrylamide-co-N-vinyl-2pyrrolidone-co-itaconic acid) hydrogel: an interesting probe, part II, J. Macromol. Sci. 44 (2007) 153–157. [13] J.J. Kim, K. Park, Molecular insulin delivery from glucose-sensitive hydrogel dosage forms, J. Control. Release 77 (2001) 39–47. [14] K. Moriyama, N. Yui, Regulated insulin release from biodegradable dextran hydrogels containing poly(ethylene glycol), J. Control. Release 42 (1996) 237–248. [15] Y.J. Kim, S. Choi, J.J. Koh, M. Lee, K.S. Ko, S.W. Kim, Controlled release of insulin from injectable biodegradable triblock copolymer, Pharm. Res. 18 (2001) 548–550. [16] M.T. Ende, N.A. Pappas, Transport of ionizable drugs and proteins in crosslinked poly(acrylic acid) and poly(acrylic acid-co-2-hydroxyethyl methacrylate) hydrogels. II. diffusion and release studies, J. Control. Release 48 (1997) 47–56. [17] S. Choi, M. Baudys, S.W. Kim, Control of blood glucose by novel GLP-1 delivery using biodegradable triblock copolymer of PLGA-PEG-PLGA in type 2 diabetic rats, Pharm. Res. 21 (2004) 827–831. [18] W.S. Shim, J.S. Yoo, Y.H. Bae, D.S. Lee, Novel injectable pH and temperature sensitive block copolymer hydrogel, Biomacromolecules 6 (2005) 2930–2934. [19] D.P. Huynh, W.S. Shim, J.H. Kim, D.S. Lee, pH/temperature sensitive poly(ethylene glycol)-based biodegradable polyester block copolymer hydrogels, Polymer 47 (2006) 7918–7926. [20] D.P. Huynh, M.K. Nguyen, B.S. Pi, M.S. Kim, S.Y. Chae, K.C. Lee, B.S. Kim, S.W. Kim, D. S. Lee, Functionalized injectable hydrogel for controlled insulin delivery, Biomaterials 29 (2008) 2527–2534. [21] D.P. Huynh, M.K. Nguyen, M.S. Kim, B.S. Kim, D.S. Lee, Molecular design of novel pH/temperature sensitive injectable hydrogels, Polymer (in press). [22] M. Takenaga, Y. Yamaguchi, A. Kitagawa, Y. Ogawa, Y. Izushima, R. Igarashi, A novel sustained-release formulation of insulin with dramatic reduction in initial rapid release, J. Control. Release 79 (2002) 80–91. [23] S. Choi, S.W. Kim, Controlled release of insulin from injectable biodegradable triblock copolymer depot in ZDF rats, Pharm. Res. 20 (2003) 2008–2010.
Contents lists available at ScienceDirect
Journal of Controlled Release j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j c o n r e l
Controlled release of insulin from pH/temperature-sensitive injectable pentablock copolymer hydrogel Dai Phu Huynh a,1, Guang Jin Im a, Su Young Chae b, Kang Choon Lee b, Doo Sung Lee a,⁎ a b
Department of Polymer Science and Engineering, Sungkyunkwan University, Suwon, Gyeonggi-do 440-746, Republic of Korea College of Pharmacy, Sungkyunkwan University, Republic of Korea
a r t i c l e
i n f o
Article history: Received 11 November 2008 Accepted 27 February 2009 Available online 12 March 2009 Keywords: Insulin release pH/temperature-sensitive hydrogel Pentablock copolymer β-amino ester STZ-induced diabetic rat
a b s t r a c t A pH- and temperature-sensitive hydrogel of poly(β-amino ester)-poly(ε-caprolactone)-poly (ethylene glycol)-poly(ε-caprolactone)-poly(β-amino ester) (PAE-PCL-PEG-PCL-PAE) pentablock copolymer was evaluated as a sustained injectable insulin delivery system. Insulin was readily loaded into the matrix, forming an ionically linked insulin–PAE complex. Complex mixtures containing various concentrations of insulin and copolymer were subcutaneously injected into male Sprague–Dawley rats to study the profile of insulin release in vivo. The insulin-release profile showed that insulin was maintained at a constant steadystate level for 15 days, and further demonstrated that insulin levels were controlled by the amount of insulin loaded into the copolymer and the copolymer concentration in the hydrogel. The effect of the insulin–gel complex was further investigated in the streptozotocin (STZ) diabetic rat model. After subcutaneously injecting complex mixtures into STZ-induced diabetic rats, blood glucose and plasma insulin levels were measured. The results showed that the diabetic rats could be treated for more than 1 week with a single injection of the complex mixture containing 10 mg/mL insulin in a 30 wt.