Fabrication of cell-benign inverse opal hydrogels for three-dimensional cell culture

Accepted Manuscript Fabrication of Cell-Benign Inverse Opal Hydrogels for Three-Dimensional Cell Culture Pilseon Im, Dong Hwan Ji, Min Kyung Kim, Jaeyun Kim PII: DOI: Reference:

S0021-9797(17)30136-4 http://dx.doi.org/10.1016/j.jcis.2017.01.108 YJCIS 22014

To appear in:

Journal of Colloid and Interface Science

Received Date: Revised Date: Accepted Date:

12 October 2016 26 January 2017 27 January 2017

Please cite this article as: P. Im, D.H. Ji, M.K. Kim, J. Kim, Fabrication of Cell-Benign Inverse Opal Hydrogels for Three-Dimensional Cell Culture, Journal of Colloid and Interface Science (2017), doi: http://dx.doi.org/10.1016/ j.jcis.2017.01.108

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Fabrication of Cell-Benign Inverse Opal Hydrogels for Three-Dimensional Cell Culture

Pilseon Im,1 Dong Hwan Ji,1 Min Kyung Kim,1 and Jaeyun Kim1,2*

1

School of Chemical Engineering, Sungkyunkwan University (SKKU), Suwon 16419, Republic

of Korea 2

Samsung Advanced Institute for Health Science & Technology (SAIHST), Sungkyunkwan

University (SKKU), Suwon 16419, Republic of Korea. * To whom correspondence should be addressed. Jaeyun Kim, Ph. D. Associate Professor School of Chemical Engineering, Samsung Advanced Institute for Health Science & Technology (SAIHST), Sungkyunkwan University (SKKU), Suwon 16419, Republic of Korea Phone: +82-31-290-7252 Fax: +82-31-290-7272 E-mail: [email protected]

1

Abstract Inverse opal hydrogels (IOHs) for cell culture were fabricated and optimized using calciumcrosslinked alginate microbeads as sacrificial template and gelatin as a matrix. In contrast to traditional three-dimensional (3D) scaffolds, the gelatin IOHs allowed the utilization of both the macropore surface and inner matrix for cell co-culture. In order to remove templates efficiently for the construction of 3D interconnected macropores and to maintain high cell viability during the template removal process using EDTA solution, various factors in fabrication, including alginate viscosity, alginate concentration, alginate microbeads size, crosslinking calcium concentration, and gelatin network density were investigated. Low viscosity alginate, lower crosslinking calcium ion concentration, and lower concentration of alginate and gelatin were found to obtain high viability of cells encapsulated in the gelatin matrix after removal of the alginate template by EDTA treatment by allowing rapid dissociation and diffusion of alginate polymers. Based on the optimized fabrication conditions, gelatin IOHs showed good potential as a cell co-culture system, applicable to tissue engineering and cancer research. Keywords Gelatin, Inverse Opal Hydrogel, Co-culture, Macroporous Scaffold, Alginate

2

1. Introduction The cells grown in three-dimensional (3D) microenvironments more closely mimic natural tissues and organs than cells grown in traditional two-dimensional (2D) plates. Therefore, generating artificial 3D cellular microenvironments is important for in vitro cell culture and its applications to a variety of biomedical studies, including drug screening, stem cell research, and tissue engineering1-6. Generally, high porosity with uniform macropores and homogeneously interconnected architectures are required for enhanced adhesion and migration of cells within the scaffold, and effective mass exchange of oxygen, nutrients, and metabolite wastes for proliferation of cells7,8. To fulfill these requirements, several strategies have been employed to generate 3D microenvironments based on biomaterials through gas foaming, salt leaching, freeze drying, and electrospinning9-14. However, most of these technologies have limitations related to uniform pore structure, size, and shape as well as poor connectivity15,16. All previous methods have difficulties in the generation of uniform macropores with precise control of pore size and shape due to the uncontrolled morphology of porogens, such as gas bubbles, salt crystals, and ice crystals. In addition, due to use of high pressure of gases, organic solvents, and freeze-drying processes, cells cannot be encapsulated within the matrix of macroporous biomaterials prepared in conventional methods. In order to overcome the challenges associated with uniform pore structure and interconnectivity, inverse opal structures based on sacrificial templates have been exploited to modulate 3D architecture with uniform pores17-19. The reported inverse opal structures for cell culture can be categorized into two types depending on the physical state of the composing matrix. The first type is the typical inverse opal scaffold composed of a dense matrix, such as silica, polyacrylamide, chitosan, poly(ethylene glycol), or poly(lactic-co-glycolic acid), prepared

