High-Strength Photoresponsive Hydrogels Enable Surface-Mediated

Sep 8, 2014 - ... by the photoinitiated copolymerization of acrylamide (AAm, hydrophilic hydrogen bonding monomer), 2-vinyl-4,6-diamino-1,3,5-triazine...
0 downloads 0 Views 7MB Size
Download PDF
Article pubs.acs.org/Langmuir

High-Strength Photoresponsive Hydrogels Enable Surface-Mediated Gene Delivery and Light-Induced Reversible Cell Adhesion/ Detachment Ning Wang, Yongmao Li, Yinyu Zhang, Yue Liao, and Wenguang Liu* School of Materials Science and Engineering, Tianjin Key Laboratory of Composite and Functional Materials, Tianjin University, Tianjin 300072, PR China S Supporting Information *

ABSTRACT: In the present study, high-strength photoresponsive hydrogels were prepared by the photoinitiated copolymerization of acrylamide (AAm, hydrophilic hydrogen bonding monomer), 2-vinyl-4,6-diamino-1,3,5-triazine (VDT, hydrophobic hydrogen bonding monomer), and spiropyrancontaining monomer (SPAA) in the presence of cross-linker poly(ethylene glycol) diacrylate (PEG575DA, Mn = 575). The double hydrogen bondings from AAm−AAm and diaminotriazine−diaminotriazine contributed to the considerable enhancement in tensile and compressive properties of the hydrogels, which showed an excellent ability to resist a variety of external forces. Fifteen minutes of UV (365 nm) irradiation led to the detachment of adhered cells due to the increased surface hydrophilicity caused by the isomerization of spiropyran moieties. Furthermore, repeated attachment/detachment of cells was realized by the alternate illumination of visible and UV light. Reverse gene transfection was carried out successfully by anchoring the PVDT/pDNA complex nanoparticles on the gel surface through hydrogen bonding between diaminotriazine motifs prior to cell seeding. Importantly, fibronectin (FN) modification combined with supplementing PVDT/pDNA complex nanoparticles after the first cycle of reverse gene transfection, so-called sandwich gene transfection, further increased the gene transfection level. A short time of UV light exposure could result in the nonharmful detachment of gene-modified cells from the gel surface. This high-strength photosensitive hydrogel holds potential as a reusable soft−wet platform for cell harvesting as well as gene transfection operation at higher efficiency.

■

INTRODUCTION Hydrogels have emerged as promising materials in the biomedical field because of their similarity to the native extracellular matrix (ECM).1,2 However, conventional synthetic hydrogels are generally mechanically weak, which severely limits their applications.3 In the past decade, many efforts were devoted to fabricating high-performance hydrogels, such as slide-ring (SR) hydrogels,4 double-network (DN) hydrogels,5 nanocomposite (NC) hydrogels,6macromolecular microsphere composite (MCC) hydrogels,7 tetra-arm poly(ethylene glycol) (tetra-PEG) hydrogels,8 hydrogen bonding/dipole−dipole strengthening hydrogels,9,10 and so on. Nevertheless, little consideration was involved in introducing biofunctions into high-strength hydrogel systems. In addition, the mechanically strong hydrogels fail to mimic the dynamics of the environment because of the lack of stimuli responsiveness.11 Previously, our group synthesized strong hydrogels based on the multiple hydrogen bonding strengthening principle of interdiaminotriazine (DAT).9,11 The hydrogels showed both high tensile and high compressive strengths. Moreover, the copolymerization of N-isopropylacrylamide or 2-(2-methoxyethoxy) ethyl methacrylate with 2-vinyl-4,6-diamino-1,3,5© 2014 American Chemical Society

triazine (VDT) resulted in strong, temperature-sensitive hydrogels that could tune enzyme-free cell detachment as a result of the switchable surface hydrophilicity change caused by temperature variation. This multifunctional strong hydrogel was shown to be a promising robust soft−wet platform for manipulating gene delivery as well as cell adhesion. However, the detachment of cells could not be controlled regionally, and the time of temperature treatment for cell detachment was as long as 1 h. To control cell detachment more rapidly and more flexibly, we constructed a photosensitive hydrogel with good compressive strength by introducing spiropyran moieties into PVDT-based hydrogels.12 Fifteen minutes of UV irradiation (365 nm) could lead to the increased surface hydrophilicity of the hydrogel and the accompanying nonharmful detachment of cells. Furthermore, the selective detachment of cells could be controlled by applying UV light illumination to the unmasked gel surface region. Also, the PVDT/pDNA nanocomplexes could be adsorbed on a hydrogel surface through self hydrogen Received: July 23, 2014 Revised: September 4, 2014 Published: September 8, 2014 11823

dx.doi.org/10.1021/la502916j | Langmuir 2014, 30, 11823−11832

Langmuir

Article

Figure 1. Molecular structure of P(AAm-co-VDT-co-SPAA) hydrogels and a schematic depiction of the double hydrogen bonding strengthening mechanism.

