Colloids and Surfaces B: Biointerfaces 126 (2015) 575–579
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Short Communication
Partially photodegradable hybrid hydrogels with elasticity tunable by light irradiation Fumiki Yanagawa a,1 , Takeomi Mizutani b,1 , Shinji Sugiura a,∗,1 , Toshiyuki Takagi a , Kimio Sumaru a , Toshiyuki Kanamori a a Research Center for Stem Cell Engineering, National Institute of Advanced Industrial Science and Technology (AIST), Central 5th, 1-1-1 Higashi, Tsukuba, Ibaraki 305-8565, Japan b Faculty of Advanced Life Science, Hokkaido University, Kita 10 Nishi 8, Kita-ku, Sapporo 060-0810, Japan
a r t i c l e
i n f o
Article history: Received 1 September 2014 Received in revised form 22 October 2014 Accepted 13 November 2014 Available online 4 December 2014 Keywords: Hydrogel Crosslinker Photodegradable Elasticity
a b s t r a c t This paper reports a simple technique to synthesize elasticity tunable hybrid hydrogels using photocleavable (N-hydroxysuccinimide terminated photocleavable tetra-arm poly(ethylene glycol); NHS-PC-4armPEG) and non-photocleavable (N-hydroxysuccinimide terminated tetra-arm poly(ethylene glycol); NHS-4armPEG) activated-ester type crosslinkers. Partially photodegradable hybrid hydrogels were synthesized by reacting the crosslinker mixture with amino-terminated tetra-arm poly(ethylene glycol) (amino-4armPEG). The photocleavable crosslinks are cleaved by irradiating light while the non-photocleavable crosslinks remain intact, resulting in decreased elasticity. We demonstrate that hydrogel elasticity can be controlled by adjusting the ratio of photocleavable NHS-PC-4armPEG and non-photocleavable NHS-4armPEG, and by varying the light exposure energy. We also show how micropatterned elasticity can be obtained in the hydrogels by irradiating with micropatterned light. These techniques could provide a novel platform to tailor the elasticity of hydrogels with microscale precision for biological studies in the near future. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Hydrogels are attractive materials due to their biocompatibility, efficient mass transfer, and ability to encapsulate bioactive molecules and cells [1]. Mechanical properties of hydrogels are essential for cell adhesion, migration, maintenance, and differentiation. Therefore, mechanical properties should be readily customized to satisfy the requirements for particular applications [2]. Cells such as fibroblasts, endothelial cells, smooth muscle cells, and mesenchymal stem cells [3–6] behave differently depending on the mechanical properties of the neighboring extracellular matrix (ECM) [7,8]. It is therefore important to reproduce soft and stiff tissue microenvironments for these cells to exhibit the desired functions. Several research groups have previously reported techniques to control hydrogel stiffness [9,10] and stiffness gradients [11–13]. In previous studies, stiffness gradients were generated using gradient mixers [14], microfluidic gradient generators [15], and
∗ Corresponding author. Tel.: +81 29 861 6286; fax: +81 29 861 6278. E-mail address: [email protected] (S. Sugiura). 1 These authors contributed equally to this work. http://dx.doi.org/10.1016/j.colsurfb.2014.11.020 0927-7765/© 2014 Elsevier B.V. All rights reserved.
diffusion chambers [16]. Photolithography is a useful tool to fabricate microstructures in hydrogels and has been used to construct microscale 3D tissue models [17–19]. To date, several techniques have been proposed to control mechanical properties by light irradiation onto photoresponsive materials [6,20–22]. Especially, photodegradable hydrogels recently emerged as powerful materials for various applications in biotechnology [23,24] due to their ability to fabricate microstructures in hydrogels [25]. The elasticity of these photodegradable hydrogels was tuned and the elasticity gradient was formed by controlling exposure energy and irradiation gradient [13,26]. More recently, we reported an activated-ester-type photocleavable crosslinker, N-hydroxysuccinimide terminated photocleavable tetra-arm poly(ethylene glycol) (NHS-PC-4armPEG), as a convenient material to prepare photodegradable hydrogels. Our NHS-PC-4armPEG forms photodegradable hydrogels by reacting with biocompatible polymers containing amino moieties, such as amino-terminated tetra-arm poly(ethylene glycol) (amino4armPEG) and gelatin [27]. Microstructures can be fabricated in these hydrogels by light irradiation through photomasks. In this study, we demonstrate that the elasticity of hybrid hydrogels prepared with photocleavable NHS-PC-4armPEG and non-photocleavable N-hydroxysuccinimide terminated tetra-arm
576
F. Yanagawa et al. / Colloids and Surfaces B: Biointerfaces 126 (2015) 575–579
Fig. 1. Schematic diagram of the synthesis of the partially photodegradable hybrid hydrogels with photocleavable NHS-PC-4armPEG and non-photocleavable NHS-4armPEG, and the subsequent tuning of elasticity by light irradiation.