3

by removing the sacrificial colloidal templates, such as silica, polystyrene, and poly(methyl methacrylate), using organic solvent or highly acidic solution20-28. These inverse opal scaffolds provide only the surface of macroporous structures for cell adhesion and cannot encapsulate the cells inside the matrix because of the use of toxic solvents to remove the opal templates. The second type of inverse opal structure recently reported is composed of hydrogels (e.g., gelatin), which can encapsulate the cells29. In contrast to the previous approaches for fabrication of inverse opal scaffolds, the second type of structure utilizes calcium-crosslinked alginate microbeads as a sacrificial opal template that can be removed using the calciumchelating agent ethylenediaminetetraacetic acid (EDTA), and gelatin is employed as a component of the scaffold matrix. In this structure, both the macropore surface and the inner matrix can be utilized for cell culture; this may allow the structures to mimic natural tissues and organs or facilitate investigation of biological processes, such as paracrine signaling in stem cell niches or cancer-stroma interactions. However, fabrication of this type of scaffold has not yet been optimized to achieve efficient removal of templates for the construction of 3D interconnected macropores and to maintain high cell viability during template removal. Longer treatment in EDTA solution is beneficial for complete removal of alginate microbeads but can be harmful to the cells encapsulated in the gelatin matrix. In this report, various factors, including alginate viscosity, alginate concentration, alginate size, calcium concentration for preparation of alginate microbeads, EDTA treatment time, and gelatin concentration, were controlled to achieve high cell viability and successful generation of a 3D inverse opal structure in order to establish a 3D co-culture system (Figure 1). Low viscosity alginate, lower crosslinking calcium ion concentration, and lower concentration of alginate and gelatin were found to obtain high viability of cells encapsulated in the gelatin matrix

4

after removal of the alginate template by EDTA treatment by allowing rapid dissociation and diffusion of alginate polymers. Based on the optimized fabrication conditions, gelatin IOHs showed good potential as a cell co-culture system, applicable to tissue engineering and cancer research.

Figure 1. Schematic presentation of gelatin inverse opal hydrogels prepared via cell-benign template removal method using alginate microbeads as template.

2. Experimental Section Materials. Gelatin (type A, ~300 g bloom from porcine skin), methacrylic anhydride (MA), 2hydroxy-4′-(2-hydroxyethoxy)-2-methylpropiophenone (I2959), alginic acid sodium salt from brown algae (medium viscosity alginate, MVA and low viscosity alginate, LVA), calcium chloride dehydrate (≥ 99%), ethylenediaminetetraacetic acid (EDTA), Dulbecco’s modified eagle’s medium (DMEM), and alpha minimum essential medium (Alpha MEM) were purchased from Sigma (St. Louis, MO, USA) and used without further purification. Live/Dead assay kit was purchased from Molecular probe (Eugene, OR, USA). OxiSelectTM cellular UV-induced DNA damage staining kit was purchased from Cell biolabs (San Diego, CA, USA). Synthesis of gelatin methacrylamide (GelMA). GelMA was synthesized as previously described30. First, 10 wt% gelatin solution was prepared by dissolving 1 g of gelatin powder in

5

Dulbecco’s phosphate-buffed saline (DPBS, pH 7.4) at 50–60°C. One milliliter of MA was added to the gelatin solution at a rate of 0.25 mL/min, and the resulting solution reacted for 1 h under vigorous stirring. Subsequently, the reaction mixture was diluted with distilled water to terminate the reaction, precipitated twice in a large excess of ethanol at 4°C, and dialyzed in distilled water using dialysis tubing with a 12–14-kDa cutoff for 5–7 days at 40°C for purification. After dialysis, the resulting GelMA was freeze-dried for 1 week and stored at -80°C until use. Fabrication of gelatin hydrogels. 10 wt% GelMA solution with 0.3 wt% I2959 was poured into custom-designed acryl mold (d: 15 mm, h: 5 mm) and photo-cured by UV irradiation for 10 min. The resulting gelatin hydrogels were separated from the mold and washed with alpha MEM medium at 37°C for 20 min to remove the photoinitiator and by-products. To seed cells on the surface of gelatin hydrogel, MC3T3 cells (ATCC, VA, USA) were seeded on surface of gelatin hydrogel at a concentration of 1.0 × 106 cells/mL, and incubated at 37°C. Preparation of calcium-crosslinked alginate microbeads. Alginate microbeads were prepared using the electrostatic droplet extrusion method, as previously described31. Alginate solutions with different concentrations (1.5–4 wt%, depending on alginate type) were prepared by dissolving sodium alginate into distilled water and slowly rotating for 12 h. The sodium alginate solution was dropped into a 100-mM calcium dichloride solution using a syringe pump (Pump 11 Series, Harvard Apparatus) through a blunt needle (24G) under 12 kV using a high voltage power supply (230-30R, Bertan). According to the electrostatic potential difference between the needle and the calcium solution, small alginate drops were produced and dropped into the calcium solution. The alginate droplets were immediately crosslinked with calcium ions to form alginate microbeads. The resulting alginate microbeads were sterilized by immersion in 70 wt%