were reported.13,14 In fact, apart from other physical crosslinkings, the cooperative hydrogen bonding from amides also plays an important role in reinforcing hydrogels.14 In our previous work,15 we fabricated a series of PVDT-based hydrogels by using poly(ethyelene glycol) dimethacrylate (PEGDA) cross-linkers with different molecular weights and found that the hydrogel cross-linked with PEG575DA had higher tensile and compressive strengths than did other hydrogels cross-linked with higher-molecular-weight PEGDA (PEG4kDA, PEG10kDA, and PEG35kDA). In this work, to construct a strong hydrogel with a superior multifunction of photoresponsiveness, surface-mediated gene transfection, and cell detachment, VDT (hydrophobic hydrogen bonding monomer), AAm (hydrophilic hydrogen bonding monomer),

bonding of diaminotriazine, thereby achieving reverse gene transfection, and the short time of UV light irradiation could induce the release of gene-modified cells. However, the tensile strength was merely on the order of kilopascals, so the hydrogels were easily damaged during actual operation; this may restrain the reusability of the hydrogels. VDT and spiropyran monomers were essentially hydrophobic, and thus the spiropyran monomer units are prone to blend into the microregions of diaminotriazine hydrogen bondings, which may result in the segregation of diaminotriazine−diaminotriazine hydrogen bondings, consequently causing the deterioration of the tensile strength of this photorepsonsive PVDT-based hydrogel.12 Recently, very tough and stretchable polyacrylamide (PAAm)-based hydrogels 11824

dx.doi.org/10.1021/la502916j | Langmuir 2014, 30, 11823−11832

Langmuir

Article

Figure 2. Schematic illustration of PAVSP hydrogel surface-mediated gene transfection. (a) Reverse gene transfection. (b) Fibronectin-modified reverse gene transfection. (c) Sandwich gene transfection. y-z stands for the P(AAm-co-VDT-co-SPAA) hydrogel with a weight ratio of AAm/VDT/SPAA (x/y/z). Similarly, P(AAm-co-VDT) hydrogel (PAVx-y) with a weight ratio of AAm/VDT (x/y), crosslinked acrylamide hydrogel (cr-PAAm), and cross-linked VDT (crPVDT) were prepared. Measurement of Mechanical Properties. The tensile and compressive properties of the hydrogels were measured on a WDW-05 electromechanical tester (Time Group Inc., China) at room temperature. All of the hydrogel samples were tested after full equilibration in deionized water. For tensile tests, the hydrogel strips with a thickness of 0.4 mm were cut into a rectangle (20 mm in length and 5 mm in width). The cylinder-shaped hydrogels (4 mm in diameter and 6 mm thick) were used for compression tests. The crosshead speed was set at 100 mm/min for the tensile test and 10 mm/min for the compression test. Three specimens were tested for each hydrogel sample. Characterization of the UV−Visible Light Response. To investigate the response of PAVSP hydrogels to visible and UV light, we used a UV−visible spectrophotometer (TU-1810, PERSEE, Beijing, China) to record the UV−vis adsorption spectrum of the sample after visible light irradiation for 1 h. Then the sample was exposed to UV light (365 nm, 2 mW/cm2) for 15 min and scanned by the same apparatus with the same mode. By contrast, the same tests were performed on the PAV hydrogel sample. Measurement of the Contact Angle (CA). The water contact angles on the hydrogel surfaces were measured by using the sessile drop method at room temperature. First, after 1 h of visible light illumination, the equilibrated hydrogel sample was placed on a glass slide and dried superficially with filter paper. The CA was recorded by a commercial contact angle meter (JC200D4, POWEREARCH, Shanghai, China) after a 4 μL drop of deionized water was placed carefully on the hydrogel surface. Herein, at least six different spots were measured on each sample. Then the hydrogel sample was placed under UV light (365 nm, 2 mW/cm2) for 15 min and tested by the contact angle meter with the same method. To investigate the reversibility of CA change in response to alternating light irradiation,

and 2-[1-acrylate-3′,3′-dimethyl-6-nitrospiro (indoline-2′,2[2H-1]-benzopyran)] acrylamide (SPAA, light-sensitive monomer) were copolymerized in one step in the presence of hydrophilic cross-linker PEG575DA. The mechanical properties of the hydrogels and photoresponsive behavior will be determined; the repeatable light-induced selective cell detachment on the hydrogel surface and surface-mediated gene transfection will be examined in this study. This superior combination of multiple functions is promising in extending the application scope of strong hydrogels.