poly(ethylene glycol) (NHS-4armPEG) can be controlled by light irradiation, and that geometrical elasticity micropatterns can be formed in the hydrogels by irradiating with micropatterned light. 2. Materials and methods 2.1. Materials All reagents were purchased from Sigma–Aldrich (St. Louis, MO) unless otherwise specified. NHS-PC-4armPEG (Mw = 12,062), whose structural formula is shown in Scheme S1 in the Supporting Information (SI), was synthesized according to the protocol reported in our previous study [27]. NHS-4armPEG (Scheme S2 in the SI, Mw = 10,998) and amino-4armPEG (Scheme S3 in the SI, Mw = 9617) were obtained from NOF Corp. (Tokyo, Japan). 2.2. Hydrogel preparation Fig. 1 illustrates the method used for synthesizing hydrogels using crosslinking amino-terminated polymers. A prepolymer solution including 10 mM amino-4armPEG was prepared in a 1:1 (v/v) mixture with phosphate buffered saline (PBS, Invitrogen Corp., Carlsbad, CA) and 300 mM 4-(2-hydroxyethyl)1-piperazineethanesulfonic acid (HEPES, Wako Pure Chemical Industries, Osaka, Japan) buffer (pH 6.0). A photocleavable crosslinker solution containing 10 mM synthesized NHS-PC4armPEG and a non-photocleavable crosslinker solution containing 10 mM NHS-4armPEG were prepared in a 10 mM phthalate acid buffer (pH 4.0, Wako) with 140 mM NaCl (Wako). ‘Premixedcrosslinker’ solutions were prepared by mixing the photocleavable and non-photocleavable crosslinker solutions at ratios of 1:9–9:1 (v/v), resulting in the ‘Premixed-crosslinker’ solutions containing
10 to 90 mol% of the photocleavable crosslinker. The prepolymer and ‘premixed-crosslinker’ solutions were then mixed at a 1:1 (v/v) ratio. Immediately after mixing, 10–30 L of the mixture was pipetted onto an amino-coated glass slide (MAS coat, Matsunami Glass Corp., Osaka, Japan), and was covered with a cover slip. To control the thickness of the hydrogels, polyethylene terephthalate films (thickness = 25 m) were used as spacers to separate the glass slide and the cover slip. The sample was then placed in a humid incubator for 24 h at 37 ◦ C, allowing the crosslinking reaction between the NHS-activated-ester moieties and the amino moieties to produce hydrogel samples (Fig. 1).
2.3. Light irradiation After the incubation for hydrogel formation, the hydrogel sample was placed topside-down on the sample stage of a light irradiation unit. The sample was then exposed to light (365 nm, 30 mW/cm2 , 10–40 s) from an ultra violet (UV) light source (UVE251S, San-Ei Electric, Osaka, Japan) using a combination of optical filters (long-pass filter = 350 nm, short-pass filter = 385 nm), allowing the photocleavage of o-nitrobenzyl moieties in the hydrogel (Fig. 1). The sample was then immersed in MilliQ water (EMD Milipore, Billerica, USA) for 24 h at 37 ◦ C to erode the photocleaved polymer. For micropatterned light irradiation, a mask-free irradiation system was used, in which a computer controlled microprojection device (DESM-01, Engineering System, Co., Matsumoto, Japan) was attached to an inverted microscope (IX71, Olympus, Tokyo, Japan). A binary micropattern image with arrayed circles (circle diameter: 10 m, pitch: 50 m) was designed using Adobe Illustrator. Following the incubation, the hydrogel sample was immersed in MilliQ water for 24 h at 37 ◦ C, and then placed in a covered Petri
F. Yanagawa et al. / Colloids and Surfaces B: Biointerfaces 126 (2015) 575–579
577
dish on the sample stage to be exposed to micropatterned light through ×10 objective lens (Olympus, numerical aperture = 0.3). After adjusting the focus plane to project a clear image on its surface, the hydrogel sample was exposed to micropatterned light (365 nm, 125 mW/cm2 , 7.2 s) from the computer controlled microprojection device. The sample was then immersed in MilliQ water for 24 h at 37 ◦ C to erode the photocleaved polymer. 2.4. Elasticity measurements A force mapping mode of an atomic force microscopy (AFM) was used to measure the elasticity of the hydrogels with micrometerscale resolution as reported previously [28,29]. We used a commercial AFM equipped with a piezo scanner (maximal scan range: 100 m for the x and y-axes, 10 m for the z-axis; Seiko Instruments Inc., Chiba Japan) and a silicon-nitride cantilever (Olympus; spring constant = 0.02 N/m). After preparing the samples as described in Sections 2.2 and 2.3, excess surface water was removed with a pipette prior to loading the hydrogel on the sample stage. Immediately after sample loading, the pyramidal tip located on the cantilever was indented on the hydrogel surface, and the force acting on the cantilever (here called ‘force curve’) was recorded at room temperature as a function of the displacement of the piezo scanner. Typical force curves for different hydrogels are shown in Fig. S1 in the SI. Force curves were fitted to the Hertzian elastic contact model for a pyramidal indenter giving the local elasticity (Young’s modulus) of the measured points as a fitting parameter. For statistical comparisons of sample elasticity, force curves from more than 1000 different points in the scan area (90 m × 90 m) were analyzed for each sample and frequency distributions of elasticity were plotted (examples are shown in Fig. S2 in the SI). The mean values and standard deviations of elasticity were obtained by fitting the Gaussian distribution to the frequency distribution.
Fig. 2. Comparison of the mean elasticities of the exposed (at 900 mJ/cm2 exposure) and unexposed hydrogels composed of 5 mM amino-4armPEG and 5 mM crosslinker with 10, 30, 50, 70, and 90 mol% of NHS-PC-4armPEG. Error bars denote the standard deviations of over 1000 different measurement points.
3. Results and discussion Partially photodegradable hybrid hydrogels composed of NHS-PC-4armPEG, NHS-4armPEG, and amino-4armPEG were synthesized. The hydrogels were exposed to light to examine its effect on their elasticity. The light exposure conditions (365 nm, 30 mW/cm2 , 30 s) were chosen based on our previous study on the photocleavage behavior of o-nitrobenzyl moiety [27]. The hydrogels became yellowish after light exposure owing to the photocleavage of o-nitrobenzyl moiety. Fig. 2 shows the mean elasticities of exposed and unexposed hydrogels of different compositions. The mean elasticity of the unexposed hydrogels decreased with increasing proportions of photocleavable NHS-PC4armPEG. This change was probably due to the different reactivity of the NHS-activated-ester moiety in NHS-PC-4armPEG and NHS4armPEG to the amino moiety on amino-4armPEG (details of each molecular formula are available in Schemes S1–S3 in the SI). In the case of 30 and 50 mol% photocleavable NHS-PC4armPEG content, a significant decrease in elasticity was observed in the exposed hydrogels while their shape was retained. As expected, higher concentrations of NHS-PC-4armPEG led to a greater decrease in elasticity after exposure. When the photocleavable NHS-PC-4armPEG content was greater than 70 mol%, the hydrogels degraded completely and lost their shape. According to our previous study on photodegradable hydrogels, the degradation depth is determined by the exposure energy [27]. To assess whether light irradiation can be used to control hydrogel elasticity, this was investigated for hydrogels prepared with a crosslinker containing 50 mol% of NHS-PC-4armPEG and amino4armPEG. As shown in Fig. 3, the elasticity decreased monotonically
Fig. 3. Effect of light exposure energy on the mean elasticities of the hydrogels composed of 5 mM amino-4armPEG and 5 mM crosslinker with 50 mol% of NHSPC-4armPEG. Error bars denote the standard deviations of over 1000 different measurement points.