6

ethanol for 1 h, washing with PBS five times, and dispersion in sterile PBS until use. The size of calcium-crosslinked alginate microbeads was measured with optical microscope (Primo Vert, Carl Zeiss). To evaluate Young’s modulus of alginate hydrogels prepared based on LVA in calcium solution with different concentration, cylindrical alginate hydrogels were prepared. First, 3 wt% LVA was poured into a mold (diameter 10 mm, height 5 mm) and carefully incubated at 50, 100, 200, or 500 mM calcium solution for 3 h. The resulting calcium-crosslinked, cylindrical alginate hydrogels were retrieved from the mold and washed with PBS several times. Compressive test of these samples was performed in air condition at room temperature using 5kgf load cell at a rate of 1 mm/min (Cometech, QC-508E). Fabrication of inverse opal hydrogels (IOHs). Uniform alginate microbeads were packed in the disc-shaped acryl mold (d: 5 mm, h: 3 mm). The hydrogel precursor solution composed of 10 wt% GelMA and 0.3 wt% photoinitiator (I2959) in PBS was infiltrated into the interstitial space of pre-packed alginate microbeads in the mold. Then, the GelMA was photo-crosslinked under UV (365 nm) irradiation for 10 min. The resulting alginate microbeads/gelatin composites were retrieved from the mold and immersed in 100 mM EDTA solution on a shaker for 2 h at 37°C in order to remove the alginate microbeads. The resulting gelatin inverse opal hydrogels were washed with PBS three times to remove excess EDTA, photo-initiator and other by-products. The gelatin IOHs were kept in PBS until use. Fabrication of IOHs encapsulating cells. Sterilized alginate microbeads were packed in a mold (d: 5 mm, h: 3 mm). HCT116 cells were mixed in 10 wt% GelMA solution with 0.3 wt% I2959 at a concentration of 1.0 × 106 cells/mL. The resulting GelMA solution containing HCT116 cells was infiltrated into the interparticle spaces of the close-packed alginate microbeads to fill interparticle space between alginate microbeads and cross-linked under UV (365 nm) irradiation

7

for 10 min. The resulting alginate microbead/gelatin composite hydrogels were separated from the mold and incubated in complete DMEM at 37°C for 20 min to remove photoinitiator and byproducts. The composite hydrogels were then immersed in 100 mM EDTA (pH 7.4) for different periods under shaking to remove the alginate microbeads. The resulting IOHs after complete removal of alginate microbeads were washed with complete DMEM three times and incubated at 37°C in 6-well plates. Co-culture of different cell types in IOHs. Using the protocol described above, sterilized alginate microbeads were prepared and packed in a mold. Mouse MSCs (D1-mCherry) were mixed in 10 wt% GelMA solution at a concentration of 6.9 × 106 cells/mL. GelMA solution with 0.3 wt% I2959 containing D1-mCherry was infiltrated into the interparticle spaces of the closepacked alginate beads in the mold (d: 5 mm, h: 3mm) and crosslinked under UV (365 nm) irradiation for 10 min. The resulting alginate microbead/gelatin composite hydrogels were separated from the mold and incubated in complete DMEM at 37°C for 20 min to remove photoinitiator and by-products. To remove the alginate microbeads, composite hydrogels were immersed in 100 mM EDTA (pH 7.4) for different times under shaking. The resulting IOHs after complete removal of alginate microbeads were washed with complete DMEM three times and incubated at 37°C in 6-well plates. Other types of cells were additionally seeded on the surface of macropores in IOHs. Specifically, C166-GFP mouse endothelial cells in DMEM (9.6 × 106 cells/mL) were dropped on the top of IOHs encapsulating D1-mCherry cells and dispersed by pipetting to help the cells infiltrate into the macropores of IOHs. Cell seeding and culture on the surface of gelatin IOHs without encapsulation of cells was also conducted using the same method as that used for empty gelatin IOHs. Next, 5 mL DMEM was carefully added to wells containing the IOHs, and the IOHs were incubated at 37°C. After culturing for 1, 4, or 7 days,