% copolymer solution, suggesting that this pH/temperature-sensitive insulin–hydrogel complex system may have therapeutic potential. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Diabetes frequency has increased in recent years and disease prevalence is expected to continue to increase [1], making diabetes one of the most dangerous epidemics in the world [2]. The standard treatment for insulin-dependent diabetic patients remains periodical parenteral injections of insulin, a treatment regimen that is associated with poor control of blood glucose level and poor patient compliance [3]. Improved glycemic control and prevention or minimization of diabetic complications requires an insulin delivery strategy that mimics normal physiologic pattern of insulin secretion. Multiple insulin injections do provide a constant basal insulin level; however, they pose a heavy burden for diabetic patients that need insulin therapy. A controlled insulin delivery system that served as artificial pancreas was introduced for the treatment of diabetes in the 1980s, and many additional systems have been developed since then [4–10]. However, inadequate insulin release and stability issues, requiring single or multiple daily doses of insulin to control blood glucose level in diabetic patients, have remained critical impediments for sustained insulin delivery systems. Biodegradable polymeric systems, especially biodegradable hydrogels [11–17], have been attractive candidates for use in insulin ⁎ Corresponding author. Tel.: +82 31 290 7282; fax: +82 31 292 8790. E-mail address: [email protected] (D.S. Lee). 1 Present address: National Key Laboratory of Polymer and Composite Materials, HoChiMinh City University of Technology, HoChiMinh City, Vietnam. 0168-3659/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jconrel.2009.02.021
formulations because of their biodegradability and biocompatibility. Drug release from hydrogels can be regulated by controlling several hydrogel properties, including pore size, hydrophobicity and biodegradability, as well as by controlling certain functionalities that foster interaction between the matrix and the drug. Among the insulin delivery formulations considered, those based on injectable temperature-sensitive hydrogels [11,15,17] have shown several advantages for once-weekly treatment. The mixture of insulin and copolymer solution remains in a liquid state at low temperature and changes to a gel state when the temperature increases after being injected into the body. Insulin is then slowly released from the hydrogels by the combination of diffusion and hydrogel degradation. However, there are drawbacks that have limited the application of these systems for the treatment of diabetic patients. The first is clogging during injection: when a temperature-sensitive hydrogel is subcutaneously injected into the body via a syringe, the warmth of the body increases the viscosity of the solution inside the needle, causing a gel to form and making injection difficult [18,19]. The second is inefficient encapsulation of insulin into hydrogels. Although enhancing agents, such as zinc [15,17], have been used to improve insulin encapsulation, the amount of insulin ultimately loaded has been limited due to physical absorption by hydrophobic–hydrophobic interactions or isolation of the drug inside the voids in the hydrogel matrix. Recently, a novel pH/temperature-sensitive injectable hydrogel of PAE-PCL-PEG-PCL-PAE pentablock copolymers was developed [20,21]. This hydrogel exhibits a sol–gel transition that is responsive to both temperature and pH and, importantly, poses no clogging problems