for exposures up to 300 mJ/cm2 , beyond which the elasticity varied little. Under these conditions, most of the o-nitrobenzyl moieties in the hydrogels were cleaved such that the hydrogel structure was probably maintained by the remaining polymer network composed of amino-4armPEG and non-photocleavable NHS-4armPEG. Hence, for sufficiently high exposures, the elasticity of the exposed hydrogels was determined by the composition of the crosslinker. Based on rubber elasticity theory, the elasticity of hydrogels is highly dependent on the effective number of crosslinking points [30]. Our results thereby suggest that light irradiation successfully decreased the crosslinking density in the hydrogel by photocleaving the o-nitrobenzyl moieties in photocleavable NHS-PC-4armPEG while leaving the non-photocleavable crosslinks formed by NHS4armPEG intact. As a proof of concept, we demonstrated micropatterning of elasticity in the partially photodegradable hybrid hydrogels. The hydrogels prepared with a crosslinker containing 50 mol% NHS-PC-4armPEG were exposed to micropatterned light (365 nm, 125 mW/cm2 , 7.2 s) using a mask-free irradiation system, through which an array of circles (resolution: 1.4 m, circles diameter: 10 m, pitch: 50 m) was projected and formed on the hydrogel samples. Fig. 4 shows the elasticity mapping images of the exposed hydrogel. The elasticity pattern on the hydrogel reflects
578
F. Yanagawa et al. / Colloids and Surfaces B: Biointerfaces 126 (2015) 575–579
Fig. 4. (a) The array of circles (diameter: 10 m, pitch: 50 m) used as a binary micropattern image. (b) The elasticity micropattern formed in the hydrogels by mask-free irradiation with micropatterned light. A hydrogel composed of 5 mM amino-4armPEG and 5 mM crosslinker with 50 mol% NHS-PC-4armPEG was exposed to 125 mW/cm2 micropatterned light at a wavelength of 365 nm for 7.2 s.
the micropatterning image, with the elasticity in the exposed regions decreasing as a result of decreased crosslinking density caused by photocleavage of the o-nitrobenzyl moieties. Fig. 4 also shows that the elasticity in the unexposed regions was not affected, demonstrating that irradiation with micropatterned light alters the hydrogel elasticity with microscale precision. The micropatterned elasticity shown in Fig. 4 exhibited different elasticity at different locations; the elasticity in the two circle areas in the right side was lower than that in the circles in the left side. We think that at least the following three factors are possibly affecting micropatterning elasticity. Firstly, focusing is very important to project a clear image on the sample surface using our microprojection device, while precise focusing on the transparent hydrogel sample is generally difficult. Secondly, it is also difficult to prepare flat samples with micrometer scale precision by using polyethylene terephthalate film spacers. Finally, the swelling of the hydrogel also alters the thickness of the hydrogel and makes it difficult to adjust the focus precisely. Therefore, inevitable defocus might cause the inhomogeneous illumination. Especially, if the sample was not flat, a clear image is projected only on the focused plane and a defocused image is projected at different z-positions. These effects potentially cause defocused projection on the sample surface at different locations. The elasticity of the partially photodegradable hybrid hydrogels presented in this work can be tuned by varying their composition while microscale geometrical elasticity patterns can be designed using micropatterned light irradiation. In the current study, we prepared PEG-based hydrogels composed of NHS-PC-4armPEG, NHS-4armPEG, and amino-4armPEG. Since PEG itself is highly inert and lacks biolological functionalities, few cells adhered onto the PEG-based partially photodegradable hybrid hydrogels as expected (Fig. S3 in the SI). No effect of light irradiation was observed on cell adhesion on these hydrogels. In order to enhance cell adhesion on the PEG-based partially photodegradable hybrid hydrogels, introduction of RGD (arginine-glycine-aspartic acid) peptide into the hydrogel is probably effective as reported for other PEG-based hydrogels [31,32]. There have been a couple of reports to control hydrogel elasticity by light irradiation onto the photodegradable hydrogels [13,26]. Compared to these studies, our approach to use the partially photodegradable hybrid hydrogel enables us to define the initial and resulting elasticity without precise control of exposure energy. Furthermore, our technique is theoretically applicable to synthesize biomaterial-based elasticity-tunable hybrid hydrogels because
the activated-ester-type crosslinker can react with a variety of polymers containing primary amine moieties [27]. As an example, we previously reported that NHS-PC-4armPEG can react with gelatin to generate a photodegradable hydrogel. Since both the photocleavable and non-photocleavable crosslinkers used in this study can react with amino-moieties in biocompatible polymers, the technique may be used to investigate cell behavior on hydrogels with tailored stiffness patterns composed of gelatin, collagen, fibronectin and chitosan. We expect that adherent cells adhere onto these biomaterial-based, partially photodegradable hybrid hydrogels. Photocleavage reaction takes place immediately upon exposure to light. Although the erosion of the polymer and the swelling of the hydrogel takes some time, the elasticity of our partially photodegradable hybrid hydrogels can be controlled on-demand. They could therefore potentially be applied to the dynamic control of elasticity in long-term cell culture processes, as reported in a previous study with a similar approach [6]. As described above, hydrogels with geometrically patterned elasticity and dynamically tunable elasticity could find novel applications in tissue engineering by mimicking soft and stiff tissue microenvironments in vivo. We therefore envision that in the near future, this simple technique could find many applications in the fields of cell manipulation and tissue engineering. 4. Conclusion We have presented a simple technique to synthesize elasticity tunable hybrid hydrogels with photocleavable NHS-PC-4armPEG and non-photocleavable NHS-4armPEG. The approach presented in this study provides hydrogels whose elasticity can be tailored by varying their composition, and by exposing them to light. We also demonstrated elasticity micropatterning by irradiating the hydrogels with micropatterned light. This technique could provide a novel platform for the precise control of hydrogel elasticity that should prove useful for biological studies in the near future. Author contributions These authors contributed equally. FY, TM, and SS conceived and planned the research, and wrote the paper. TT synthesized the materials. FY, TM, and SS designed the experimental protocols. FY prepared hydrogel samples and TM performed AFM measurements.