8

cells in IOHs were imaged on a fluorescence microscope (Ti-u, Nikon). To encapsulate two different types of cells in IOHs, GFP-expressing C166 and mCherry-expressing D1 cells (5.0 × 106 cells/mL each) were mixed in GelMA solution, and IOHs were fabricated according to the procedure described above. Cell viability and toxicity test. The viability of encapsulated cells in gelatin hydrogel or gelatin IOHs in 100 mM EDTA solution was analyzed over time through cell counting. Cell viability was calculated by the ratio of live cells to all cells using live/dead assay (Molecular probe, Eugene, OR, USA). To investigate optimal EDTA treatment time for cell viability, gelatin bulk hydrogels encapsulating HCT116 cells were incubated with 100 mM EDTA solution for different durations of 30, 60, 120, and 300 min. Then, gelatin hydrogels encapsulating cells were washed with complete medium (DMEM) three times to remove remaining EDTA. Next, gelatin hydrogels encapsulating cells was incubated with a mixture solution of 2 μM calcein AM and 4μM EthD-1 for 10 min. Then the hydrogels was washed with complete medium three times and the cells encapsulated in IOHs were imaged on a fluorescence microscope (Ti-u, Nikon). The same method was used to assess long-term cell viability for 1, 4, 7, and 14 days. UV-induced cell damage were evaluated using OxiSelectTM cellular UV-induced DNA damage staining kit (Cell biolabs, San Diego, CA, USA) that detects cyclobutane pyrimidine dimers (CPD) produced in DNA after UV irradiation via immunofluorescent assay. Briefly, HCT116 cells were seeded in 96 well plate overnight and exposed under UV-irradiation at 365 nm for 10 min. Then, for fixation and denaturation, 100μL of MtOH/Aceton at 4:1 ratio was added into cells for 30 min, and next, 100 μL of 70 % EtOH was added for 30 min. After washing above solution by DPBS, denaturing solution also was added into cells, step by step. For staining, anti-CPD antibody and FITC conjugated antibody was added, for 1 h respectively. After

9

staining, cells were washed by DPBS 4 times and imaged on fluorescence microscope using FITC filter (Ti-u, Nikon). Statistical analysis. All numerical data were presented as mean ± standard deviation (SD). Statistical comparisons were implemented using a one-way ANOVA with Scheffe`s post hoc test. Differences at a p-value (<0.05) were considered to be reliable. All error bars were expressed as standard deviations.

3. Results and Discussion The key concept in the fabrication of cell-friendly, macroporous IOHs is the use of Ca2+crosslinked alginate hydrogel microbeads as sacrificial templates that can be dissolved by EDTA, a calcium chelator. Although a mixture of EDTA and trypsin is widely used to detach cells from culture plates or to prepare discrete cell suspensions, longer EDTA treatment could induce cell cytotoxicity. First, we evaluated the viability of cells encapsulated in bare gelatin hydrogels after EDTA treatment for different times to determine the maximum EDTA treatment time that could be used while maintaining high cell viability. HCT116 human colorectal carcinoma cells were encapsulated in 10 wt% gelatin hydrogels by photopolymerization of GelMA. The cellcontaining gelatin hydrogels, a disc shape measuring 5 mm in height and 15 mm in diameter, were then incubated in 100 mM EDTA at 37°C for 30, 60, 120, or 300 min, respectively, and the viability of HCT116 cells was measured based on fluorescent live-dead assay on days 1, 4, 7, and 14 (Figure 2). There were no significant differences in cell viability after EDTA treatment for up to 120 min compared with that of the control group (no EDTA treatment) within 14 days. In contrast, EDTA treatment for 300 min resulted in significantly decreased cell viability over time compared to other conditions. These results indicated that EDTA treatment for longer than

10

300 min could be toxic to the encapsulated cells. The gelatin/alginate composite hydrogels could be treated with 100 mM EDTA solution for up to 2 h to remove alginate beads, allowing fabrication of gelatin IOHs encapsulating live cells in gelatin matrix. Because alginate and gelatin are the only materials used in IOH fabrication, we investigated several parameters related to the alginate microbeads and gelatin hydrogel in order to achieve complete removal of alginate microbeads within a maximum of 2 h of EDTA treatment, which was the longest time that could be used to maintain high cell viability. The basic characteristics of alginate microbeads that allow their use as opal templates are their mechanical properties, which allow proper formation of an opal structure, and their dissolution speed in EDTA solution. For optimization of alginate microbeads, the following factors should be considered: molecular weight of alginate, concentration of alginate solution, and concentration of calcium ions for ionic crosslinking of alginate. First, we investigated the concentration of crosslinking calcium ions on alginate bead removal from the alginate/gelatin composite hydrogels. Alginate microbeads were prepared using 3 wt% LVA in 50, 100, 200, or 500 mM calcium chloride solution. The alginate microbeads prepared in 50 mM calcium solution did not have mechanical property required to generate the opal structure. In contrast, calcium concentrations above 100 mM yielded microbeads with appropriate mechanical properties for use as IOH templates. As the small size and spherical shape of microbeads was inappropriate to measure Young’s modulus in compressive mechanical test, instead we prepared cylindrical alginate hydrogels in 50, 100, 200, or 500 mM calcium chloride solution using a mold (d: 10 mm, h: 5 mm). The corresponding Young’s moduli of the alginate hydrogel obtained from compressive test were 33.3, 75.3, 116, and 141 kPa, respectively. Assuming there is no significant difference of modulus depending on hydrogel size, the modulus over 75.3 kPa would

11

be the minimum modulus of alginate microbeads to prepare stable 3D microbeads packing while maintaining interparticle space between alginate microbeads for infiltration of the second hydrogel precursor.

