- Email: [email protected]
pH-responsive swelling behavior of collagen gels prepared by novel crosslinkers based on naturally derived di- or tricarboxylic acids
Acta Biomaterialia 3 (2007) 89–94 www.actamat-journals.com
pH-responsive swelling behavior of collagen gels prepared by novel crosslinkers based on naturally derived di- or tricarboxylic acids Hirofumi Saito a,b, Tetsushi Taguchi c,*, Hirokatsu Aoki b, Shun Murabayashi a, Yoshinori Mitamura a, Junzo Tanaka c, Tetsuya Tateishi c a
Graduate School of Information Science and Technology, The University of Hokkaido, N-14 W-9, Kita-ku, Sapporo 060-0814, Japan b Furuuchi Chemical Corporation, 6-17-17 Minami-oi, Shinagawa-ku, Tokyo 140-0013, Japan c Biomaterials Center, National Institute for Materials Science, 1-1 Namiki, Tsukuba, Ibaraki 305-0044, Japan Received 28 April 2006; received in revised form 14 July 2006; accepted 16 August 2006
Abstract The aim of this study was to compare the physicochemical properties of alkali-treated collagen (AlCol) gels prepared using two kinds of naturally derived crosslinkers made from citric and malic acids (CAD and MAD, respectively) that we have developed. From the crosslinking reaction between active ester groups and amino groups of AlCol, we successfully obtained AlCol gels, named AlColCAD and AlCol-MAD, prepared using CAD and MAD, respectively. The gelation time of the AlCol solution containing CAD initially decreased with increasing CAD concentration up to 70 mM, and then increased as the CAD concentration increased further. The gelation time reached its minimum and began to increase. On the other hand, for AlCol-MAD solution, gelation occurred within 40 s at any MAD concentration. Moreover, the residual amino groups in AlCol-CAD and AlCol-MAD were found to decrease with increasing CAD or MAD concentrations, whereas increased residual carboxyl groups were detected only in the case of AlCol-CAD. The swelling ratio of AlCol-CAD significantly increased at CAD concentrations above 50 mM. On the other hand, AlCol-MAD showed little increase in swelling ratio with increasing MAD concentration. Also, AlCol-CAD was swollen when the gels were immersed in a solution with high pH. On the other hand, no significant increase in swelling ratio was observed when AlCol-MAD was immersed in a similar solution. These results suggest that the different amounts of carboxyl groups in AlCol-CAD affected the swelling behavior of gels and that this pH-responsive AlCol-CAD has potential for drug delivery systems and tissue engineering. 2006 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. Keywords: Crosslinker; Biopolymer; Citric acid derivative; Malic acid derivative; Gel
1. Introduction pH-responsive hydrogels are polymer networks that have pendant acidic or basic functional groups which either accept or release protons [1–4]. Hydrogels of this kind that contain carboxyl or sulfonic acid groups show significant changes in their swelling behavior as a result of changing the external pH [5,6]. pH-responsive hydrogels are divided into two categories according to whether their constituent polymers are natu*
Corresponding author. Tel.: +81 29 860 4498; fax: +81 29 851 4714. E-mail address: [email protected] (T. Taguchi).
ral or synthetic. Synthetic polymers, however, are not biologically degradable by either hydrolytic or enzymatic mechanisms, and so these polymers have limited use as biomedical implant devices. Natural polymers, such as collagen and gelatin, have been used for tissue engineering and drug delivery systems [7–10]. These biodegradable polymers are often crosslinked with various compounds to obtain water-insoluble hydrogel matrices. Several crosslinkers have been used, including glutaraldehyde [11,12], carbodiimide [13] and epoxy compounds [14]. However, these crosslinkers, which remain in the resulting hydrogels, usually show high toxicity [15,16]. Therefore, it is necessary to develop an alternative, low-toxicity crosslinker. In order
1742-7061/$ - see front matter 2006 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.actbio.2006.08.003
90
H. Saito et al. / Acta Biomaterialia 3 (2007) 89–94
to overcome these obstacles, we developed a novel crosslinker known as citric acid derivative (CAD) with three active ester groups [17–21]. We confirmed that CAD can crosslink biodegradable polymers, which show excellent cytocompatibility compared with commercially available crosslinkers such as glutaraldehyde [18,19]. Furthermore, a tissue adhesive consisting of CAD and collagen had high bonding strength and excellent biocompatibility in vivo [17–19]. In our current study, we developed a novel crosslinker, malic acid derivative (MAD), which contains two active ester groups, in order to prepare alkali-treated collagen (AlCol) gels. We then compared the physicochemical properties (gelation time, storage moduli (G 0 ), swelling ratio, residual amino and carboxyl groups) of AlCol gels prepared using CAD and MAD. We also investigated the pH-responsive AlCol-CAD and AlCol-MAD. 2. Experimental
2.3. Preparation of AlCol gels AlCol, whose isoelectoric point is 5, has carboxyl groups generated by the hydrolysis of residual amide groups that exit in asparagine and glutamine of atelocollagen. AlCol was first dissolved in DMSO to obtain a 20% w/v solution. Then, 50 ll of CAD or MAD solution at crosslinker concentrations from 10 to 200 mM was added to a 200 ll of 20% w/v AlCol solution. The crosslinking reaction continued for 24 h at 37 C. The resulting AlCol gels were subsequently immersed in excess 0.1 M phosphate buffer solution (PBS) at different pHs (5.8–8.0) for 24 h at 37 C to remove DMSO. The AlCol gels were weighed under the equilibrium swollen state, and freeze-dried at 20 C for 48 h to determine the swelling ratio using the following equation: Swelling ratio ¼ ðW 0 W d Þ=W d ; where Wd and W0 are the weights of the dried and immersed AlCol gels, respectively.