F. Yanagawa et al. / Colloids and Surfaces B: Biointerfaces 126 (2015) 575–579
FY, TM and SS analyzed the data. KS and TK supervised the research. All authors revised the manuscript and agreed on its final contents. Funding sources This research was supported by KAKENHI (24106512 and 24106502). Acknowledgments We thank Dr. T. Sato and Ms. K. Morishita for technical help and helpful comments on rheological characterization. Appendix A. Supplementary data Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.colsurfb. 2014.11.020. References [1] B.V. Slaughter, S.S. Khurshid, O.Z. Fisher, A. Khademhosseini, N.A. Peppas, Adv. Mater. 21 (2009) 3307. [2] D.E. Ingber, Int. J. Dev. Biol. 50 (2006) 255. [3] Y.S. Pek, A.C. Wan, J.Y. Ying, Biomaterials 31 (2010) 385. [4] C. Yang, M.W. Tibbitt, L. Basta, K.S. Anseth, Nat. Mater. 13 (2014) 645. [5] J.R. Tse, A.J. Engler, PLoS ONE 6 (2011). [6] R.A. Marklein, J.A. Burdick, Soft Matter 6 (2010) 136. [7] D.E. Discher, D.J. Mooney, P.W. Zandstra, Science 324 (2009) 1673. [8] G. Huang, L. Wang, S. Wang, Y. Han, J. Wu, Q. Zhang, F. Xu, T.J. Lu, Biofabrication 4 (2012) 042001.
579
[9] H. Shin, B.D. Olsen, A. Khademhosseini, Biomaterials 33 (2012) 3143. [10] C. Cha, S.R. Shin, X. Gao, N. Annabi, M.R. Dokmeci, X.S. Tang, A. Khademhosseini, Small 10 (2014) 514. [11] J.S. Martinez, A.M. Lehaf, J.B. Schlenoff, T.C. Keller 3rd, Biomacromolecules 14 (2013) 1311. [12] H.J. Kong, D. Kaigler, K. Kim, D.J. Mooney, Biomacromolecules 5 (2004) 1720. [13] A.M. Kloxin, J.A. Benton, K.S. Anseth, Biomaterials 31 (2010) 1. [14] S. Nemir, H.N. Hayenga, J.L. West, Biotechnol. Bioeng. 105 (2010) 636. [15] J.A. Burdick, A. Khademhosseini, R. Langer, Langmuir 20 (2004) 5153. [16] J.K. He, Y.A. Du, J.L. Villa-Uribe, C.M. Hwang, D.C. Li, A. Khademhosseini, Adv. Funct. Mater. 20 (2010) 131. [17] T. Kawano, S. Kidoaki, Biomaterials 32 (2011) 2725. [18] C. Monge, N. Saha, T. Boudou, C. Pozos-Vasquez, V. Dulong, K. Glinel, C. Picart, Adv. Funct. Mater. 23 (2013) 3432. [19] S.R. Shin, B. Aghaei-Ghareh-Bolagh, T.T. Dang, S.N. Topkaya, X.G. Gao, S.Y. Yang, S.M. Jung, J.H. Oh, M.R. Dokmeci, X.W. Tang, A. Khademhosseini, Adv. Mater. 25 (2013) 6385. [20] Y.K. Cheung, E.U. Azeloglu, D.A. Shiovitz, K.D. Costa, D. Seliktar, S.K. Sia, Angew. Chem. Int. Ed. 48 (2009) 7188. [21] S. Khetan, J.A. Burdick, Biomaterials 31 (2010) 8228. [22] R. Sunyer, A.J. Jin, R. Nossal, D.L. Sackett, PLoS ONE 7 (2012). [23] A.M. Kloxin, A.M. Kasko, C.N. Salinas, K.S. Anseth, Science 324 (2009) 59. [24] S. Murayama, M. Kato, Anal. Chem. 82 (2010) 2186. [25] C.A. DeForest, K.S. Anseth, Nat. Chem. 3 (2011) 925. [26] A.M. Kloxin, M.W. Tibbitt, A.M. Kasko, J.A. Fairbairn, K.S. Anseth, Adv. Mater. 22 (2010) 61. [27] F. Yanagawa, S. Sugiura, T. Takagi, K. Sumaru, G. Camci-Unal, A. Patel, A. Khademhosseini, T. Kanamori, Adv. Healthc. Mater. (2014), http://dx.doi.org/ 10.1002/adhm.201400180 (in press). [28] T. Mizutani, H. Haga, K. Kawabata, Cell Motil. Cytoskelet. 59 (2004) 242. [29] M. Radmacher, M. Fritz, C.M. Kacher, J.P. Cleveland, P.K. Hansma, Biophys. J. 70 (1996) 556. [30] K.S. Anseth, C.N. Bowman, L. Brannon-Peppas, Biomaterials 17 (1996) 1647. [31] K.H. Park, M.H. Kim, S.H. Park, H.J. Lee, I.K. Kim, H.M. Chung, Biosci. Biotechnol. Biochem. 68 (2004) 2224. [32] C.N. Salinas, K.S. Anseth, J. Tissue Eng. Regen. Med. 2 (2008) 296.