Figure 2. Effects of EDTA treatment time on viability of cells encapsulated in gelatin hydrogels. (a) Cell viability based on live/dead staining of HCT116 cells encapsulated in 10wt% gelatin hydrogels after 100 mM EDTA treatment for different times (normalized to the cell viability in the control on day 1). (b) The representative fluorescent images of HCT116 cells after live/dead staining on Day 1 and 14, respectively. Scale bar, 100μm.

The alginate beads were then used to prepare composite hydrogels to test alginate dissolution. The modulus of composite gelatin/alginate hydrogels prepared using alginate microbeads obtained from 100, 200, and 500 mM calcium solution were 34.3, 38.8, and 40.8 kPa, respectively. The composite hydrogels composed of alginate microbeads and gelatin hydrogel filled in interparticle space between close-packed alginate microbeads were treated with 100 mM

12

EDTA solution to remove Ca-crosslinked alginate microbeads and the hydrogels were observed over time on optical microscope to check the removal of alginate microbeads from the composite hydrogels (Figure 3). For all samples, the alginate microbeads were gradually removed from outside to inside of the composite hydrogels because of diffusion of Ca-chelating EDTA from the external solution. The results clearly showed that longer EDTA treatment times were

Figure 3. Optimization of fabrication conditions for gelatin inverse opal hydrogels using alginate microbeads as sacrificial templates. Effects of calcium ion concentrations used in the preparation of alginate microbeads (scale bar, 3 mm).

required when higher calcium concentrations were used in microbead formation. Specifically, alginate microbeads crosslinked in 100 or 200 mM calcium solution could be removed from the composite hydrogel within 90 min of incubation in 100 mM EDTA solution. In contrast, microbeads formed in 500 mM calcium solution required at least 4 h of EDTA treatment, which did not guarantee cell viability as shown in Figure 2. These results suggested that a suitable degree of ionic crosslinking of alginate microbeads was required for the fabrication of cell-

13

friendly IOHs. Based on this result, we used 100 mM calcium solution for the formation of ionically crosslinked alginate microbeads in the following experiments. We also examined different sizes of alginate microbeads in the fabrication of IOHs. LVA microbeads with different sizes (360, 730, and 1150 μm, Figure S1) were used to form alginate/gelatin composite hydrogels and subsequently treated with 100 mM EDTA solution (Figure 4). Longer treatment times were required for the smaller alginate microbeads, as expected. The tested alginate microbeads were all completely removed within 2 h, demonstrating that these microbeads were suitable for the encapsulation of cells in IOHs. The macropore sizes of the IOHs prepared using 360, 730, and 1150 μm alginate microbeads were 315, 710, and 1060 μm, respectively, representing the successful removal of template alginate microbeads in IOHs. The thereotical macroporosity based on the close-packed alginate microbeads template was around 74%.

Figure 4. Effects of alginate microbead size on the fabrication of IOHs. a) Representative photographs of alginate/gelatin composite hydrogels with various sizes (360, 730, and 1150 μm) in 100 mM EDTA solution over time (scale bar, 1 mm). b) Average time required to fully dissolve alginate microbeads (n=4). Error bars, mean ± s.d. *p < 0.05. The experiment was repeated 4 times.

14

The other parameter of alginate microbeads related to EDTA dissolution is the molecular weight and concentration of alginate solution. Alginate formulations with two different molecular weights were used to fabricate the microbeads. Uniform 1800-μm alginate microbeads were prepared using 1.5 wt% medium viscosity alginate (MVA), 3 wt% low viscosity alginate (LVA), and 4 wt% LVA. After the alginate beads were closely packed in a disc-shaped acryl mold, photocurable GelMA solution was infiltrated into the spaces between the microbeads. UV irradiation for 15 min resulted in opaque alginate microbead/gelatin hydrogel composites (Figure 5a). The composite hydrogels were then immersed in 100 mM EDTA solution, and the removal of alginate microbead templates was observed under an optical microscope over time. Interestingly, significantly different EDTA treatment times were required for the complete removal of the microbeads. It was impossible to completely remove the alginate microbeads in the 1.5 wt% MVA sample. In contrast, both 3 wt% and 4 wt% LVA microbeads were completely removed, yielding interconnected 3D macroporous IOHs within 2 h. The EDTA treatment time could be shortened to 1 h when the 3 wt% LVA microbeads were used. Because the viscosity of alginate is directly related to the molecular weight of alginate, LVA has a lower molecular weight than MVA. Accordingly, LVA also has a shorter chain length than MVA, and LVA microbeads could be easily dissociated into linear alginate chains. In addition, the dissolved shorter alginate chains had higher mass transport capacity, allowing them to be more easily released through the macropores of the IOHs compared with MVA microbeads. In contrast, the longer linear chains of alginate in MVA microbeads limited diffusion from the gelatin matrix due to insufficient mobility and a higher degree of ionic crosslinking. Following optimization of the fabrication conditions for the alginate microbeads as the template, the parameters of the gelatin matrix were investigated in order to further optimize the