■

EXPERIMENT

Materials. 2-[1-Acrylate-3′,3′-dimethyl-6-nitrospiro (indoline-2′,2[2H-1]-benzopyran)] acrylamide (SPAA) was prepared according to our previous work.12 Acrylamide (AAm, 98%) and 2-vinyl-4,6diamino-1,3,5-triazine (VDT, ≥95%) were obtained from Tokyo Kasei Kogyo. Poly(ethylene glycol) diacrylate (PEG575DA, Mn = 575, ≥99%), phenylbis(2,4,6-trimethyl benzoyl) phosphine oxide (PBPO, 97%), and fibronectin (FN) were provided by Sigma-Aldrich (USA). Dimethyl sulfoxide (DMSO, analytical grade) was purchased from Tianjin Guangfu Fine Chemical Research (China). Synthesis of P(AAm-co-VDT-co-SPAA) (PAVSP) Hydrogels. The recipes for preparing hydrogels are listed in Table S1. AAm, VDT, and SPAA were dissolved in DMSO, with a total monomer concentration of 10 wt %. After cross-linker PEGDA and photoinitiator PBPO were added, the homogeneous solution was poured into a plastic mold and put under white light for 30 min. The white light source (LED lamp, 600 lm in intensity) was placed 20 cm from the mold. Herein, the initial feed ratio of monomers/cross-linker/ initiator was fixed at 3:2:0.1 (w/w/w), and the weight ratio of photosensitive monomer to nonphotosensitive monomers was kept at 1:20. After light irradiation, the product was removed from the mold and then soaked in distilled water for a week, with daily water exchange to remove any impurities. The molecular structure of the hydrogel is depicted in Figure 1. It is noted that in Table S1 PAVSPx11825

dx.doi.org/10.1021/la502916j | Langmuir 2014, 30, 11823−11832

Langmuir

Article

another cycle of visible−UV light was applied to the sample and the CAs were recorded accordingly. Cell Culture and Detachment Assay. After being sterilized in 75% medical alcohol overnight, the hydrogel samples were placed into a 96-well culture plate and immersed in pH 7.4 phosphate buffer solution (PBS) for 8 h to reach swelling equilibrium. Then the mouse fibroblast L929 cells were seeded onto the hydrogels with a density of about 1 × 104 cells/well and cultured with MEM/EBSS culture medium supplemented with 10% horse serum (HS) at 37 °C in 5% CO2 for 24 h to guarantee the adhesion and spreading of cells. To evaluate the cell adhesion rate, the cell-adhered hydrogel was transplanted carefully into a new well and tested by MTT assay using the cells with the same density in the absence of hydrogels as the control. The cell adhesion rate (mean ± SD, n = 3) was calculated as follows: cell adhesion rate(%) =