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Short Communication
Partially photodegradable hybrid hydrogels with elasticity tunable by light irradiation Fumiki Yanagawa a,1 , Takeomi Mizutani b,1 , Shinji Sugiura a,∗,1 , Toshiyuki Takagi a , Kimio Sumaru a , Toshiyuki Kanamori a a Research Center for Stem Cell Engineering, National Institute of Advanced Industrial Science and Technology (AIST), Central 5th, 1-1-1 Higashi, Tsukuba, Ibaraki 305-8565, Japan b Faculty of Advanced Life Science, Hokkaido University, Kita 10 Nishi 8, Kita-ku, Sapporo 060-0810, Japan
a r t i c l e
i n f o
Article history: Received 1 September 2014 Received in revised form 22 October 2014 Accepted 13 November 2014 Available online 4 December 2014 Keywords: Hydrogel Crosslinker Photodegradable Elasticity
a b s t r a c t This paper reports a simple technique to synthesize elasticity tunable hybrid hydrogels using photocleavable (N-hydroxysuccinimide terminated photocleavable tetra-arm poly(ethylene glycol); NHS-PC-4armPEG) and non-photocleavable (N-hydroxysuccinimide terminated tetra-arm poly(ethylene glycol); NHS-4armPEG) activated-ester type crosslinkers. Partially photodegradable hybrid hydrogels were synthesized by reacting the crosslinker mixture with amino-terminated tetra-arm poly(ethylene glycol) (amino-4armPEG). The photocleavable crosslinks are cleaved by irradiating light while the non-photocleavable crosslinks remain intact, resulting in decreased elasticity. We demonstrate that hydrogel elasticity can be controlled by adjusting the ratio of photocleavable NHS-PC-4armPEG and non-photocleavable NHS-4armPEG, and by varying the light exposure energy. We also show how micropatterned elasticity can be obtained in the hydrogels by irradiating with micropatterned light. These techniques could provide a novel platform to tailor the elasticity of hydrogels with microscale precision for biological studies in the near future. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Hydrogels are attractive materials due to their biocompatibility, efficient mass transfer, and ability to encapsulate bioactive molecules and cells [1]. Mechanical properties of hydrogels are essential for cell adhesion, migration, maintenance, and differentiation. Therefore, mechanical properties should be readily customized to satisfy the requirements for particular applications [2]. Cells such as fibroblasts, endothelial cells, smooth muscle cells, and mesenchymal stem cells [3–6] behave differently depending on the mechanical properties of the neighboring extracellular matrix (ECM) [7,8]. It is therefore important to reproduce soft and stiff tissue microenvironments for these cells to exhibit the desired functions. Several research groups have previously reported techniques to control hydrogel stiffness [9,10] and stiffness gradients [11–13]. In previous studies, stiffness gradients were generated using gradient mixers [14], microfluidic gradient generators [15], and
∗ Corresponding author. Tel.: +81 29 861 6286; fax: +81 29 861 6278. E-mail address: [email protected] (S. Sugiura). 1 These authors contributed equally to this work. http://dx.doi.org/10.1016/j.colsurfb.2014.11.020 0927-7765/© 2014 Elsevier B.V. All rights reserved.
diffusion chambers [16]. Photolithography is a useful tool to fabricate microstructures in hydrogels and has been used to construct microscale 3D tissue models [17–19]. To date, several techniques have been proposed to control mechanical properties by light irradiation onto photoresponsive materials [6,20–22]. Especially, photodegradable hydrogels recently emerged as powerful materials for various applications in biotechnology [23,24] due to their ability to fabricate microstructures in hydrogels [25]. The elasticity of these photodegradable hydrogels was tuned and the elasticity gradient was formed by controlling exposure energy and irradiation gradient [13,26]. More recently, we reported an activated-ester-type photocleavable crosslinker, N-hydroxysuccinimide terminated photocleavable tetra-arm poly(ethylene glycol) (NHS-PC-4armPEG), as a convenient material to prepare photodegradable hydrogels. Our NHS-PC-4armPEG forms photodegradable hydrogels by reacting with biocompatible polymers containing amino moieties, such as amino-terminated tetra-arm poly(ethylene glycol) (amino4armPEG) and gelatin [27]. Microstructures can be fabricated in these hydrogels by light irradiation through photomasks. In this study, we demonstrate that the elasticity of hybrid hydrogels prepared with photocleavable NHS-PC-4armPEG and non-photocleavable N-hydroxysuccinimide terminated tetra-arm
576
F. Yanagawa et al. / Colloids and Surfaces B: Biointerfaces 126 (2015) 575–579
Fig. 1. Schematic diagram of the synthesis of the partially photodegradable hybrid hydrogels with photocleavable NHS-PC-4armPEG and non-photocleavable NHS-4armPEG, and the subsequent tuning of elasticity by light irradiation.