15

fabrication of gelatin IOHs. The concentration of GelMA precursor is an important factor as it is related to the density of the hydrogel network, the resulting diffusion properties of EDTA, and the dissolved alginate polymer during template removal. To evaluate this in our system, three concentrations of GelMA (7, 10, and 15 wt%) were prepared with LVA microbeads formed in 100 mM calcium solution. The treatment of the alginate/gelatin composite hydrogels in 100 mM EDTA solution resulted in complete removal of alginate microbeads after 60, 90, and 210 min for the 7, 10, and 15 wt% gelatin samples, respectively (Figure 5b). These results indicated that the GelMA solution with 7–10 wt% concentration was optimal for fabrication of IOHs within 2 h. Additionally, we examined the effects of the concentration of initiator in GelMA solution that may be related to the length of crosslinked GelMA. Our results showed that the concentration of initiator had only a minor effect on the dissolution time of alginate microbeads from the composite hydrogels.

Figure 5. Optimization of fabrication conditions for gelatin inverse opal hydrogels using alginate microbeads as sacrificial templates. Effects of a) alginate viscosity and b) gelatin concentration, respectively, on the time required for the complete removal of alginate templates. Alginate/gelatin composite hydrogels were incubated in 100 mM EDTA solution to remove the alginate microbeads completely (n=4). Error bars, mean ± s.d. *p < 0.05. The experiment was repeated 4 times.

16

Because gelatin possesses a cell-binding moiety, such as RGD, there was no need to modify the surface of IOHs for cell seeding. To evaluate cell attachment on the surface of the gelatin hydrogel, the 10 wt% gelatin hydrogel was fabricated in a disc shape measuring 5 mm in height and 10 mm in diameter. MC3T3 fibroblasts (3.0 × 106 cells/mL) were then seeded on the top of the gelatin hydrogel. MC3T3 fibroblasts were cultured on the hydrogel for different times and then observed under a fluorescent microscope after live cell staining. As shown in Figure 6, MC3T3 fibroblasts were successfully attached on the surface of gelatin hydrogel on day 1 and proliferated over time. On day 7, the surface of the gelatin hydrogel was fully covered by the cells. The cells on the surface migrated, stretched, and interconnected with neighboring cells over time.

Figure 6. Fluorescent images of MC3T3 fibroblasts cultured on the surface of 10 wt% gelatin hydrogel on days 1, 4, and 7 after live/dead staining. Scale bar, 100 μm.

Next, we investigated the capacity of gelatin IOHs to support cell growth on the inner and outer surfaces of the inverse opal hydrogels to mimic a complicated cellular microenvironment. GFP-expressing C166 and mCherry-expressing D1 cells (mesenchymal stem cells, MSCs) were used to visualize the different locations of the two cell types in co-culture. The mCherryexpressing D1 cells were encapsulated in IOHs based on gelatin during UV-polymerization. Cells were then immersed in 100 mM EDTA for 2 h to remove alginate beads and incubated and washed in excess cell medium several times to wash away the EDTA. GFP-expressing C166

17

cells were seeded on the surface of IOHs after washing and incubated at 37°C for 7 days. D1 cells encapsulated in IOHs and C166 cells seeded on the surface of IOHs showed continuous proliferation over time (Figure 7a). These results showed that the optimized fabrication method for IOHs was suitable for co-culture systems requiring different physical locations of cultured cells, representing a major benefit of IOH hydrogels compared with dense inverse opal structures. The different types of cells could be encapsulated simultaneously in the gelatin matrix. D1mCherry and C166-GFP cells were both encapsulated in gelatin IOHs, and the cells were monitored over time, showing that the cells survived and proliferated, even after removal of alginate microbeads by EDTA treatment for around 2 h (Figure 7b). To evaluate possible UVinduced cell damage with UV irradiation, UV-induced formation of cyclobutane pyrimidine dimers (CPD) in DNA was accessed using immunofluorescence assay after irradiation of HCT166 cells with UV (365 nm) for 10 min (Figure S2). CPD produced after UV irradiation for 10 min was very little and thus current UV irradiation to prepare IOH has little genotoxic effect in the cells, which is consistent with several previous reports on high cell viability over 95% in UVcrosslinked gelatin hydrogels.32,33 Taken together, these data showed that optimization of the fabrication method for gelatin IOHs was achieved by controlling several parameters, including alginate viscosity, alginate concentration, size of sacrificial beads, calcium concentration used during the preparation of alginate microbeads, EDTA treatment time, and gelatin concentration, for establishment of cell-friendly IOHs. IOH could generate cell-cell contact between the cells incorporated within the hydrogel matrix and the cells attached on the surface of macropores, which might allow a possibility to use IOH as a model hydrogel system in the study of the particular cell arrangement such as extrinsic MSC line adjacent fibroblast. Also, the IOHs

18

generated through this method will have potential applications as a material platform to study paracrine effects and cell-cell interactions.