hydrogels were washed gently with pH 7.4 PBS. Then African green monkey kidney cells (COS-7) were seeded onto the hydrogel surface at 2 × 104 cells/well and cultured with DMEM medium supplemented with 10% fetal bovine serum (FBS) for 48 h to complete reverse gene transfection. Herein, the medium was replaced with fresh DMEM plus 10% FBS when the transfection was carried out for 24 h. For the FN-modified reverse gene transfection, a 50 μL solution of FN (0.02 mg/mL) was added to the surface of the hydrogel sample for adsorption. Twelve hours later, the residual solution was removed and the hydrogel surface was washed gently with PBS. After that, reverse gene transfection was performed as previously mentioned (Figure 2b). On the basis of the FN-modified reverse gene transfection, a sandwich transfection was further carried out. That is, after the COS-7 cells were cultured on the FN-modified and PVDT/pDNA complex adsorbed hydrogel surface for 24 h, another dose of PVDT/pDNA complex nanoparticles was added, and the transfection was performed for another 24 h. The process is shown in Figure 2c. Measurement of Transfection Efficiency. The transfected cells were released from the hydrogels by UV illumination for 15 min. The obtained cell suspension was centrifuged for 5 min, and then the medium was discarded. After being washed with PBS twice, the cells were treated with 100 μL of reporter lysis buffer (RLB, Promega, USA) for 15 min, followed by a freeze−thaw cycle to ensure complete lysis. The resulting lysate was centrifuged, and the supernatant was used for the measurement of relative light units (RLU) and protein concentration. A 50 μL portion of the supernatant of each sample was mixed with 50 μL of the luciferase reagent (Bright-Glo luciferase assay system, Promega, USA) and was tested with a 1420 multilabel counter (Wallac, USA) to measure the number of relative light units (RLU). The protein concentration was measured with a bicinchoninic acid (BCA) protein assay (Pierce, USA). The luciferase activity was expressed as the number of RLUs per milligram of cell protein (mean ± SD, n = 3).12 Statistical Analysis. All of the values are represented as the mean ± standard deviation (SD). Statistical analysis was performed using the two-population Student t test. Differences were considered statistically significant when p < 0.05. At least three specimens were tested for each sample.

Abs(attached cells on hydrogels) × 100% Abs(control)

In the cell detachment assay, after cell culture on the hydrogel surface for 24 h, the cell-adhered hydrogel was irradiated by UV light (365 nm, 2 mW/cm2) for 15 min to induce a hydrophilicity change on the surface. Then the irradiated hydrogel was washed gently with pH 7.4 PBS to promote the detachment of cells from the hydrogel surface. To quantify the degree of cell detachment on the hydrogels, we performed an MTT assay to measure the cell detachment efficiency. After UV irradiation, the detached cells and the hydrogel with the residual cells were transferred to new wells and were separately tested by the MTT method to record their absorbance values. The cell detachment efficiency (mean ± SD, n = 3) was calculated as follows:

cell detachment efficiency(%) =

Abs(detached cells) [Abs(detached cells)

+ Abs(residual cells on the surface of hydrogels)] × 100% To investigate the reversibility and reusability of the photosensitive PAVSP hydrogels for controlling cell adhesion and detachment, the cell-detached hydrogels were placed under visible light for 1 h to induce the SPAA moieties to convert back to the hydrophobic closedring SP structure. Then the L929 cells were seeded for another cycle of cell adherence and detachment. In this work, three adhere−detach cycles were carried out, and the detachment efficiency in each cycle was measured by using the MTT method. To study the influence of UV irradiation on cell apoptosis, L929 cells were stained with Annexin V-FITC/PI after UV irradiation for 0, 15, and 30 min and then were evaluated by flow cytometry (FACS Calibur, BD, USA). Regional Control of Cell Detachment. To control cell detachment regionally, we shielded the cell-adhered hydrogel with a light-proof photomask on one half. Then the whole piece of hydrogel was put under UV light for 15 min. As the light source and photomask were removed, the hydrogel surface was washed gently with pH 7.4 PBS to rinse the detached cells away. The residual cells on the hydrogel were observed and recorded with a phase-contrast microscope (Olympus, Japan). Effect of Fibronectin (FN) Modification on Cell Behavior. A 50 μL solution of FN (0.02 mg/mL) was added to the surface of each sterilized hydrogel sample for adsorption. Twelve hours later, the residual solution was removed and the hydrogel surface was washed with PBS. Then L929 cells were seeded on the hydrogel surface, and light-tuned cell attachment and detachment was performed in the same way as above. The adhesion rate and detachment efficiency were also tested with the MTT method. Hydrogel Surface-Mediated Gene Transfection. Surfacemediated or reverse gene transfection was performed according to the schematic description shown in Figure 2a. The sterilized hydrogels were placed on a 96-well plate immersed in pH 7.4 PBS solution, and a 50 μL solution of PVDT/pDNA nanocomplexes (containing 1 μg of DNA, PVDT/pDNA = 0.31 (w/w)) was added to each well in the 96well culture plate to replace the PBS. After adsorption for 12 h, the