poly(ethylene glycol) (NHS-4armPEG) can be controlled by light irradiation, and that geometrical elasticity micropatterns can be formed in the hydrogels by irradiating with micropatterned light. 2. Materials and methods 2.1. Materials All reagents were purchased from Sigma–Aldrich (St. Louis, MO) unless otherwise specified. NHS-PC-4armPEG (Mw = 12,062), whose structural formula is shown in Scheme S1 in the Supporting Information (SI), was synthesized according to the protocol reported in our previous study [27]. NHS-4armPEG (Scheme S2 in the SI, Mw = 10,998) and amino-4armPEG (Scheme S3 in the SI, Mw = 9617) were obtained from NOF Corp. (Tokyo, Japan). 2.2. Hydrogel preparation Fig. 1 illustrates the method used for synthesizing hydrogels using crosslinking amino-terminated polymers. A prepolymer solution including 10 mM amino-4armPEG was prepared in a 1:1 (v/v) mixture with phosphate buffered saline (PBS, Invitrogen Corp., Carlsbad, CA) and 300 mM 4-(2-hydroxyethyl)1-piperazineethanesulfonic acid (HEPES, Wako Pure Chemical Industries, Osaka, Japan) buffer (pH 6.0). A photocleavable crosslinker solution containing 10 mM synthesized NHS-PC4armPEG and a non-photocleavable crosslinker solution containing 10 mM NHS-4armPEG were prepared in a 10 mM phthalate acid buffer (pH 4.0, Wako) with 140 mM NaCl (Wako). ‘Premixedcrosslinker’ solutions were prepared by mixing the photocleavable and non-photocleavable crosslinker solutions at ratios of 1:9–9:1 (v/v), resulting in the ‘Premixed-crosslinker’ solutions containing
10 to 90 mol% of the photocleavable crosslinker. The prepolymer and ‘premixed-crosslinker’ solutions were then mixed at a 1:1 (v/v) ratio. Immediately after mixing, 10–30 L of the mixture was pipetted onto an amino-coated glass slide (MAS coat, Matsunami Glass Corp., Osaka, Japan), and was covered with a cover slip. To control the thickness of the hydrogels, polyethylene terephthalate films (thickness = 25 m) were used as spacers to separate the glass slide and the cover slip. The sample was then placed in a humid incubator for 24 h at 37 ◦ C, allowing the crosslinking reaction between the NHS-activated-ester moieties and the amino moieties to produce hydrogel samples (Fig. 1).
2.3. Light irradiation After the incubation for hydrogel formation, the hydrogel sample was placed topside-down on the sample stage of a light irradiation unit. The sample was then exposed to light (365 nm, 30 mW/cm2 , 10–40 s) from an ultra violet (UV) light source (UVE251S, San-Ei Electric, Osaka, Japan) using a combination of optical filters (long-pass filter = 350 nm, short-pass filter = 385 nm), allowing the photocleavage of o-nitrobenzyl moieties in the hydrogel (Fig. 1). The sample was then immersed in MilliQ water (EMD Milipore, Billerica, USA) for 24 h at 37 ◦ C to erode the photocleaved polymer. For micropatterned light irradiation, a mask-free irradiation system was used, in which a computer controlled microprojection device (DESM-01, Engineering System, Co., Matsumoto, Japan) was attached to an inverted microscope (IX71, Olympus, Tokyo, Japan). A binary micropattern image with arrayed circles (circle diameter: 10 m, pitch: 50 m) was designed using Adobe Illustrator. Following the incubation, the hydrogel sample was immersed in MilliQ water for 24 h at 37 ◦ C, and then placed in a covered Petri
F. Yanagawa et al. / Colloids and Surfaces B: Biointerfaces 126 (2015) 575–579
577
dish on the sample stage to be exposed to micropatterned light through ×10 objective lens (Olympus, numerical aperture = 0.3). After adjusting the focus plane to project a clear image on its surface, the hydrogel sample was exposed to micropatterned light (365 nm, 125 mW/cm2 , 7.2 s) from the computer controlled microprojection device. The sample was then immersed in MilliQ water for 24 h at 37 ◦ C to erode the photocleaved polymer. 2.4. Elasticity measurements A force mapping mode of an atomic force microscopy (AFM) was used to measure the elasticity of the hydrogels with micrometerscale resolution as reported previously [28,29]. We used a commercial AFM equipped with a piezo scanner (maximal scan range: 100 m for the x and y-axes, 10 m for the z-axis; Seiko Instruments Inc., Chiba Japan) and a silicon-nitride cantilever (Olympus; spring constant = 0.02 N/m). After preparing the samples as described in Sections 2.2 and 2.3, excess surface water was removed with a pipette prior to loading the hydrogel on the sample stage. Immediately after sample loading, the pyramidal tip located on the cantilever was indented on the hydrogel surface, and the force acting on the cantilever (here called ‘force curve’) was recorded at room temperature as a function of the displacement of the piezo scanner. Typical force curves for different hydrogels are shown in Fig. S1 in the SI. Force curves were fitted to the Hertzian elastic contact model for a pyramidal indenter giving the local elasticity (Young’s modulus) of the measured points as a fitting parameter. For statistical comparisons of sample elasticity, force curves from more than 1000 different points in the scan area (90 m × 90 m) were analyzed for each sample and frequency distributions of elasticity were plotted (examples are shown in Fig. S2 in the SI). The mean values and standard deviations of elasticity were obtained by fitting the Gaussian distribution to the frequency distribution.
Fig. 2. Comparison of the mean elasticities of the exposed (at 900 mJ/cm2 exposure) and unexposed hydrogels composed of 5 mM amino-4armPEG and 5 mM crosslinker with 10, 30, 50, 70, and 90 mol% of NHS-PC-4armPEG. Error bars denote the standard deviations of over 1000 different measurement points.
3. Results and discussion Partially photodegradable hybrid hydrogels composed of NHS-PC-4armPEG, NHS-4armPEG, and amino-4armPEG were synthesized. The hydrogels were exposed to light to examine its effect on their elasticity. The light exposure conditions (365 nm, 30 mW/cm2 , 30 s) were chosen based on our previous study on the photocleavage behavior of o-nitrobenzyl moiety [27]. The hydrogels became yellowish after light exposure owing to the photocleavage of o-nitrobenzyl moiety. Fig. 2 shows the mean elasticities of exposed and unexposed hydrogels of different compositions. The mean elasticity of the unexposed hydrogels decreased with increasing proportions of photocleavable NHS-PC4armPEG. This change was probably due to the different reactivity of the NHS-activated-ester moiety in NHS-PC-4armPEG and NHS4armPEG to the amino moiety on amino-4armPEG (details of each molecular formula are available in Schemes S1–S3 in the SI). In the case of 30 and 50 mol% photocleavable NHS-PC4armPEG content, a significant decrease in elasticity was observed in the exposed hydrogels while their shape was retained. As expected, higher concentrations of NHS-PC-4armPEG led to a greater decrease in elasticity after exposure. When the photocleavable NHS-PC-4armPEG content was greater than 70 mol%, the hydrogels degraded completely and lost their shape. According to our previous study on photodegradable hydrogels, the degradation depth is determined by the exposure energy [27]. To assess whether light irradiation can be used to control hydrogel elasticity, this was investigated for hydrogels prepared with a crosslinker containing 50 mol% of NHS-PC-4armPEG and amino4armPEG. As shown in Fig. 3, the elasticity decreased monotonically
Fig. 3. Effect of light exposure energy on the mean elasticities of the hydrogels composed of 5 mM amino-4armPEG and 5 mM crosslinker with 50 mol% of NHSPC-4armPEG. Error bars denote the standard deviations of over 1000 different measurement points.