Figure 7. Fluorescent images of co-cultures of two different types of cells in gelatin IOHs. The cells were cultured and then imaged on days 1, 4, and 7. (a) mCherry-expressing D1 cells were encapsulated in IOHs, and GFP-expressing C166 cells were seeded on the surface of IOHs. (b) mCherry-expressing D1 cells and GFP-expressing C166 cells were simultaneously encapsulated in the gelatin IOHs. Scale bar, 100 μm.

4. Conclusion We have demonstrated the optimized fabrication of cell-friendly IOHs with uniform pore structure, 3D interconnectivity, and high cell viability, fulfilling many of the necessary conditions for a 3D scaffold. Compared to the previous strategies to generate 3D microenvironments through gas foaming, salt leaching, freeze drying, and electrospinning,9-16 gelatin IOHs provide uniform pore structure, size, and shape and enhanced pore connectivity. In addition, gelatin IOHs prepared by using cell-benign removal of alginate microbead template allow both the macropore surface and the inner matrix can be utilized for cell culture, which is different from other 3D scaffold systems including other inverse opal structures prepared by

19

using templates removable by toxic solvent.20-28 To fabricate cell-friendly IOHs, it was found that low-molecular-weight alginate and low concentrations of alginate and gelatin-methacrylate solution were necessary to guarantee high cell viability after removal of the alginate template by EDTA treatment for up to 2 h. Based on the optimized fabrication conditions, gelatin-based IOHs may have applications in tissue regeneration and as an in vitro three-dimensional model to study paracrine effects. Furthermore, an appropriate IOH with good mechanical properties could be used as a biomaterial-based platform for in vivo models in tissue regeneration and cancer development.

Acknowledgements This work was supported by grants funded by the National Research Foundation (NRF) under the Ministry of Science, ICT & Future Planning, Republic of Korea (2010-0027955, 2015R1A2A2A01005548), and a grant from the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (HI14C0211).

References (1) Dutta, R.C.; Dutta A.K. Cell-interactive 3D-scaffold; advances and applications. Biotechnol. Adv. 2009, 27, 334-339. (2) Langer, R.; Vacanti, J.P. Tissue engineering: the design and fabrication of living replacement devices for surgical reconstruction and transplantation. Lancet 1993, 354, 3233.

20

(3) Yang, S.; Leong, K.F.; Du, Z; Chua, C.K. The design of scaffolds for use in tissue engineering, Tissue Eng. 2001, 7, 679-89. (4) Owen, S.C.; Scoichet M.S. Design of three-dimensional biomimetic scaffolds. J Biomed Mater Res A 2010, 94, 1321-1331. (5) Hoffman, A.S. Hydrogels for biomedical applications. Adv. Drug Deliv. Rev. 2012, 64, 1823. (6) Lutolf, M.P. Biomaterials: Spotlight on hydrogels. Nat. Mater. 2009, 8, 451-453. (7) Nelson, C.M.; Liu, W.F.; Chen, C.S. Manipulation of cell-cell adhesion using bowtieshaped microwells, Methods Mol. Biol. 2007, 370, 1-10. (8) Choi, S.W.; Xie, J.; Xia, Y. Chitosan-based inverse opals: Three-dimensional scaffolds with uniform pore structure for cell culture. Adv. Mater. 2009, 21, 2997-3001. (9) Yoon, J.J.; Park, T.G. Degradation behaviors of biodegradable macroporous scaffolds prepared by gas foaming of effervescent salts. J. Biomed. Mater. Res. 2001, 55, 401-408. (10) Liao, C.J; Chen, C.F.; Chen, J.H.; Chiang, S.F.; Lin, Y.J.; Chang, K.Y. Fabrication of porous biodegradable polymers scaffolds using a solvent merging/particulate leaching method. J. Biomed. Mater. Res. 2002, 59, 676. (11) Ma, P.X.; Zhang, R. Synthetic nano-scale fibrous extracellular matrix. J. Biomed. Mater. Res. 1999, 46, 60-72. (12) Whang, K.; Goldstick, T.K.; Healy, K.E. A biodegradable polymer scaffold for delivery of osteotropic factors. Biomaterials 2000, 21, 2545-2551.