2.1. Materials AlCol derived from pig skin was provided by Nitta Gelatin Inc. (Osaka, Japan). Citric acid, malic acid, Nhydroxysuccinimide (HOSu), tetrahydrofuran (THF), 2,4,6-trinitrobenzenesulfonic acid (TNBS), disodium hydrogenphosphate, sodium hydrogenphosphate, toluidine blue-O, dimethylsulfoxide (DMSO), ethanol, HCl and acetic acid were purchased from Wako Pure Chemical Industrials Ltd. (Osaka, Japan). Dicyclohexylcarbodiimide (DCC) was purchased from Kokusan Chemical Co., Ltd. (Tokyo, Japan). All other chemicals were used without further purification. 2.2. Preparation of CAD and MAD CAD and MAD were prepared by the method previously reported [22]. Briefly speaking, citric or malic acid was first dissolved in THF, and then HOSu and DCC were added. After mixing for 30 min, the mixture was concentrated with rotary evaporation under a reduced pressure to remove THF. The resulting mixture was recrystallized to yield pure CAD or MAD. Characterization of CAD and MAD was performed using 1H-NMR (JEOL EX300) and elemental analysis. The 1H-NMR and elemental analysis results of CAD were follows: 1H-NMR (DMSO-d6) d = 2.8 ppm (s, 12H, succinimidyl esters CH2 · 6), 3.4 ppm (s, 4H, CH2 · 2), 7.2 ppm (s, 1H, OH). Analysis. Calculated for C18H17N3O13: C, 44.73; H, 3.55; N, 8.69. Found: C, 44.83; H, 3.45; N, 8.58. The 1H-NMR and elemental analysis results of MAD were follows: 1H-NMR (DMSO-d6) d = 2.8 ppm (s, 8H, succinimidyl esters CH2 · 4), 3.1–3.3 ppm (m, 2 H, CH2), 4.8–4.9 ppm (m, 1H, CH), 6.73 ppm (d, 1H, OH). Analysis. Calculated for C12H12N2O9: C, 43.91; H, 3.68; N, 8.53. Found: C, 43.76; H, 3.51; N, 8.40.
2.4. Determination of residual amino and carboxyl group content in AlCol gels Determination of residual amino groups in AlCol gels was performed by a spectrophotometric method using TNBS [23]. AlCol gels prepared using different CAD or MAD concentrations were placed in 5 ml tubes. The next stage was to add 1 ml of 4% NaHCO3 and 1 ml of 0.1% TNBS to AlCol gels, which were then incubated for 2 h at 37 C. Then, after adding 3 ml of 6 N HCl, the AlCol gels were autoclaved for 1 h at 120 C to hydrolyze them. The mixed solutions were spectrophotometrically measured at 340 nm using a microplate-reader (GENios A-5082, Tecan Japan, Japan). The residual carboxyl groups in AlCol gels were estimated by staining with toluidine blue-O [24]. AlCol gels, prepared using CAD or MAD at crosslinker concentrations from 10 to 200 mM, were immersed in excess 0.1 M PBS at pH 7.0 at 37 C for 24 h. The AlCol gels were then stained with 5 · 104 M toluidine blue-O in NaHCO3/ Na2CO3 buffer solution (pH 10) for 3 h, and rinsed three times with NaHCO3/NaCO3 buffer solution (pH 10). The toluidine blue-O in AlCol gels was extracted with 50% v/v acetic acid solution. The spectrophotometric measurement of the extracts was carried out at 633 nm using ultraviolet–visible spectroscopy (Hitachi U-2001, Hitachi, Japan). 2.5. Rheological measurements The gelation time of AlCol solution containing CAD or MAD was determined with a rheometer (Rheostress RS1, Haake, Germany). The viscoelastic meter was equipped with plate–plate tools of 20 mm in diameter with a gap length of 1 mm. The temperature of the sample chamber was maintained at 37 C. AlCol solution (500 ll)
H. Saito et al. / Acta Biomaterialia 3 (2007) 89–94
91
containing CAD or MAD was first placed onto the stage of the viscoelastic meter. Time-dependent storage (G 0 ) and loss moduli (G00 ) were then measured at a frequency of 1 Hz and an oscillatory shear stress of 1.0 Pa. Gelation time was defined as the time at which G 0 = G00 , as reported in the literature [25,26]. 3. Results and discussion 3.1. Rheological measurements In order to compare the reactivities of CAD or MAD, the gelation time was determined by monitoring the G 0 and G00 of gel precursor solutions consisting of AlCol and CAD, or AlCol and MAD. Fig. 1 shows the dependency of the gelation time on CAD or MAD concentrations ranging from 10 to 300 mM. The gelation time of the AlCol solution containing CAD initially decreased with increasing CAD concentration up to 70 mM and then began to increase with increasing CAD concentration. This suggests that the crosslinking reaction was inhibited by CAD at elevated CAD concentration. There are two kinds of active ester groups in CAD, and their different reactivities may cause the formation of CADbearing AlCol as shown in Fig. 7(a). On the other hand, gelation of AlCol and MAD mixture solutions occurred within 40 s at any MAD concentration. These behaviors were similar to that of commercially available crosslinkers such as glutaraldehyde under the same conditions. Fig. 2 shows the effect of crosslinker concentration on the behavior of G 0 of AlCol-CAD and AlCol-MAD. In AlCol-CAD, G 0 decreased with increasing CAD concentration. In general, the increase of G 0 is due to the increasing number of chemical junctions responsible for the formation of the amide bonds [27]. That is, the decrease of G 0 indicates a reduction in crosslinking density. Therefore, the results from Fig. 2 suggest that the reduction in the crosslinking density of AlCol-CAD was due to the formation of CAD-bearing AlCol at elevated CAD concentrations
Fig. 1. Dependence of gelation time on the crosslinker concentration of CAD (d) or MAD (n) at 37 C. Error bars represent standard deviations; n = 3.
Fig. 2. Storage moduli (G 0 ) of AlCol gels prepared using CAD (d) or MAD (n) at different crosslinker concentrations. Error bars represent standard deviations; n = 3.
(Fig. 7(a)). On the other hand, the G 0 of AlCol-MAD showed no effect of G 0 with increasing MAD concentration. The possibility of MAD-bearing AlCol being formed in AlCol-MAD was considered; however, it was thought that most of the active ester groups in MAD reacted with the amino groups of AlCol without forming MAD-bearing AlCol as shown in Fig. 7(b). 3.2. Effect of crosslinker concentration on the swelling behavior of AlCol gels Fig. 3 shows the equilibrium swelling ratio of AlCol gels at different crosslinker concentrations of CAD and MAD. It was observed that the swelling ratio of AlCol-CAD and AlCol-MAD was saturated within 24 h. The swelling ratio of AlCol-CAD had slightly increased with increasing CAD concentration up to 50 mM, after which the swelling ratio showed a sharp increase. In general, the higher the crosslinking density of a gel, the lower the swelling ratio of a gel becomes [28]. That is, swelling ratio decreases with increasing crosslinker concentration. However, unusual swelling behavior was observed only in the case of AlCol-CAD as shown
Fig. 3. The swelling ratio of AlCol gels prepared using CAD (d) or MAD (n) at different crosslinker concentrations. Error bars represent standard deviations; n = 3.