for exposures up to 300 mJ/cm2 , beyond which the elasticity varied little. Under these conditions, most of the o-nitrobenzyl moieties in the hydrogels were cleaved such that the hydrogel structure was probably maintained by the remaining polymer network composed of amino-4armPEG and non-photocleavable NHS-4armPEG. Hence, for sufficiently high exposures, the elasticity of the exposed hydrogels was determined by the composition of the crosslinker. Based on rubber elasticity theory, the elasticity of hydrogels is highly dependent on the effective number of crosslinking points [30]. Our results thereby suggest that light irradiation successfully decreased the crosslinking density in the hydrogel by photocleaving the o-nitrobenzyl moieties in photocleavable NHS-PC-4armPEG while leaving the non-photocleavable crosslinks formed by NHS4armPEG intact. As a proof of concept, we demonstrated micropatterning of elasticity in the partially photodegradable hybrid hydrogels. The hydrogels prepared with a crosslinker containing 50 mol% NHS-PC-4armPEG were exposed to micropatterned light (365 nm, 125 mW/cm2 , 7.2 s) using a mask-free irradiation system, through which an array of circles (resolution: 1.4 m, circles diameter: 10 m, pitch: 50 m) was projected and formed on the hydrogel samples. Fig. 4 shows the elasticity mapping images of the exposed hydrogel. The elasticity pattern on the hydrogel reflects
578
F. Yanagawa et al. / Colloids and Surfaces B: Biointerfaces 126 (2015) 575–579
Fig. 4. (a) The array of circles (diameter: 10 m, pitch: 50 m) used as a binary micropattern image. (b) The elasticity micropattern formed in the hydrogels by mask-free irradiation with micropatterned light. A hydrogel composed of 5 mM amino-4armPEG and 5 mM crosslinker with 50 mol% NHS-PC-4armPEG was exposed to 125 mW/cm2 micropatterned light at a wavelength of 365 nm for 7.2 s.
the micropatterning image, with the elasticity in the exposed regions decreasing as a result of decreased crosslinking density caused by photocleavage of the o-nitrobenzyl moieties. Fig. 4 also shows that the elasticity in the unexposed regions was not affected, demonstrating that irradiation with micropatterned light alters the hydrogel elasticity with microscale precision. The micropatterned elasticity shown in Fig. 4 exhibited different elasticity at different locations; the elasticity in the two circle areas in the right side was lower than that in the circles in the left side. We think that at least the following three factors are possibly affecting micropatterning elasticity. Firstly, focusing is very important to project a clear image on the sample surface using our microprojection device, while precise focusing on the transparent hydrogel sample is generally difficult. Secondly, it is also difficult to prepare flat samples with micrometer scale precision by using polyethylene terephthalate film spacers. Finally, the swelling of the hydrogel also alters the thickness of the hydrogel and makes it difficult to adjust the focus precisely. Therefore, inevitable defocus might cause the inhomogeneous illumination. Especially, if the sample was not flat, a clear image is projected only on the focused plane and a defocused image is projected at different z-positions. These effects potentially cause defocused projection on the sample surface at different locations. The elasticity of the partially photodegradable hybrid hydrogels presented in this work can be tuned by varying their composition while microscale geometrical elasticity patterns can be designed using micropatterned light irradiation. In the current study, we prepared PEG-based hydrogels composed of NHS-PC-4armPEG, NHS-4armPEG, and amino-4armPEG. Since PEG itself is highly inert and lacks biolological functionalities, few cells adhered onto the PEG-based partially photodegradable hybrid hydrogels as expected (Fig. S3 in the SI). No effect of light irradiation was observed on cell adhesion on these hydrogels. In order to enhance cell adhesion on the PEG-based partially photodegradable hybrid hydrogels, introduction of RGD (arginine-glycine-aspartic acid) peptide into the hydrogel is probably effective as reported for other PEG-based hydrogels [31,32]. There have been a couple of reports to control hydrogel elasticity by light irradiation onto the photodegradable hydrogels [13,26]. Compared to these studies, our approach to use the partially photodegradable hybrid hydrogel enables us to define the initial and resulting elasticity without precise control of exposure energy. Furthermore, our technique is theoretically applicable to synthesize biomaterial-based elasticity-tunable hybrid hydrogels because
the activated-ester-type crosslinker can react with a variety of polymers containing primary amine moieties [27]. As an example, we previously reported that NHS-PC-4armPEG can react with gelatin to generate a photodegradable hydrogel. Since both the photocleavable and non-photocleavable crosslinkers used in this study can react with amino-moieties in biocompatible polymers, the technique may be used to investigate cell behavior on hydrogels with tailored stiffness patterns composed of gelatin, collagen, fibronectin and chitosan. We expect that adherent cells adhere onto these biomaterial-based, partially photodegradable hybrid hydrogels. Photocleavage reaction takes place immediately upon exposure to light. Although the erosion of the polymer and the swelling of the hydrogel takes some time, the elasticity of our partially photodegradable hybrid hydrogels can be controlled on-demand. They could therefore potentially be applied to the dynamic control of elasticity in long-term cell culture processes, as reported in a previous study with a similar approach [6]. As described above, hydrogels with geometrically patterned elasticity and dynamically tunable elasticity could find novel applications in tissue engineering by mimicking soft and stiff tissue microenvironments in vivo. We therefore envision that in the near future, this simple technique could find many applications in the fields of cell manipulation and tissue engineering. 4. Conclusion We have presented a simple technique to synthesize elasticity tunable hybrid hydrogels with photocleavable NHS-PC-4armPEG and non-photocleavable NHS-4armPEG. The approach presented in this study provides hydrogels whose elasticity can be tailored by varying their composition, and by exposing them to light. We also demonstrated elasticity micropatterning by irradiating the hydrogels with micropatterned light. This technique could provide a novel platform for the precise control of hydrogel elasticity that should prove useful for biological studies in the near future. Author contributions These authors contributed equally. FY, TM, and SS conceived and planned the research, and wrote the paper. TT synthesized the materials. FY, TM, and SS designed the experimental protocols. FY prepared hydrogel samples and TM performed AFM measurements.