■

RESULTS AND DISCUSSION Mechanical Properties and EWCs of Hydrogels. In our previous work,12 we reported photoresponsive P(OEGMA-coVDT-co-SPAA) hydrogels. The gels exhibited better compressive strengths but failed to withstand stretching even under a lower stress. As described in the Introduction, spiropyran units could weaken the hydrogen bonding interaction between diaminotriazine (DAT) motifs as a result of the miscible blending in DAT−DAT microdomains. In this work, to improve both the tensile and compressive strengths of the photoresponsive hydrogels, we proposed a dual hydrogen bonding strengthening strategy. As depicted in Figure 1, the hydrogel consists of two types of hydrogen bonding units. Diaminotriazine (DAT) residue in VDT can form DAT−DAT hydrogen bonds, and amide groups in acrylamide can form AAm−AAm hydrogen bonds.16,17 Our previous studies have demonstrated that the hydrophobic microdomain of inter-DAT multiple hydrogen bonding was so stable in an aqueous medium that it could contribute to a remarkable increase in the mechanical properties of the gels.9,16 For amide hydrogen bondings, the calculation of free energies suggests that these hydrophilic H-bondings are rather unstable in the aqueous phase but become very stable in a hydrophobic environment.17 On the basis of the nature of these two types of H-bondings, we hypothesize that the H-bondings between AAm units could be stabilized while they are surrounded by the DAT−DAT 11826

dx.doi.org/10.1021/la502916j | Langmuir 2014, 30, 11823−11832

Langmuir

Article

Table 1. EWCs and Mechanical Properties of Hydrogels tensile properties hydrogel cr-PAAm PAV1-1 PAVSP1-1-0.1 PAV1-2 PAVSP1-2-0.15 PAV1-3 PAVSP1-3-0.2 cr-PVDT

EWC (%) 92.2 85.8 85.4 75.6 74.2 64.4 58.6 50.4

± ± ± ± ± ± ± ±

0.1 0.9 0.4 2.5 0.2 0.5 2.0 2.0

tensile strength (kPa) 4.5 11.4 12.7 100.4 111.2 240.4 250.0 110.4

± ± ± ± ± ± ± ±

compression properties

elongation at break (%)

3.2 3.3 4.7 15.8 16.7 18.6 20.1 20.8

19.3 43.2 41.6 230.4 213.0 274.8 268.4 100.3

micromilieu and thus act as an effective physical cross-linker (Figure 1). Table 1 clearly shows that the cross-linked polyacrylamide hydrogel (cr-PAAm) in the absence of VDT is very weak; its tensile strength and compressive strength are only 4.5 and 137 kPa, respectively. Also, the gel is rather brittle with merely 19% breaking strain. By copolymerizing with VDT, the mechanical properties of the P(VDT-AAm) gels are increased considerably. In trial-and-error experiments, we found that the hydrogels with VDT/AAm ratios of less than 1/1 were too weak to operate against the mechanical tester. However, the hydrogels with VDT/AAm ratios above 3/1 were shown to be highly hydrophobic and rather opaque and thus unsuitable for subsequent application. So in this work, the PAVSP hydrogels with AAm/VDT ratios from 1/1 to 1/3 were studied. In particular, at AAm/VDT ratios of 1/2 and 1/3, the tensile strengths of the P(VDT-AAm) hydrogel are enhanced by 25fold and 60-fold, respectively, compared to that of cr-PAAm; correspondingly, the P(VDT-AAm) gel exhibits 20-fold and 300-fold increases in compressive strength, respectively. Furthermore, when the ratio of VDT is dominant, the hydrogel becomes elastic with over 200% elongation at break. These results show that the multiple hydrogen bonding originating from diaminotriazine motifs contributes to a remarkable increase in mechanical strength.9,15,18 Compared to single hydrogen bonding poly(OEGMA-co-VDT) hydrogel systems previously reported,12 the dual hydrogen bonding P(VDTAAm) hydrogels demonstrate a 0.7−7-fold increase in the compressive strengths over the 1/1 to 1/3 range of AAm/VDT ratios. The result proves that the AAm−AAm hydrogen bonding is truly stabilized by the microenvironment of diaminotriazine−diaminotrazine hydrogen bonding, and these two types of H-bonding lead to increased physical cross-linking so the comprehensive strengths are enhanced accordingly. By contrast, we also find that at a 1/3 AAm/VDT ratio the mechanical properties of dual hydrogen bonding hydrogels are higher than those of single hydrogen bonding cr-PVDT. We note that at the same AAm/VDT ratio the P(AAm-co-VDT-coSPAA) hydrogels exhibit similar mechanical properties to those of P(VDT-AAm) gels. This indicates that introducing SPAA into the hydrogel network did not influence the mechanical strength because of the aided reinforcing role of AAm Hbonding. Figure 3 clearly demonstrates that the PAVSP1-2-0.15 hydrogel can bear knotting, twisting, and stretching without any damage, implying its excellent ability to withstand handling and manipulation under external force as a soft−wet platform. It is necessary to point out that the PEGDA575-cross-linked SPAA hydrogel was too weak to resist the clamp of the electromechanical tester. Although SPAA contains amide bonds in its