D.P. Huynh et al. / Journal of Controlled Release 137 (2009) 20–24
during injection. At pH 3–4, nitrogen in the amino groups of PAE combines with H+ ions in the buffer solution to form positively charged nitrogen, increasing the solubility of hydrogel. When insulin is included with the copolymer at this low pH, most of the biomolecules can interact ionically with the PAE blocks, forming ionic linkages between the positive charges in the ionized PAE and negatively charged amino acids in insulin [20]. In addition, the hydrophilic/hydrophobic balance of PEG/PCL shifts to a more hydrophilic character at lower temperature, resulting in a hydrogel that stays in a sol state at lower pH and lower temperature. When pH and temperature are increased to 7.4 and 37 °C, respectively, the hydrogel forms a gel structure because the PAE block is de-ionized and becomes hydrophobic when the pH is higher than the pKb of PAE. The sol–gel phase transition of aqueous copolymers is easily controlled by adjusting the PEG molecular weight, the PCL/PEG ratios and the molecular weight of PAE [21]. The pentablock copolymers also provide various additional advantages, including ease of dissolution, easy sterilization by UV radiation, simple dose adjustment, and biocompatibility with no inflammatory reaction. In this study, we evaluated the profile of insulin release from mixtures of pH/temperature-sensitive PAE-PCL-PEG-PCL-PAE hydrogel and insulin in Sprague–Dawley (SD) rats. We also extended these investigations to include a consideration of the efficacy of insulin-loaded gel complexes in treating streptozotocin (STZ)-induced diabetic rats. 2. Materials and methods 2.1. Materials The materials used for polymer synthesis, which include poly (ethylene glycol) (PEG) (Mn = 1650), ε-caprolactone, acryloyl chloride, triethylamine, 1,4-butanediol diacrylate, 4,4′-trimethylene dipiperidine, Tween 80, Cremophor EL, and stannous octoate [Sn(Oct)2], were purchased from Sigma-Aldrich Co. (St. Louis, MO, USA). HCl 37%, NaCl, chloroform, dichloromethane, and diethyl ether were all obtained from Samchun Co. All other reagents were of analytical grade, and were used without further purification. PAE-PCL-PEG-PCL-PAE (block length, 1258–1584–1650–1584–1258) was synthesized as described previously [20]. Human insulin, STZ, plasma heparin, and phosphate buffered saline (PBS) were purchased from Sigma-Aldrich Co. 2.2. Insulin loading and release in vitro 2.2.1. Insulin loading Copolymer solutions were prepared by dissolving PAE-PCL-PEGPCL-PAE in PBS buffer containing 2–4 volume percent (vol.%) of 37% HCl and stored at 2 °C for 1–2 days. The pH of the copolymer solutions was adjusted to 3.0–4.0 by adding 5 M NaOH or 5 M HCl at 2 °C. Insulin was then loaded into the complex by mixing the drug with copolymer solutions at 2 °C for 6 h. 2.2.2. Insulin release from the gel complex A liquid gel–insulin mixture (0.5 mL) at pH 7.4 was placed in a 6-mL vial, and the solution was converted to a gel state by incubating at 37 °C for 30 min. Three milliliters of fresh release medium (2.4 wt.% Tween 80, 4 wt.% Cremophor EL in PBS buffer at pH 7.4) at 37 °C [19], added to the vial samples, provided a reservoir into which insulin could be released from the gel complex and subsequently measured. At a given time, a 1.5-mL sample of release medium was extracted from the sample vials and replaced with 1.5 mL fresh release medium. 2.2.3. Insulin measurement in vitro Insulin released into the release medium in vitro was determined by HPLC using a C18 column (250 × 4.0 mm, 5.0 µm particles) with an ACN:H2O mobile phase (28:72) containing 0.15% (v/v) trifluoroacetic acid and a flow rate of 0.5 mL/min, with UV detection at 214 nm.