92
H. Saito et al. / Acta Biomaterialia 3 (2007) 89–94
in Fig. 3. The swelling ratio of AlCol-CAD significantly increased at CAD concentrations above 50 mM. The results from Figs. 2 and 3 indicate that there was a decrease in crosslinking density of AlCol-CAD due to the formation of CAD-bearing AlCol (Fig. 7(a)). On the other hand, AlCol-MAD showed a slight increase in swelling ratio with increasing MAD concentration. This suggests that the slightly increased swelling ratio of AlCol-MAD was due to the effect of the secondary hydroxyl group of MAD covalently reacting with AlCol. 3.3. Determination of residual amino and carboxyl groups in AlCol gels Fig. 4 shows the amounts of residual amino groups in AlCol gel as functions of CAD and MAD concentrations. Spectrophotometric methods were employed to determine the residual amino and carboxyl groups in AlCol gels. Residual amino groups in AlCol-CAD decreased with increasing CAD concentration up to 15 mM. At CAD concentrations higher than 15 mM, no amino groups were detected. This means that amino groups completely reacted with the active ester groups of CAD. Theoretically, 39 mM of amino groups exist in 200 ll of 20% w/v AlCol solution. Therefore, 13 mM of CAD is required to completely react with all amino groups in the AlCol solution, because the CAD has three active ester groups per molecule. On the other hand, AlCol-MAD showed no amino groups at MAD concentrations above 20 mM. This suggests that the MAD also completely reacted with the residual amino groups of AlCol at this concentration. Fig. 5 shows the amounts of residual carboxyl groups in AlCol gels at different CAD and MAD concentrations. In AlCol-CAD, the remaining carboxyl groups in the gel increased with increasing CAD concentrations, whereas no significant change in the residual carboxyl groups was observed in AlCol-MAD. The results from Figs. 4 and 5 suggest that CAD-bearing AlCol were formed at elevated CAD concentrations as shown in Fig. 7(a).
Fig. 4. Residual amino group content in AlCol gels prepared using CAD (d) or MAD (n) at different crosslinker concentrations. Error bars represent standard deviations; n = 3.
Fig. 5. Residual carboxyl group content in AlCol gels prepared using CAD (d) or MAD (n) at different crosslinker concentrations. Error bars represent standard deviations; n = 3.
3.4. Effect of solution pH on the swelling behavior of AlCol gels To investigate the influence of environmental conditions on the swelling behavior of AlCol gels, AlCol-CAD and AlCol-MAD were immersed in buffer solutions with different pHs. Fig. 6 shows the equilibrium swelling ratio of AlColCAD and AlCol-MAD after immersion in 0.1 M PBS at different pHs for 24 h. CAD and MAD concentrations were fixed at 200 mM. pH-responsive swelling behavior was observed when AlCol-CAD was immersed in PBS, i.e. the swelling ratio of AlCol-CAD increased with increasing pH of PBS. However, pH-responsive AlCol-MAD was not observed. It is well known that ionization of carboxyl groups takes place at elevated pH of the buffer solution in anionic polymer gels containing carboxyl groups [29]. After gel ionization, the increased electric repulsions between negatively charged carboxyl groups cause the gel to swell. Therefore, as shown in Fig. 7(a) the increased swelling ratio of AlCol-CAD was due to the increased electric repulsion among the carboxyl groups of CA-bearing
Fig. 6. The swelling ratio of AlCol gels prepared using CAD (d) or MAD (n) at a crosslinker concentration of 200 mM in 0.1 M PBS at different pHs. Error bars represent standard deviations; n = 3.
H. Saito et al. / Acta Biomaterialia 3 (2007) 89–94
93
Fig. 7. The schematic illustration of crosslinking reaction of CAD (CAD concentration >70 mM) (a) or MAD (b) with AlCol.
AlCol generated from the hydrolysis of CAD-bearing AlCol, whereas the pH-responsive swelling behavior of AlCol-MAD was not observed because the formation of MA-bearing AlCol did not occur. Moreover, the isoelectric point of CA-bearing AlCol became lower than five due to the increasing levels of the carboxyl group. Therefore, it was expected that AlCol-CAD gels containing of CA-bearing AlCol were shrunk at pH < 5. The results from Figs. 3–5 indicated that the CA- or CAD-bearing AlCol formation occurs with increasing CAD concentration, resulting in the increased swelling ratio of the gels. These results support the proposal that the pH-responsive AlCol-CAD is due to the effect of CAbearing AlCol formation. On the other hand, it was found that the formation of MA-bearing AlCol did not occur in AlCol-MAD. In addition, in the present study we also performed a toxicity test by implantation of tissue adhesive consisting of AlCol-CAD. The results clearly showed that AlColCAD glue had little toxicity to tissue [18]. Therefore, we expected that CAD-bearing and CA-bearing AlCol gels would show low toxicity even if these constituents remained in vivo. These results suggest that pH-responsive AlCol-CAD has potential as a biomaterial, e.g. for drug delivery systems and tissue engineering. 4. Conclusion We have shown that pH-responsive hydrogels can be constructed from AlCol using CAD with three active ester groups. The residual amino groups in AlCol-CAD and AlCol-MAD decreased with increasing crosslinker concentration. However, only in the case of AlCol-CAD did the level of residual carboxyl groups increase with increasing crosslinker concentration. The swelling ratio of AlColCAD increased with increasing CAD concentration. AlCol-MAD showed slight increase in the swelling ratio with increasing MAD concentration. pH-responsive swelling behavior was observed when AlCol-CAD was