± ± ± ± ± ± ± ±

11.1 10.3 15.3 6.8 9.4 25.7 30.5 10.8

stress (kPa) 137.8 289.0 292.5 3179.1 3483.3 49 309.0 53 362.5 3725.0

± ± ± ± ± ± ± ±

23.0 27.3 25.0 784.3 812.0 4179.2 3488.1 1492.0

fracture strain (%) 50.2 83.4 80.4 95.1 94.0 96.3 95.1 88.0

± ± ± ± ± ± ± ±

1.7 2.2 4.6 2.1 0.9 1.4 2.2 2.0

Figure 3. Photographs of a PAVSP1-2-0.15 hydrogel experiencing (a) stretching, (b) knotting and stretching, (c) twisting and stretching, and (d) compression and recovery.

molecular structure, its bulky side group may interfere with the formation of amide−amide hydrogen bonds. Table 1 lists the EWCs of the hydrogels. The cross-linked PVDT hydrogel (cr-PVDT) in the absence of AAm is only 50%. Copolymerizing hydrophilic AAm can increase the ability of the gel to absorb water; nevertheless, along with an increment of VDT content, the EWCs of both PAV and PAVSP copolymer hydrogels decrease because of the increased compactness of the network resulting from the increased density of physical cross-linking. UV−Visible Light Responsiveness. A UV−visible spectrophotometer was used to record the absorption spectra of PAVSP and PAV hydrogels under visible light and UV irradiation, respectively (Figure 4). As shown in the figure, after UV irradiation for 15 min, an absorption peak appears at about 540 nm (Figure 4a), which is distinctly different from the initial spectrum recorded under visible light. And the maximum absorbance did not increase with prolonged irradiation time, suggesting that the equilibrium of transformation was reached. We can see that the color of the PAVSP hydrogel changes from yellow to purple after UV illumination (inset picture). The 11827

dx.doi.org/10.1021/la502916j | Langmuir 2014, 30, 11823−11832

Langmuir

Article

Figure 4. UV−vis absorbance spectra of (a) PAVSP1-3-0.2 and (b) PAV1-3 hydrogels before and after UV irradiation. (Inset) Changes in the color of hydrogels after visible and UV light irradiation.

MC form.21 And the magnitude of the CA change is larger than that for the systems previously reported.12,22,23 In contrast, the CAs of PAV hydrogels in the absence of SPAA show no significant difference after UV irradiation. To investigate the reversibility of light-induced CA changes, another visible−UV light irradiation cycle was carried out. It is observed in Figure 5 that the CAs of PAVSP hydrogels recover to almost the initial values after visible light irradiation for 1 h. And after more UV irradiation for 15 min, the CAs decrease to a similar extent as in the first cycle, implying the reversibility of PAVSP hydrogels. Cytotoxicity of PAVSP Hydrogels. The PAVSP hydrogelconditioned medium was used for the L929 cell culture to investigate the cytotoxicity of the hydrogels. After cell culture for 48 h, the measured cell viability is shown in Figure S1, where the viability of nontreated cells is set as 100%. It is found that the cell viabilities on hydrogels with all of the monomer ratios are higher than 90%, indicating the low cytotoxicity of the PAVSP hydrogel. Light-Induced Cell Detachment. Figure 6 shows the cell adhesion rate and light-induced cell detachment efficiency on PAVSP (a) and PAV (b) hydrogels with varied monomer ratios. For the PAVSP hydrogels, the cell adhesion rate increases with the increment of the VDT/AAm ratio and reaches the maximum, 95.6% on PAVSP1-3-0.2. From the CAs measured after visible light irradiation in Figure 5, we can see that the CA increases with the increase in VDT content, indicating the enhancement of hydrophobicity. It is reported that cell adhesion is highly affected by surface wettability and that the cells are apt to attach to the surface with appropriate hydrophobicity.24,25 It is rational to think that the hydrophobicity of PAVSP1-3-0.2 is suited for the adhesion of L929 cells. After UV irradiation for 15 min, the cells spreading well on the PAVSP hydrogel retract automatically and exhibit a roundlike morphology (Figure S2). After being gently blown on with a pipet, the cells detach from the surface readily. This is because UV irradiation induced the transformation of spiropyran from SP to a hydrophilic open-ring MC structure, causing the increase in hydrophilicity on the hydrogel surface. Consequently, the cells sense the surface change that is against adhesion and retract in response. As shown in Figure 6a, the detachment efficiency is over 80% for all of the PAVSP hydrogels and shows a slight tendency to decrease with increasing VDT content. This is possibly because the cells are easier to detach from the more hydrophilic PAVSP1-1-0.1