21
2.3. Controlled insulin release in male SD rats In vivo controlled insulin release experiments were performed using 5–6-week-old, male SD rats (average body weight 200 g) obtained from Hanlim Experimental Animal Laboratory (Seoul, Korea). SD rats were maintained according to the National Institutes of Health (NIH) guidelines for the care and use of laboratory animals (NIH publications 85-23, revised 1985). Before experiments, the rats were acclimatized in the animal facility (12-h light/dark cycle) for 1 week and provided free access to food and water. The rats were divided into four groups containing five rats per group. Group 1 animals were administered a complex mixture containing 0.75 mg/mL insulin in 20 wt.% copolymer solution using a 26G hypodermic needle. Groups 2, 3, and 4 animals were administered complex mixtures containing copolymer solutions with concentrations of 20, 25 and 30 wt.%, respectively, and 5 mg/mL insulin. A total of 200 µL of the complex solution at pH 7.0 and 10 °C was injected subcutaneously into the back of SD rats. 2.4. In vivo gel formation and degradation The in vivo gel formation and degradation tests were performed using 5–6-week-old male SD rats (average body weight 200 g). The complex mixture containing 5 mg/mL insulin in 20 wt.% copolymer solution were injected subcutaneously into the back of SD rats. At predetermined time, the animals were sacrificed by cervical dislocation and then the injection site was carefully cut open and the images of the formed gel was taken with a camera. 2.5. Controlled insulin release in STZ-induced diabetic rats 2.5.1. Preparation of STZ-induced diabetic rats An STZ solution (300 mg/dL), prepared by dissolving STZ in chilled 0.05 M sodium citrate buffer at pH 4.5, was intraperitoneally injected into male SD rats at a final dose of 60 mg/kg STZ. Food was withheld for 3 h prior to STZ injection. Five days after treating rats with STZ, blood glucose levels were greater than 450 mg/dL. 2.5.2. Insulin treatment in STZ-induced diabetic rats The STZ-induced diabetic rats were divided into four groups (n = 5 per group) and treated with the complex mixtures. In the control group (group 1), only a 30 wt.% copolymer solution was administered. Groups 2, 3 and 4 were administered complex mixtures containing 1 mg/mL, 5 mg/mL and 10 mg/mL insulin in 30 wt.% copolymer solution, respectively. A total of 200 µL of the complex solution, adjusted to pH 7.0 with 5 M NaOH or 5 M HCl at 10 °C, was injected subcutaneously into the back of STZ-induced diabetic rats. 2.6. Blood glucose and plasma insulin level measurements Blood was taken from the tail vein, and blood glucose levels were measured using a one-touch blood glucose monitoring system (Accucheck sensor, Roche Diagnostics Inc., Switzerland). Plasma for insulin measurements was obtained by collecting 300-μL blood samples from the rat tail vein into a polypropylene tube containing 3 µL plasma heparin, and then centrifuged at 13,000 rpm for 5 min. Insulin concentration in plasma samples was analyzed using a Mercodia Insulin ELISA kit (Mercodia AB, Sweden). All samples were stored at −10 °C before analysis. 3. Results and discussion 3.1. Controlled insulin release in vitro To evaluate the ability of the pH/temperature-sensitive pentablock hydrogel of PAE-PCL-PEG-PCL-PAE to effectively deliver insulin, we performed in vitro release studies, measuring insulin released into the
22
D.P. Huynh et al. / Journal of Controlled Release 137 (2009) 20–24
surrounding medium (pH 7.4) at 37 °C. Fig. 1 shows the release profile of insulin from the pH/temperature-sensitive hydrogel. There were no significant differences in the cumulative release profiles between complex mixtures containing 5 mg/mL insulin in 20 or 30 wt.% copolymer solutions (Fig. 1A), and between the complex mixtures containing 5 or 10 mg/mL insulin in 20 wt.% copolymer solution (Fig. 1B). Insulin was continuously released from the hydrogel with constant release kinetics over 20 days, indicating that the complex gel played an important role in the long-term sustained release of insulin. This was likely due to ionic complexation between insulin and the PAE segments of the polymer, which stabilized the insulin physically and chemically [20].
3.2. Controlled insulin release in SD rats Fig. 2 shows the steady-state release of insulin from the insulin–gel complexes after a single subcutaneous injection into an SD rat. Increasing the insulin concentration loaded into a 20 wt.% copolymer solution from 0.75 mg/mL to 5 mg/mL increased steady-state insulin levels from 282 mU/L to 546 mU/L. At a fixed insulin concentration of 5 mg/mL, increasing the copolymer concentration from 20 wt.% to 25 and 30 wt.% also increased the steady-state insulin levels from 546 mU/L to 572 and 710 mU/L, respectively, and maintained insulin levels for 15 days. These results show that there was a good correlation between in vitro and in vivo insulin-release profiles, and demonstrate that the steady-state plasma insulin levels were directly proportional to the insulin concentration loaded and the copolymer concentration used. An initial burst of insulin release was detected after a single injection of the complex mixtures containing 5 mg/mL of insulin in 20
Fig. 1. In vitro release of insulin from pentablock copolymer hydrogels containing 5 mg/mL insulin in 20 or 30 wt.% copolymer solutions (A) and those containing 5 or 10 mg/mL insulin in 20 wt.% copolymer solution (B). The results are presented as mean± standard deviation (n = 4).