immersed in PBS at elevated pHs due to the formation of CA-bearing AlCol. On the other hand, such behavior of AlCol-MAD was not observed. Acknowledgements This work was financially supported in part by Industrial Technology Research Grant Program in ’04 from New Energy and Industrial Technology Development Organization (NEDO), Innovative Technology Research Grant from Japan Science and Technology Agency (JST), Japan, the Coordination Fund for Promoting Science and Technology from Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan (Development of Artificial Organs Utilizing with Nanotechnology and Materials Science). References [1] Suzuki H, Kumagai A. A disposable biosensor employing a glucosesensitive biochemomechanical gel. Biosens Bioelectron 2003;18: 1289–97. [2] Byrne ME, Park K, Peppas NA. Molecular imprinting within hydrogels. Adv Drug Deliv Rev 2002;54:149–61. [3] Hilt JZ, Byrne ME. Configurational biomimesis in drug delivery: molecular imprinting of biologically significant molecules. Adv Drug Deliv Rev 2004;56:1599–620. [4] Lee KY, Mooney DJ. Hydrogels for tissue engineering. Chem Rev 2001;101:1869–80. [5] Kim SJ, Lee CK, Kim SI. Electrical/pH responsive properties of poly(2-acrylamido-2-methylpropane sulfonic acid)/hyaluronic acid hydrogels. J Poly Sci 2004;92:1731–6. [6] Peppas LB, Peppas NA. Equilibrium swelling behavior of pHsensitive hydrogels. Chem Eng Sci 1991;46:715–22. [7] Taylor PM, Sachlos E, Dreger SA, Chester AH, Czernuszka JT, Yacoub MH. Interaction of human valve interstitial cells with collagen matrices manufactured using rapid prototyping. Biomaterials 2006;27:2733–7. [8] Ruszczak Z, Friess W. Collagen as a carrier for on-site delivery of antibacterial drugs. Adv Drug Deliv Rev 2003;55:1679–98. [9] Vandelli MA, Romagnoli M, Monti A, Gozzi M, Guerra P, Rivasi F, et al. Microwave-treated gelatin microspheres as drug delivery system. J Control Release 2004;96:67–84.
94
H. Saito et al. / Acta Biomaterialia 3 (2007) 89–94
[10] Li X, Jin L, Balian G, Laurencin CT, Anderson DG. Demineralized bone matrix gelatin as scaffold for osteochondral tissue engineering. Biomaterials 2006;27:2426–33. [11] Bigi A, Cojazzi G, Panzavolta S, Rubini K, Roveri N. Mechanical and thermal properties of gelatin films at different degrees of glutaraldehyde crosslinking. Biomaterials 2001;22:763–8. [12] Gough JE, Scotchford CA, Downes S. Cytotoxicity of glutaraldehyde crosslinked collagen/poly(vinyl alcohol) films is by the mechanism of apoptosis. J Biomed Mater Res 2002;61:121–30. [13] Damink LHH, Dijkstra PJ, Luyn MJA, Wachem PB, Nieuwenhuis P, Feijen J. Cross-linking of dermal sheep collagen using a water-soluble carbodiimide. Biomaterial 1996;17:765–73. [14] Wachem PB, Zeeman R, Dijkstra PJ, Feijen J, Hendriks M, Cahalan PT, et al. Characterization and biocompatibility of epoxy-crosslinked dermal sheep collagens. J Biomed Mater Res 1999;47:270–7. [15] Tsai CC, Huang RN, Sung HW, Liang HC. In vitro evaluation of the genotoxicity of a naturally occurring crosslinking agent (genipin) for biologic tissue fixation. J Biomed Mater Res 2000;52:58–65. [16] Chang Y, Tsai CC, Liang HC, Sung HW. In vivo evaluation of cellular and acellular bovine pericardia fixed with a naturally occurring crosslinking agent (genipin). Biomaterials 2002;23:2447–57. [17] Taguchi T, Saito H, Iwasashi M, Sakane M, Kakinoki S, Tanaka J. Development of a novel glue consisting of naturally-derived biomolecules: citric acid and human serum albumin. J Nanosci Nanotech, in press. [18] Taguchi T, Saito H, Uchida Y, Sakane M, Kobayashi H, Kataoka K, et al. Biocompatible high-strength glue consisting of citric acid derivative and collagen. Mater Sci Eng C 2006;26:9–13. [19] Taguchi T, Saito H, Uchida Y, Sakane M, Kobayashi H, Kataoka K, et al. Bonding of soft tissues using a novel tissue adhesive consisting of a citric acid derivative and collagen. Mater Sci Eng C 2004;24:775–80.
[20] Saito H, Taguchi T, Kobayashi H, Kataoka K, Tanaka J, Murabayashi S, et al. Physicochemical properties of gelatin gels prepared using citric acid derivative. Mater Sci Eng C 2004;24: 781–5. [21] Aoki H, Taguchi T, Saito H, Uchida Y, Sakane M, Kobayashi H, et al. Rheological evaluation of gelatin gels prepared with a citric acid derivative as a novel cross-linker. Mater Sci Eng C 2004;24:787–90. [22] Anderson GW, Zimmerman JE, Callahan FM. The use of esters of Nhydroxy-succinimide in peptide synthesis. J Am Chem Soc 1964;86:1839–42. [23] Bubnis WA, Ofner III CM. The determination of e-amino groups in soluble and poorly soluble proteinaceous materials by a spectrophotometric method using trinitrobenzenesulfonic acid. Anal Biochem 1992;207:129–33. [24] Nakajima N, Ikada Y. Mechanism of amide formation by carbodiimide for bioconjugation in aqueous media. Bioconjugate Chem 1995;6:123–30. [25] Fuchs T, Richtering W, Burchard W, Kajiwara K, Kitamura S. Gel point in physical gels: rheology and light scattering from thermoreversibly gelling schizophyllan. Polym Gels networks 1997;5:541–59. [26] Winter HH. Can the gel point of a cross-linking polymer be detected by the G 0 –G00 crossover? Polym Eng Sci 1987;27:1698–701. [27] Montembault A, Viton C, Domard A. Rheometric study of the gelation of chitosan in a hydroalcoholic medium. Biomaterials 2005;26:1633–43. [28] Charulatha V, Rajaram A. Influence of different crosslinking treatments on the physical properties of collagen membranes. Biomaterials 2003;24:759–67. [29] Khare AR, Peppas NA. Swelling/deswelling of anionic copolymer gels. Biomaterials 1995;16:559–67.