phenomenon originated from the isomerization of the spiropyran group in SPAA moieties from a closed-ring, hydrophobic SP form to an open-ring, hydrophilic MC form under UV irradiation.19,20 Moreover, the absorption spectrum of the PAVSP gel could get back to the initial state, and the color could return to yellow when it was irradiated again by visible light for 1 h, demonstrating the reversible light responsiveness of PAVSP hydrogels. For comparison, the PAV hydrogel was irradiated with same UV and visible light. Figure 4b shows that there is no change in the spectrum after UV irradiation, and the hydrogel remains colorless and transparent (inset picture) during the whole process, confirming the crucial role that SPAA plays in inducing the photoresponsive changes. To investigate the hydrophilicity change of the hydrogel surface in response to light irradiation, the water contact angles (CAs) on PAV and PAVSP hydrogel surfaces before and after UV illumination were tested by using the sessile drop method, as shown in Figure 5. After UV irradiation for 15 min, the CAs of PAVSP hydrogels decrease significantly, with a maximum change of 13.3° indicating the enhancement of hydrophilicity. This is because of the isomerization of the hydrophobic SP form of spiropyran groups in SPAA moieties to the hydrophilic

Figure 5. Changes in water contact angles (CAs) on PAV and PAVSP hydrogels after two cycles of visible−UV light irradiation. Data are presented as mean ± SD. Asterisks (*) denote significant differences (p < 0.05) compared to the CA after the first round of visible light irradiation. 11828

dx.doi.org/10.1021/la502916j | Langmuir 2014, 30, 11823−11832

Langmuir

Article

Figure 6. Cell adherence rate and detachment efficiency on the surfaces of (a) PAVSP and (b) PAV hydrogels with different monomer ratios.

was washed gently with PBS and was recorded with a microscope, as shown in Figure 8. Obviously, the cells on the

surface that exhibits a lower binding force with cells. However, the cells bind tightly on the hydrophobic PAVSP1-3-0.2 hydrogel surface so that the hydrophilicity change caused by UV irradiation does not produce enough driving force for cell detachment. In contrast, the cell detachment efficiency on PAV hydrogels is only less than 10%, evidencing the photosensitivity of SPAA moieties for light-induced cell detachment. In light of excellent mechanical properties of the hydrogels and their reversibility of light-induced CA changes, we inspected the reusability of PAVSP for repeated cell adhesion and detachment. Figure 7 shows that the efficiency of cell

Figure 8. Microscopy images of L929 cells on the PAVSP hydrogel surface after regional cell detachment.

irradiated region were released from the hydrogel surface easily because of the isomerization of SPAA moieties. However, on the shielded region, the adhered cells retained their initial elongated morphology with good spreading capacity, and no cell release is observed after PBS washing, indicating the realization of regional cell detachment. In our previous work,12 the cell viability assayed by the MTT method was not affected significantly by UV irradiation for 15 min or even for 30 min. In addition, after being detached from the PAVSP hydrogel surface by UV irradiation, the cells were recultured on another plate for another 24 h. As a result, the cells restored normal growth, suggesting good viability after UV illumination. Herein, to investigate the effect of UV light on cell apoptosis, the cells were irradiated for 0, 15, and 30 min under UV light and then were tested by flow cytometry. Figure S3 shows that the fraction of apoptotic cells does not change significantly after UV irradiation for 30 min. Effect of FN Modification on Cell Behavior. Herein, the hydrogel was modified with fibronectin (FN), a known extracellular matrix protein for promoting cell adhesion. After FN modification, the cell adhesion rate and cell detachment efficiency were tested, as shown in Figure 9. At the same VDT/

Figure 7. Effect of detaching cycles on the cell detachment efficiency. Results are presented as mean ± SD. Using detachment efficiency in the first cycle as a control, no significant differences were shown in the detachment efficiency in the second and third cycles, respectively, as calculated with the two-population Student t test.