Fig. 2. Controlled insulin release in plasma in SD rats. The results are presented as mean± standard deviation (n = 5).
Fig. 3. Gel formation and degradation of the insulin–gel complexes in SD rats.
D.P. Huynh et al. / Journal of Controlled Release 137 (2009) 20–24
and 25 wt.% copolymer solutions (Fig. 2). At a copolymer concentration of 20 wt.%, the peak insulin level for the initial burst reached 1120 mU/L and caused the rats to undergo hypoglycemic shock. At this copolymer concentration, there might be insufficient copolymers to form electrostatic linkages with all the insulin loaded into the copolymer solution [20]. Thus, some insulin encapsulated into the complex mixture without ionic linkage could be rapidly released by diffusion at the early phase of the insulin-release period, resulting in a brief spike in blood insulin levels that caused a rapid decrease in blood glucose levels and induction of hypoglycemic shock. However, increasing the copolymer concentration to 25 wt.% decreased the peak insulin level during the initial burst to 967 mU/L without changing steady-state insulin concentrations; under these conditions, no animals underwent hypoglycemic shock. Increasing the copolymer concentration to 30 wt.% completely eliminated the initial burst of insulin release; thus, this formulation was selected for the subsequent experiment. 3.3. In vivo gel formation and degradation We performed in vivo gel formation and degradation studies with a complex mixture containing 5 mg/mL insulin in 20 wt.% copolymer solution. After a single subcutaneous injection of the insulin-loaded complex mixture into the back of SD rats, the gel that formed under the skin remained for more than 14 days (Fig. 3), which correlates with the results of in vivo insulin-release experiments. 3.4. Controlled insulin release in STZ-induced diabetic rats We extended these studies to investigate the possibility of using insulin-loaded complex mixtures to treat STZ-induced diabetic rats. Intraperitoneal injection of STZ into SD rats induced hyperglycemia and hypoinsulinemia due to damage caused to insulin-secretory cells in the pancreatic islets (Figs. 4 and 5). As shown in Fig. 4, the blood glucose levels in the control group rapidly increased from about 100 mg/dL to 470 mg/dL within 5 days after STZ injection, and then reached and maintained 590 ± 22 mg/dL for more than 15 days. Although endogenous insulin was still produced in STZ-induced diabetic rats, the levels were too low to control blood glucose. After injecting insulin-loaded complex mixtures into diabetic rats, the exogenously released insulin started to decrease the blood glucose levels, which dropped rapidly (i.e., within one day) after a single injection of the insulin-loaded complex mixtures and were main-
Fig. 4. Blood glucose levels in STZ-induced diabetic rats after treatment with different concentrations of insulin in a 30 wt.% copolymer solution. The insulin concentrations were 0 mg/mL in the control group (pentablock solution only), 1 mg/mL in group 1, 5 mg/mL in group 2, and 10 mg/mL in group 3. The results are presented as mean ±standard deviation (n = 5).
23
Fig. 5. Plasma insulin levels in STZ-induced diabetic rats after treatment with different concentrations of insulin in a 30 wt.% copolymer solution. The insulin concentrations were 0 mg/mL in the control group (pentablock solution only), 1 mg/mL in group 1, 5 mg/mL in group 2, and 10 mg/mL in group 3. The results are presented as mean ± standard deviation (n = 5).
tained at steady-state levels throughout the insulin-release period. For groups 1, 2 and 3, the steady-state blood glucose levels were 189 ± 22, 119 ± 22 and 104 ± 29 mg/dL, and the steady-state periods were 5, 7 and 10 days, respectively. Furthermore, steady-state blood glucose levels in groups 2 and 3, where the insulin concentrations were 5 and 10 mg/mL, respectively (both in 30 wt.% copolymer solution), dropped to normal levels. This was especially striking in group 3, where the steady-state blood glucose levels were maintained at normal levels for 1 week and then at near-normal levels for an additional 3 days. The plasma insulin levels correlated well with the blood glucose levels after a single injection of the insulin–gel complexes (Fig. 5). In group 3, insulin release from the complex mixture was constant until day 10. Four days after receiving an injection of pentablock solution only, the body weight of STZ-induced diabetic rats in the control group, which had reached the highest level of blood glucose (590 ± 22 mg/dL), started to decrease (Fig. 6). In contrast, body weight increased continuously up to 10 days after a single injection of complex mixture containing 10 mg/mL insulin in 30 wt.% copolymer solution (group 3), before gradually declining. This decline in body weight began earlier in groups 1 and 2 treated with 1 and 5 mg/mL insulin, respectively, than in group 3. These results suggest that the change in body weight in diabetic rats correlates with the amount of insulin loaded into the copolymer.