pH-responsive swelling behavior of collagen gels prepared by novel crosslinkers based on naturally derived di- or tricarboxylic acids Hirofumi Saito a,b, Tetsushi Taguchi c,*, Hirokatsu Aoki b, Shun Murabayashi a, Yoshinori Mitamura a, Junzo Tanaka c, Tetsuya Tateishi c a
Graduate School of Information Science and Technology, The University of Hokkaido, N-14 W-9, Kita-ku, Sapporo 060-0814, Japan b Furuuchi Chemical Corporation, 6-17-17 Minami-oi, Shinagawa-ku, Tokyo 140-0013, Japan c Biomaterials Center, National Institute for Materials Science, 1-1 Namiki, Tsukuba, Ibaraki 305-0044, Japan Received 28 April 2006; received in revised form 14 July 2006; accepted 16 August 2006
Abstract The aim of this study was to compare the physicochemical properties of alkali-treated collagen (AlCol) gels prepared using two kinds of naturally derived crosslinkers made from citric and malic acids (CAD and MAD, respectively) that we have developed. From the crosslinking reaction between active ester groups and amino groups of AlCol, we successfully obtained AlCol gels, named AlColCAD and AlCol-MAD, prepared using CAD and MAD, respectively. The gelation time of the AlCol solution containing CAD initially decreased with increasing CAD concentration up to 70 mM, and then increased as the CAD concentration increased further. The gelation time reached its minimum and began to increase. On the other hand, for AlCol-MAD solution, gelation occurred within 40 s at any MAD concentration. Moreover, the residual amino groups in AlCol-CAD and AlCol-MAD were found to decrease with increasing CAD or MAD concentrations, whereas increased residual carboxyl groups were detected only in the case of AlCol-CAD. The swelling ratio of AlCol-CAD significantly increased at CAD concentrations above 50 mM. On the other hand, AlCol-MAD showed little increase in swelling ratio with increasing MAD concentration. Also, AlCol-CAD was swollen when the gels were immersed in a solution with high pH. On the other hand, no significant increase in swelling ratio was observed when AlCol-MAD was immersed in a similar solution. These results suggest that the different amounts of carboxyl groups in AlCol-CAD affected the swelling behavior of gels and that this pH-responsive AlCol-CAD has potential for drug delivery systems and tissue engineering. 2006 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. Keywords: Crosslinker; Biopolymer; Citric acid derivative; Malic acid derivative; Gel
1. Introduction pH-responsive hydrogels are polymer networks that have pendant acidic or basic functional groups which either accept or release protons [1–4]. Hydrogels of this kind that contain carboxyl or sulfonic acid groups show significant changes in their swelling behavior as a result of changing the external pH [5,6]. pH-responsive hydrogels are divided into two categories according to whether their constituent polymers are natu*
Corresponding author. Tel.: +81 29 860 4498; fax: +81 29 851 4714. E-mail address: [email protected] (T. Taguchi).
ral or synthetic. Synthetic polymers, however, are not biologically degradable by either hydrolytic or enzymatic mechanisms, and so these polymers have limited use as biomedical implant devices. Natural polymers, such as collagen and gelatin, have been used for tissue engineering and drug delivery systems [7–10]. These biodegradable polymers are often crosslinked with various compounds to obtain water-insoluble hydrogel matrices. Several crosslinkers have been used, including glutaraldehyde [11,12], carbodiimide [13] and epoxy compounds [14]. However, these crosslinkers, which remain in the resulting hydrogels, usually show high toxicity [15,16]. Therefore, it is necessary to develop an alternative, low-toxicity crosslinker. In order
1742-7061/$ - see front matter 2006 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.actbio.2006.08.003
90
H. Saito et al. / Acta Biomaterialia 3 (2007) 89–94
to overcome these obstacles, we developed a novel crosslinker known as citric acid derivative (CAD) with three active ester groups [17–21]. We confirmed that CAD can crosslink biodegradable polymers, which show excellent cytocompatibility compared with commercially available crosslinkers such as glutaraldehyde [18,19]. Furthermore, a tissue adhesive consisting of CAD and collagen had high bonding strength and excellent biocompatibility in vivo [17–19]. In our current study, we developed a novel crosslinker, malic acid derivative (MAD), which contains two active ester groups, in order to prepare alkali-treated collagen (AlCol) gels. We then compared the physicochemical properties (gelation time, storage moduli (G 0 ), swelling ratio, residual amino and carboxyl groups) of AlCol gels prepared using CAD and MAD. We also investigated the pH-responsive AlCol-CAD and AlCol-MAD. 2. Experimental
2.3. Preparation of AlCol gels AlCol, whose isoelectoric point is 5, has carboxyl groups generated by the hydrolysis of residual amide groups that exit in asparagine and glutamine of atelocollagen. AlCol was first dissolved in DMSO to obtain a 20% w/v solution. Then, 50 ll of CAD or MAD solution at crosslinker concentrations from 10 to 200 mM was added to a 200 ll of 20% w/v AlCol solution. The crosslinking reaction continued for 24 h at 37 C. The resulting AlCol gels were subsequently immersed in excess 0.1 M phosphate buffer solution (PBS) at different pHs (5.8–8.0) for 24 h at 37 C to remove DMSO. The AlCol gels were weighed under the equilibrium swollen state, and freeze-dried at 20 C for 48 h to determine the swelling ratio using the following equation: Swelling ratio ¼ ðW 0 W d Þ=W d ; where Wd and W0 are the weights of the dried and immersed AlCol gels, respectively.