21

(13) Mooney, D.J; Baldwin, D.F.; SuH, N.P.; Vacanti, J.P.; Langer, R. Novel approach to fabricate porous sponges of poly (D, L-lactic-co-glycolic acid) without the use of organic solvents. Biomaterials 1996, 17, 1417-1422. (14) Matthews, J.A.; Wnek, G.E.; Simpson, D.G.; Bowlin, G. L. Electrospinning of collagen nanofibers. Biomacromolecules 2002, 3, 232-238. (15) Hollister, S.H. Porous scaffold design for tissue engineering. Nat. Mater. 2005, 4, 518-524. (16) Gallego, D.; Ferrell, N.; Sun, Y.; Hansford, D.J. Multilayer micromolding of degradable polymer tissue engineering scaffolds. Mater. Sci. Eng. C 2008, 28, 353-358. (17) Velev, O.D.; Lenhoff, A.M. Colloidal crystals as templates for porous materials. Curr. Opin. Colloid Interface Sci. 2000, 5, 56-63. (18) Xia, Y.; Gates, B.; Yin, Y.; Lu, Y. Monodispersed colloidal spheres: old materials with new applications. Adv. Mater. 2000, 12, 693-713. (19) Stein, A.; Schroden, R.C. Colloidal crystal templating of three-dimensionally ordered macroporous solid: materials for photonics and beyond. Curr. Opin. Solid State Mater. Sci. 2001, 5, 553-564. (20) Zhang, Y.S.; Choi, S.W.; Xia, Y. Inverse opal scaffolds for applications in regenerative medicine. Soft Mater 2013, 9, 9747-9754. (21) Wang, D.; Salgueirno-Maceira, V.; Liz-Marzan, L.M.; Caruso, F.

Gold-silica inverse

opals by colloidal crystal templating. Adv. Mater. 2002, 14, 908-912. (22) Wang, J.; Cao, Y.; Feng, Y.; Gao, J. Multiresponsive inverse-opal hydrogels. Adv. Mater. 2007, 19, 3865-3871.

22

(23) Choi, S.W.; Zhang, Y.; Thomopoulos, S.; Xia, Y. In vitro mineralization by preosteoblasts in Poly(Dl-lactide-co-glycolide) inverse opal scaffolds reinforced with hydroxyapatite nanoparticles. Langmuir 2010, 26, 12126-12131. (24) Stachowiak, A.N.; Irvine, D.J. Inverse opal hydrogel-collagen composite scaffolds as a supportive microenvironment for immune cell migration. J. Biomed. Mater. Res. A 2008, 85, 815-828. (25) Choi, S.W.; Zhang, Y.; Xia, Y. Three-dimensional scaffolds for tissue engineering: The importance of uniformity in pore size and structure. Langmuir 2010, 26, 19001-19006. (26) Liu, Y.; Wang, S.; Lee, J.W.; Kotov, N.A. A floating self-assembly route to colloidal crystal templates for 3D cell scaffolds. Chem. Mater. 2005, 17, 4918-4924. (27) Cassagneau, T.; Caruso, F. Conjugated polymer inverse opals for potentiometric biosensing. Adv. Mater. 2002, 14, 1837-1841. (28) Zhang, Y.; Wang, S.; Eghtedary, M.; Motamedi, M.; Nicholas, A. K. Inverted-colloidalcrystal hydrogel matrices as three-dimensional cell scaffolds. Adv. Fuct. Mater. 2005, 15, 725-731. (29) Kim, J.; Bencherif, S.A.; Li, W.A.; Mooney, D.J. Cell-friendly inverse opal-like hydrogels for a spatially separated co-culture system. Macromol. Rapid. Commun. 2014, 35, 15781586. (30) Nichol, J.W.; Koshy, S.T.; Bae, H.; Hwang, C.M.; Yamanlar, S.; Khademhosseini, A. Cellladen microengineered gelatin methacrylate hydrogel. Biomaterials 2010, 31, 5536-5544. (31) Kim, J.; Arifin, D.R.; Muja, N.; Kim, T.; Gilad, A.A.; Kim, H.; Arepally, A.; Hyeon, T.; bulte, J.W.M. Multifunctional capsule-in-capsules for immunoprotection and trimodal imaging. Angew. Chem. Int. Ed. 2011, 50, 2317-2321.

23

(32) Billiet, T.; Gevaert, E.; De Schryver, T.; Cornelissen, M.; Dubruel, P. The 3D printing of gelatin methacrylamide cell-laden tissue-engineered constructs with high cell viability. Biomaterials 2014, 35, 49-62. (33) Loessner, D.; Meinert, C.; Kaemmerer, E.; Martine, L.C.; Yue, K.; Levett, P.A.; Klein, T.J.; Melchels, F.P.; Khademhosseini, A.; Hutmacher, D.W.; Functionalization, preparation and use of cell-laden gelatin methacryloyl-based hydrogels as modular tissue culture platforms. Nat. Protoc. 2016, 11, 727-746.

24

For Table of Contents Fabrication of Cell-Benign Inverse Opal Hydrogels for Three Dimensional Cell Culture Pilseon Im,1 Dong Hwan Ji,1 Min Kyung Kim,1 and Jaeyun Kim1,2* 1

School of Chemical Engineering, Sungkyunkwan University (SKKU), Suwon 16419, Republic

of Korea 2

Samsung Advanced Institute for Health Science & Technology (SAIHST), Sungkyunkwan

University (SKKU), Suwon 16419, Republic of Korea. * To whom correspondence should be addressed. E-mail: [email protected]

25