Fig. 6. Body weight changes in STZ-induced diabetic rats after treatment with different concentrations of insulin in a 30 wt.% copolymer solution. The insulin concentrations were 0 mg/mL in the control group (pentablock solution only), 1 mg/ml in group 1, 5 mg/mL in group 2, and 10 mg/mL in group 3. The results are presented as mean ± standard deviation (n = 5).
24
D.P. Huynh et al. / Journal of Controlled Release 137 (2009) 20–24
Sustained-release formulations of insulin prepared with a biodegradable matrix have been recently developed. Takenaga et al. [22] reported that insulin-loaded microcapsules made of poly(D,L-lactic-coglycolic acid) (PLGA) increased plasma insulin levels in STZ-induced diabetic rats, but the glucose levels of diabetic rats were not reported. In addition, drug delivery systems using such polymers involve the use of organic solvents or heat, which may denature proteins and lead to a loss of bioactivity. Choi and Kim [23] used a water soluble, temperaturesensitive biodegradable triblock copolymer of PLGA-PEG-PLGA for controlled release of insulin, and reported that steady-state blood glucose levels were maintained in a euglycemic range in diabetic rats for 1 week. The triblock copolymer, however, has several drawbacks, including clogging during injection and inefficient encapsulation of insulin into hydrogels, as noted above. Here, we used the pH/ temperature-sensitive, injectable PAE-PCL-PEG-PCL-PAE pentablock copolymer hydrogel, which, in addition to circumventing clogging problems during injection [20], maintains a constant level of blood glucose within a normal range for more than 1 week in diabetic rats. And it dose so after only a single subcutaneous injection of a complex mixture containing 10 mg/mL insulin in 30 wt.% of pentablock copolymer solution, and without a loss of body weight.
4. Conclusions The controlled insulin-release experiments in SD rats performed using complex formed from mixtures of pH/temperature-sensitive hydrogel of PAE-PCL-PEG-PCL-PAE pentablock copolymer and insulin demonstrated that the major factors that influence the insulin-release profile are the amount of insulin loaded into the copolymer and the copolymer concentration in the hydrogel. After a single subcutaneous injection into STZ-induced diabetic rats, a complex mixture containing 10 mg/mL insulin in 30 wt.% copolymer solution achieved a constant steady-state blood glucose level that was maintained in the normal range for more than 1 week without a decrease in body weight. Thus, our pH/temperature-sensitive insulin-hydrogel complex system offers advantages over other insulin delivery systems, and has the potential to improve therapeutic efficacy and diabetic patient compliance.
Acknowledgment This work was supported by Korea Research Foundation Grant KRF-2006-005-J04602.