F. Yanagawa et al. / Colloids and Surfaces B: Biointerfaces 126 (2015) 575–579
FY, TM and SS analyzed the data. KS and TK supervised the research. All authors revised the manuscript and agreed on its final contents. Funding sources This research was supported by KAKENHI (24106512 and 24106502). Acknowledgments We thank Dr. T. Sato and Ms. K. Morishita for technical help and helpful comments on rheological characterization. Appendix A. Supplementary data Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.colsurfb. 2014.11.020. References [1] B.V. Slaughter, S.S. Khurshid, O.Z. Fisher, A. Khademhosseini, N.A. Peppas, Adv. Mater. 21 (2009) 3307. [2] D.E. Ingber, Int. J. Dev. Biol. 50 (2006) 255. [3] Y.S. Pek, A.C. Wan, J.Y. Ying, Biomaterials 31 (2010) 385. [4] C. Yang, M.W. Tibbitt, L. Basta, K.S. Anseth, Nat. Mater. 13 (2014) 645. [5] J.R. Tse, A.J. Engler, PLoS ONE 6 (2011). [6] R.A. Marklein, J.A. Burdick, Soft Matter 6 (2010) 136. [7] D.E. Discher, D.J. Mooney, P.W. Zandstra, Science 324 (2009) 1673. [8] G. Huang, L. Wang, S. Wang, Y. Han, J. Wu, Q. Zhang, F. Xu, T.J. Lu, Biofabrication 4 (2012) 042001.
579
[9] H. Shin, B.D. Olsen, A. Khademhosseini, Biomaterials 33 (2012) 3143. [10] C. Cha, S.R. Shin, X. Gao, N. Annabi, M.R. Dokmeci, X.S. Tang, A. Khademhosseini, Small 10 (2014) 514. [11] J.S. Martinez, A.M. Lehaf, J.B. Schlenoff, T.C. Keller 3rd, Biomacromolecules 14 (2013) 1311. [12] H.J. Kong, D. Kaigler, K. Kim, D.J. Mooney, Biomacromolecules 5 (2004) 1720. [13] A.M. Kloxin, J.A. Benton, K.S. Anseth, Biomaterials 31 (2010) 1. [14] S. Nemir, H.N. Hayenga, J.L. West, Biotechnol. Bioeng. 105 (2010) 636. [15] J.A. Burdick, A. Khademhosseini, R. Langer, Langmuir 20 (2004) 5153. [16] J.K. He, Y.A. Du, J.L. Villa-Uribe, C.M. Hwang, D.C. Li, A. Khademhosseini, Adv. Funct. Mater. 20 (2010) 131. [17] T. Kawano, S. Kidoaki, Biomaterials 32 (2011) 2725. [18] C. Monge, N. Saha, T. Boudou, C. Pozos-Vasquez, V. Dulong, K. Glinel, C. Picart, Adv. Funct. Mater. 23 (2013) 3432. [19] S.R. Shin, B. Aghaei-Ghareh-Bolagh, T.T. Dang, S.N. Topkaya, X.G. Gao, S.Y. Yang, S.M. Jung, J.H. Oh, M.R. Dokmeci, X.W. Tang, A. Khademhosseini, Adv. Mater. 25 (2013) 6385. [20] Y.K. Cheung, E.U. Azeloglu, D.A. Shiovitz, K.D. Costa, D. Seliktar, S.K. Sia, Angew. Chem. Int. Ed. 48 (2009) 7188. [21] S. Khetan, J.A. Burdick, Biomaterials 31 (2010) 8228. [22] R. Sunyer, A.J. Jin, R. Nossal, D.L. Sackett, PLoS ONE 7 (2012). [23] A.M. Kloxin, A.M. Kasko, C.N. Salinas, K.S. Anseth, Science 324 (2009) 59. [24] S. Murayama, M. Kato, Anal. Chem. 82 (2010) 2186. [25] C.A. DeForest, K.S. Anseth, Nat. Chem. 3 (2011) 925. [26] A.M. Kloxin, M.W. Tibbitt, A.M. Kasko, J.A. Fairbairn, K.S. Anseth, Adv. Mater. 22 (2010) 61. [27] F. Yanagawa, S. Sugiura, T. Takagi, K. Sumaru, G. Camci-Unal, A. Patel, A. Khademhosseini, T. Kanamori, Adv. Healthc. Mater. (2014), http://dx.doi.org/ 10.1002/adhm.201400180 (in press). [28] T. Mizutani, H. Haga, K. Kawabata, Cell Motil. Cytoskelet. 59 (2004) 242. [29] M. Radmacher, M. Fritz, C.M. Kacher, J.P. Cleveland, P.K. Hansma, Biophys. J. 70 (1996) 556. [30] K.S. Anseth, C.N. Bowman, L. Brannon-Peppas, Biomaterials 17 (1996) 1647. [31] K.H. Park, M.H. Kim, S.H. Park, H.J. Lee, I.K. Kim, H.M. Chung, Biosci. Biotechnol. Biochem. 68 (2004) 2224. [32] C.N. Salinas, K.S. Anseth, J. Tissue Eng. Regen. Med. 2 (2008) 296.