2.1. Materials AlCol derived from pig skin was provided by Nitta Gelatin Inc. (Osaka, Japan). Citric acid, malic acid, Nhydroxysuccinimide (HOSu), tetrahydrofuran (THF), 2,4,6-trinitrobenzenesulfonic acid (TNBS), disodium hydrogenphosphate, sodium hydrogenphosphate, toluidine blue-O, dimethylsulfoxide (DMSO), ethanol, HCl and acetic acid were purchased from Wako Pure Chemical Industrials Ltd. (Osaka, Japan). Dicyclohexylcarbodiimide (DCC) was purchased from Kokusan Chemical Co., Ltd. (Tokyo, Japan). All other chemicals were used without further purification. 2.2. Preparation of CAD and MAD CAD and MAD were prepared by the method previously reported [22]. Briefly speaking, citric or malic acid was first dissolved in THF, and then HOSu and DCC were added. After mixing for 30 min, the mixture was concentrated with rotary evaporation under a reduced pressure to remove THF. The resulting mixture was recrystallized to yield pure CAD or MAD. Characterization of CAD and MAD was performed using 1H-NMR (JEOL EX300) and elemental analysis. The 1H-NMR and elemental analysis results of CAD were follows: 1H-NMR (DMSO-d6) d = 2.8 ppm (s, 12H, succinimidyl esters CH2 · 6), 3.4 ppm (s, 4H, CH2 · 2), 7.2 ppm (s, 1H, OH). Analysis. Calculated for C18H17N3O13: C, 44.73; H, 3.55; N, 8.69. Found: C, 44.83; H, 3.45; N, 8.58. The 1H-NMR and elemental analysis results of MAD were follows: 1H-NMR (DMSO-d6) d = 2.8 ppm (s, 8H, succinimidyl esters CH2 · 4), 3.1–3.3 ppm (m, 2 H, CH2), 4.8–4.9 ppm (m, 1H, CH), 6.73 ppm (d, 1H, OH). Analysis. Calculated for C12H12N2O9: C, 43.91; H, 3.68; N, 8.53. Found: C, 43.76; H, 3.51; N, 8.40.
2.4. Determination of residual amino and carboxyl group content in AlCol gels Determination of residual amino groups in AlCol gels was performed by a spectrophotometric method using TNBS [23]. AlCol gels prepared using different CAD or MAD concentrations were placed in 5 ml tubes. The next stage was to add 1 ml of 4% NaHCO3 and 1 ml of 0.1% TNBS to AlCol gels, which were then incubated for 2 h at 37 C. Then, after adding 3 ml of 6 N HCl, the AlCol gels were autoclaved for 1 h at 120 C to hydrolyze them. The mixed solutions were spectrophotometrically measured at 340 nm using a microplate-reader (GENios A-5082, Tecan Japan, Japan). The residual carboxyl groups in AlCol gels were estimated by staining with toluidine blue-O [24]. AlCol gels, prepared using CAD or MAD at crosslinker concentrations from 10 to 200 mM, were immersed in excess 0.1 M PBS at pH 7.0 at 37 C for 24 h. The AlCol gels were then stained with 5 · 104 M toluidine blue-O in NaHCO3/ Na2CO3 buffer solution (pH 10) for 3 h, and rinsed three times with NaHCO3/NaCO3 buffer solution (pH 10). The toluidine blue-O in AlCol gels was extracted with 50% v/v acetic acid solution. The spectrophotometric measurement of the extracts was carried out at 633 nm using ultraviolet–visible spectroscopy (Hitachi U-2001, Hitachi, Japan). 2.5. Rheological measurements The gelation time of AlCol solution containing CAD or MAD was determined with a rheometer (Rheostress RS1, Haake, Germany). The viscoelastic meter was equipped with plate–plate tools of 20 mm in diameter with a gap length of 1 mm. The temperature of the sample chamber was maintained at 37 C. AlCol solution (500 ll)
H. Saito et al. / Acta Biomaterialia 3 (2007) 89–94
91
containing CAD or MAD was first placed onto the stage of the viscoelastic meter. Time-dependent storage (G 0 ) and loss moduli (G00 ) were then measured at a frequency of 1 Hz and an oscillatory shear stress of 1.0 Pa. Gelation time was defined as the time at which G 0 = G00 , as reported in the literature [25,26]. 3. Results and discussion 3.1. Rheological measurements In order to compare the reactivities of CAD or MAD, the gelation time was determined by monitoring the G 0 and G00 of gel precursor solutions consisting of AlCol and CAD, or AlCol and MAD. Fig. 1 shows the dependency of the gelation time on CAD or MAD concentrations ranging from 10 to 300 mM. The gelation time of the AlCol solution containing CAD initially decreased with increasing CAD concentration up to 70 mM and then began to increase with increasing CAD concentration. This suggests that the crosslinking reaction was inhibited by CAD at elevated CAD concentration. There are two kinds of active ester groups in CAD, and their different reactivities may cause the formation of CADbearing AlCol as shown in Fig. 7(a). On the other hand, gelation of AlCol and MAD mixture solutions occurred within 40 s at any MAD concentration. These behaviors were similar to that of commercially available crosslinkers such as glutaraldehyde under the same conditions. Fig. 2 shows the effect of crosslinker concentration on the behavior of G 0 of AlCol-CAD and AlCol-MAD. In AlCol-CAD, G 0 decreased with increasing CAD concentration. In general, the increase of G 0 is due to the increasing number of chemical junctions responsible for the formation of the amide bonds [27]. That is, the decrease of G 0 indicates a reduction in crosslinking density. Therefore, the results from Fig. 2 suggest that the reduction in the crosslinking density of AlCol-CAD was due to the formation of CAD-bearing AlCol at elevated CAD concentrations
Fig. 1. Dependence of gelation time on the crosslinker concentration of CAD (d) or MAD (n) at 37 C. Error bars represent standard deviations; n = 3.
Fig. 2. Storage moduli (G 0 ) of AlCol gels prepared using CAD (d) or MAD (n) at different crosslinker concentrations. Error bars represent standard deviations; n = 3.
(Fig. 7(a)). On the other hand, the G 0 of AlCol-MAD showed no effect of G 0 with increasing MAD concentration. The possibility of MAD-bearing AlCol being formed in AlCol-MAD was considered; however, it was thought that most of the active ester groups in MAD reacted with the amino groups of AlCol without forming MAD-bearing AlCol as shown in Fig. 7(b). 3.2. Effect of crosslinker concentration on the swelling behavior of AlCol gels Fig. 3 shows the equilibrium swelling ratio of AlCol gels at different crosslinker concentrations of CAD and MAD. It was observed that the swelling ratio of AlCol-CAD and AlCol-MAD was saturated within 24 h. The swelling ratio of AlCol-CAD had slightly increased with increasing CAD concentration up to 50 mM, after which the swelling ratio showed a sharp increase. In general, the higher the crosslinking density of a gel, the lower the swelling ratio of a gel becomes [28]. That is, swelling ratio decreases with increasing crosslinker concentration. However, unusual swelling behavior was observed only in the case of AlCol-CAD as shown
Fig. 3. The swelling ratio of AlCol gels prepared using CAD (d) or MAD (n) at different crosslinker concentrations. Error bars represent standard deviations; n = 3.