References [1] American Diabetes Association, National Diabetes Fact Sheet, , October 5 2004 www.diabetes.org. [2] H. King, R. Aubert, W. Herman, Global burden of diabetes, 1995–2025: prevalence, numerical estimates, and projections, Diabetes Care 21 (9) (1998) 1414–1431. [3] K. Zhang, X.Y. Wu, Modulated insulin permeation across a glucose-sensitive polymeric composite membrane, J. Control. Release 80 (2002) 169–178. [4] P. Tyagi, Insulin delivery systems: present trends and the future direction, Indian J. Pharmacol. 34 (2002) 379–389. [5] K. Ishihara, M. Kobayashi, I. Shinohara, Control of insulin permeation through a polymer membrane with responsive function for glucose, Makromol. Chem., Rapid Commun. 4 (1983) 327–331. [6] C.M. Hassan, F.J. Doyle III, N.A. Peppas, Dynamic behavior of glucose-responsive poly(methacrylic acid-g-ethylene glycol), Macromolecular 30 (1997) 6166–6173. [7] T. Traitel, Y. Cohen, J. Kost, Characterization of glucose-sensitive insulin release systems in stimulated in vivo conditions, Biomaterials 21 (2000) 1679–1687. [8] K. Zhang, X.Y. Wu, L.Y. Chu, Y. Li, J.H. Zhu, H.D. Wang, Y.J. Liang, Control of pore size and permeability of a glucose-responsive gating membrane for insulin delivery, J. Control. Release 97 (2004) 43–53. [9] T. Uchiyama, Y. Kiritoshi, J. Watanabe, K. Ishihara, Degradation of phospholipid polymer hydrogel by hydrogen peroxide aiming at insulin release device, Biomaterials 24 (2003) 5183–5190. [10] J. Varshosaz, Insulin Delivery Systems for Controlling Diabetes, Recent patents on Endocrine, Metabolic & Immune Drug Discovery, vol. 1, 2007, pp. 25–40. [11] P. Calicetia, S. Salmaso, A. Lante, M. Yoshida, R. Katakai, F. Martellini, L.H.I. Mei, M. Carenza, Controlled release of biomolecules from temperature-sensitive hydrogels prepared by radiation polymerization, J. Control. Release 75 (2001) 173–181. [12] S.K. Bajpai, S.S. Saggu, Insulin release behavior of poly(methacrylamide-co-N-vinyl-2pyrrolidone-co-itaconic acid) hydrogel: an interesting probe, part II, J. Macromol. Sci. 44 (2007) 153–157. [13] J.J. Kim, K. Park, Molecular insulin delivery from glucose-sensitive hydrogel dosage forms, J. Control. Release 77 (2001) 39–47. [14] K. Moriyama, N. Yui, Regulated insulin release from biodegradable dextran hydrogels containing poly(ethylene glycol), J. Control. Release 42 (1996) 237–248. [15] Y.J. Kim, S. Choi, J.J. Koh, M. Lee, K.S. Ko, S.W. Kim, Controlled release of insulin from injectable biodegradable triblock copolymer, Pharm. Res. 18 (2001) 548–550. [16] M.T. Ende, N.A. Pappas, Transport of ionizable drugs and proteins in crosslinked poly(acrylic acid) and poly(acrylic acid-co-2-hydroxyethyl methacrylate) hydrogels. II. diffusion and release studies, J. Control. Release 48 (1997) 47–56. [17] S. Choi, M. Baudys, S.W. Kim, Control of blood glucose by novel GLP-1 delivery using biodegradable triblock copolymer of PLGA-PEG-PLGA in type 2 diabetic rats, Pharm. Res. 21 (2004) 827–831. [18] W.S. Shim, J.S. Yoo, Y.H. Bae, D.S. Lee, Novel injectable pH and temperature sensitive block copolymer hydrogel, Biomacromolecules 6 (2005) 2930–2934. [19] D.P. Huynh, W.S. Shim, J.H. Kim, D.S. Lee, pH/temperature sensitive poly(ethylene glycol)-based biodegradable polyester block copolymer hydrogels, Polymer 47 (2006) 7918–7926. [20] D.P. Huynh, M.K. Nguyen, B.S. Pi, M.S. Kim, S.Y. Chae, K.C. Lee, B.S. Kim, S.W. Kim, D. S. Lee, Functionalized injectable hydrogel for controlled insulin delivery, Biomaterials 29 (2008) 2527–2534. [21] D.P. Huynh, M.K. Nguyen, M.S. Kim, B.S. Kim, D.S. Lee, Molecular design of novel pH/temperature sensitive injectable hydrogels, Polymer (in press). [22] M. Takenaga, Y. Yamaguchi, A. Kitagawa, Y. Ogawa, Y. Izushima, R. Igarashi, A novel sustained-release formulation of insulin with dramatic reduction in initial rapid release, J. Control. Release 79 (2002) 80–91. [23] S. Choi, S.W. Kim, Controlled release of insulin from injectable biodegradable triblock copolymer depot in ZDF rats, Pharm. Res. 20 (2003) 2008–2010.