92
H. Saito et al. / Acta Biomaterialia 3 (2007) 89–94
in Fig. 3. The swelling ratio of AlCol-CAD significantly increased at CAD concentrations above 50 mM. The results from Figs. 2 and 3 indicate that there was a decrease in crosslinking density of AlCol-CAD due to the formation of CAD-bearing AlCol (Fig. 7(a)). On the other hand, AlCol-MAD showed a slight increase in swelling ratio with increasing MAD concentration. This suggests that the slightly increased swelling ratio of AlCol-MAD was due to the effect of the secondary hydroxyl group of MAD covalently reacting with AlCol. 3.3. Determination of residual amino and carboxyl groups in AlCol gels Fig. 4 shows the amounts of residual amino groups in AlCol gel as functions of CAD and MAD concentrations. Spectrophotometric methods were employed to determine the residual amino and carboxyl groups in AlCol gels. Residual amino groups in AlCol-CAD decreased with increasing CAD concentration up to 15 mM. At CAD concentrations higher than 15 mM, no amino groups were detected. This means that amino groups completely reacted with the active ester groups of CAD. Theoretically, 39 mM of amino groups exist in 200 ll of 20% w/v AlCol solution. Therefore, 13 mM of CAD is required to completely react with all amino groups in the AlCol solution, because the CAD has three active ester groups per molecule. On the other hand, AlCol-MAD showed no amino groups at MAD concentrations above 20 mM. This suggests that the MAD also completely reacted with the residual amino groups of AlCol at this concentration. Fig. 5 shows the amounts of residual carboxyl groups in AlCol gels at different CAD and MAD concentrations. In AlCol-CAD, the remaining carboxyl groups in the gel increased with increasing CAD concentrations, whereas no significant change in the residual carboxyl groups was observed in AlCol-MAD. The results from Figs. 4 and 5 suggest that CAD-bearing AlCol were formed at elevated CAD concentrations as shown in Fig. 7(a).
Fig. 4. Residual amino group content in AlCol gels prepared using CAD (d) or MAD (n) at different crosslinker concentrations. Error bars represent standard deviations; n = 3.
Fig. 5. Residual carboxyl group content in AlCol gels prepared using CAD (d) or MAD (n) at different crosslinker concentrations. Error bars represent standard deviations; n = 3.
3.4. Effect of solution pH on the swelling behavior of AlCol gels To investigate the influence of environmental conditions on the swelling behavior of AlCol gels, AlCol-CAD and AlCol-MAD were immersed in buffer solutions with different pHs. Fig. 6 shows the equilibrium swelling ratio of AlColCAD and AlCol-MAD after immersion in 0.1 M PBS at different pHs for 24 h. CAD and MAD concentrations were fixed at 200 mM. pH-responsive swelling behavior was observed when AlCol-CAD was immersed in PBS, i.e. the swelling ratio of AlCol-CAD increased with increasing pH of PBS. However, pH-responsive AlCol-MAD was not observed. It is well known that ionization of carboxyl groups takes place at elevated pH of the buffer solution in anionic polymer gels containing carboxyl groups [29]. After gel ionization, the increased electric repulsions between negatively charged carboxyl groups cause the gel to swell. Therefore, as shown in Fig. 7(a) the increased swelling ratio of AlCol-CAD was due to the increased electric repulsion among the carboxyl groups of CA-bearing
Fig. 6. The swelling ratio of AlCol gels prepared using CAD (d) or MAD (n) at a crosslinker concentration of 200 mM in 0.1 M PBS at different pHs. Error bars represent standard deviations; n = 3.
H. Saito et al. / Acta Biomaterialia 3 (2007) 89–94
93
Fig. 7. The schematic illustration of crosslinking reaction of CAD (CAD concentration >70 mM) (a) or MAD (b) with AlCol.
AlCol generated from the hydrolysis of CAD-bearing AlCol, whereas the pH-responsive swelling behavior of AlCol-MAD was not observed because the formation of MA-bearing AlCol did not occur. Moreover, the isoelectric point of CA-bearing AlCol became lower than five due to the increasing levels of the carboxyl group. Therefore, it was expected that AlCol-CAD gels containing of CA-bearing AlCol were shrunk at pH < 5. The results from Figs. 3–5 indicated that the CA- or CAD-bearing AlCol formation occurs with increasing CAD concentration, resulting in the increased swelling ratio of the gels. These results support the proposal that the pH-responsive AlCol-CAD is due to the effect of CAbearing AlCol formation. On the other hand, it was found that the formation of MA-bearing AlCol did not occur in AlCol-MAD. In addition, in the present study we also performed a toxicity test by implantation of tissue adhesive consisting of AlCol-CAD. The results clearly showed that AlColCAD glue had little toxicity to tissue [18]. Therefore, we expected that CAD-bearing and CA-bearing AlCol gels would show low toxicity even if these constituents remained in vivo. These results suggest that pH-responsive AlCol-CAD has potential as a biomaterial, e.g. for drug delivery systems and tissue engineering. 4. Conclusion We have shown that pH-responsive hydrogels can be constructed from AlCol using CAD with three active ester groups. The residual amino groups in AlCol-CAD and AlCol-MAD decreased with increasing crosslinker concentration. However, only in the case of AlCol-CAD did the level of residual carboxyl groups increase with increasing crosslinker concentration. The swelling ratio of AlColCAD increased with increasing CAD concentration. AlCol-MAD showed slight increase in the swelling ratio with increasing MAD concentration. pH-responsive swelling behavior was observed when AlCol-CAD was
immersed in PBS at elevated pHs due to the formation of CA-bearing AlCol. On the other hand, such behavior of AlCol-MAD was not observed. Acknowledgements This work was financially supported in part by Industrial Technology Research Grant Program in ’04 from New Energy and Industrial Technology Development Organization (NEDO), Innovative Technology Research Grant from Japan Science and Technology Agency (JST), Japan, the Coordination Fund for Promoting Science and Technology from Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan (Development of Artificial Organs Utilizing with Nanotechnology and Materials Science). References [1] Suzuki H, Kumagai A. A disposable biosensor employing a glucosesensitive biochemomechanical gel. Biosens Bioelectron 2003;18: 1289–97. [2] Byrne ME, Park K, Peppas NA. Molecular imprinting within hydrogels. Adv Drug Deliv Rev 2002;54:149–61. [3] Hilt JZ, Byrne ME. Configurational biomimesis in drug delivery: molecular imprinting of biologically significant molecules. Adv Drug Deliv Rev 2004;56:1599–620. [4] Lee KY, Mooney DJ. Hydrogels for tissue engineering. Chem Rev 2001;101:1869–80. [5] Kim SJ, Lee CK, Kim SI. Electrical/pH responsive properties of poly(2-acrylamido-2-methylpropane sulfonic acid)/hyaluronic acid hydrogels. J Poly Sci 2004;92:1731–6. [6] Peppas LB, Peppas NA. Equilibrium swelling behavior of pHsensitive hydrogels. Chem Eng Sci 1991;46:715–22. [7] Taylor PM, Sachlos E, Dreger SA, Chester AH, Czernuszka JT, Yacoub MH. Interaction of human valve interstitial cells with collagen matrices manufactured using rapid prototyping. Biomaterials 2006;27:2733–7. [8] Ruszczak Z, Friess W. Collagen as a carrier for on-site delivery of antibacterial drugs. Adv Drug Deliv Rev 2003;55:1679–98. [9] Vandelli MA, Romagnoli M, Monti A, Gozzi M, Guerra P, Rivasi F, et al. Microwave-treated gelatin microspheres as drug delivery system. J Control Release 2004;96:67–84.
94
H. Saito et al. / Acta Biomaterialia 3 (2007) 89–94
[10] Li X, Jin L, Balian G, Laurencin CT, Anderson DG. Demineralized bone matrix gelatin as scaffold for osteochondral tissue engineering. Biomaterials 2006;27:2426–33. [11] Bigi A, Cojazzi G, Panzavolta S, Rubini K, Roveri N. Mechanical and thermal properties of gelatin films at different degrees of glutaraldehyde crosslinking. Biomaterials 2001;22:763–8. [12] Gough JE, Scotchford CA, Downes S. Cytotoxicity of glutaraldehyde crosslinked collagen/poly(vinyl alcohol) films is by the mechanism of apoptosis. J Biomed Mater Res 2002;61:121–30. [13] Damink LHH, Dijkstra PJ, Luyn MJA, Wachem PB, Nieuwenhuis P, Feijen J. Cross-linking of dermal sheep collagen using a water-soluble carbodiimide. Biomaterial 1996;17:765–73. [14] Wachem PB, Zeeman R, Dijkstra PJ, Feijen J, Hendriks M, Cahalan PT, et al. Characterization and biocompatibility of epoxy-crosslinked dermal sheep collagens. J Biomed Mater Res 1999;47:270–7. [15] Tsai CC, Huang RN, Sung HW, Liang HC. In vitro evaluation of the genotoxicity of a naturally occurring crosslinking agent (genipin) for biologic tissue fixation. J Biomed Mater Res 2000;52:58–65. [16] Chang Y, Tsai CC, Liang HC, Sung HW. In vivo evaluation of cellular and acellular bovine pericardia fixed with a naturally occurring crosslinking agent (genipin). Biomaterials 2002;23:2447–57. [17] Taguchi T, Saito H, Iwasashi M, Sakane M, Kakinoki S, Tanaka J. Development of a novel glue consisting of naturally-derived biomolecules: citric acid and human serum albumin. J Nanosci Nanotech, in press. [18] Taguchi T, Saito H, Uchida Y, Sakane M, Kobayashi H, Kataoka K, et al. Biocompatible high-strength glue consisting of citric acid derivative and collagen. Mater Sci Eng C 2006;26:9–13. [19] Taguchi T, Saito H, Uchida Y, Sakane M, Kobayashi H, Kataoka K, et al. Bonding of soft tissues using a novel tissue adhesive consisting of a citric acid derivative and collagen. Mater Sci Eng C 2004;24:775–80.
[20] Saito H, Taguchi T, Kobayashi H, Kataoka K, Tanaka J, Murabayashi S, et al. Physicochemical properties of gelatin gels prepared using citric acid derivative. Mater Sci Eng C 2004;24: 781–5. [21] Aoki H, Taguchi T, Saito H, Uchida Y, Sakane M, Kobayashi H, et al. Rheological evaluation of gelatin gels prepared with a citric acid derivative as a novel cross-linker. Mater Sci Eng C 2004;24:787–90. [22] Anderson GW, Zimmerman JE, Callahan FM. The use of esters of Nhydroxy-succinimide in peptide synthesis. J Am Chem Soc 1964;86:1839–42. [23] Bubnis WA, Ofner III CM. The determination of e-amino groups in soluble and poorly soluble proteinaceous materials by a spectrophotometric method using trinitrobenzenesulfonic acid. Anal Biochem 1992;207:129–33. [24] Nakajima N, Ikada Y. Mechanism of amide formation by carbodiimide for bioconjugation in aqueous media. Bioconjugate Chem 1995;6:123–30. [25] Fuchs T, Richtering W, Burchard W, Kajiwara K, Kitamura S. Gel point in physical gels: rheology and light scattering from thermoreversibly gelling schizophyllan. Polym Gels networks 1997;5:541–59. [26] Winter HH. Can the gel point of a cross-linking polymer be detected by the G 0 –G00 crossover? Polym Eng Sci 1987;27:1698–701. [27] Montembault A, Viton C, Domard A. Rheometric study of the gelation of chitosan in a hydroalcoholic medium. Biomaterials 2005;26:1633–43. [28] Charulatha V, Rajaram A. Influence of different crosslinking treatments on the physical properties of collagen membranes. Biomaterials 2003;24:759–67. [29] Khare AR, Peppas NA. Swelling/deswelling of anionic copolymer gels. Biomaterials 1995;16:559–67.