detachment in three cycles of cell adherence−detachment on the same hydrogel surface still remains over 80%. And there is no significant difference in the cell detachment efficiency among the three repeats for the selected hydrogel samples, indicating the good reusability of PAVSP hydrogels. It is noted that the mechanical properties (e.g., tensile and compressive strength) of PAVSP hydrogels showed no significant change after three cycles of light-induced cell attachment−detachment assays (data not shown) and neither did their shapes and sizes. As a photoresponsive material, PAVSP is able to allow spatiotemporal control of the surface pattern. This characteristic can be translated into tuning the detachment of cells in selective regions. In our experiment, the cells seeded onto the PAVSP hydrogel surface were partially shielded and then exposed to UV light. After UV irradiation, the hydrogel surface 11829

dx.doi.org/10.1021/la502916j | Langmuir 2014, 30, 11823−11832

Langmuir

Article

Figure 9. Cell adherence rate and detachment efficiency on the surface of fibronectin-modified PAVSP hydrogels. Figure 10. Expression level of reverse gene transfection in COS-7 cells mediated by the hydrogel surface. Results are presented as the mean ± SD. Asterisks (*) denote significant differences (p < 0.05) calculated with the two-population Student t test using the expression level on the PAVSP1-0-0.05 hydrogel as a control.

AAm ratio, the cell adhesion rate of the FN gel is higher than that of gel without the FN coating (Figure 6a), demonstrating that cell adhesion is enhanced notably by FN modification resulting from cell interactions with FN fibrils via integrin cell− surface receptors.26,27 After UV irradiation, the cells can also be detached automatically, with the detachment efficiency remaining above 80%. Unlike the cell detachment from the non-FN-modified gel as shown in Figure 6a, the cell detachment efficiency on FN-modified gel increases with the increment of VDT/AAm and reaches a maximum at PAVSP13-0.2. This is because FN is apt to be adsorbed on the more hydrophobic PAVSP1-3-0.2 surface via hydrophobic interaction and can be released by UV irradiation as a result of the enhancement of hydrophilicity,28,29 which provides an additional driving force for cell detachment. Hydrogel Surface-Mediated Gene Transfection. Recently, reverse gene transfection has been carried out by culturing cells on a plasmid DNA-anchored substrate to prolong the contact time of DNA with cells, thus increasing the cellular uptake of polyplexes, in contrast to conventional bolus gene transfection.30 In this work, we used linear PVDT (prepared in our previous work31) as a gene vector to form complex nanoparticles with plasmid DNA via hydrogen bondings between diaminotriazine groups and base pairs. Then the complex particles were adsorbed on PAVSP hydrogel surfaces through hydrogen bonding among the diaminotriazines in between PVDT and PAVSP, followed by culturing COS-7 cells on the complex particle-anchored hydrogel surface to perform gene transfection (Figure 2a). The transfected cells were released from the hydrogel surface by UV irradiation for 15 min, and the transfection efficiency was measured (Figure 10). It is observed that the transfection efficiency is enhanced with the increase in the VDT/AAm ratio, which is ascribed to the increased number of PVDT/pDNA complex particles adsorbed on the gel surface. In Figure S4, the PVDT/YOYO-1labeled DNA complex particles adsorbed on PAVSP hydrogels display an increasing trend along with an increase in the VDT content in hydrogels due to the increased hydrogen bonding between PVDT vectors and PAVSP hydrogels. For PAVSP1-00.05 without VDT, a very small number of fluorescent spots are observed in Figure S4a, indicating a small number of PVDT/ pDNA nanocomplexes adsorbed via physical adsorption, which eventually results in a low transfection efficiency of 5.5 × 105 RLU/mg protein. As for PAVSP1-3-0.2 that exhibits maximal PVDT/pDNA adsorption, the highest transfection efficiency, 6.6 × 106 RLU/mg protein, is achieved; it is 11 times higher than that on PAVSP1-0-0.05 and much higher than that with

the traditional liquid transfection method using PVDT as a vector as reported in our previous work.31 To examine the possible negative effect of UV light irradiation on pDNA, we measured the adsorption spectra of an aqueous solution of pDNA before and after UV irradiation by using a UV−vis spectrophotometer. The results showed that the adsorption spectra of pDNA remained unchanged after UV irradiation (data not shown), indicating that no damage occurred to DNA. In our previous work,12 we also demonstrated that UV light treatment did not influence the level of gene expression, showing that UV irradiation had no influence on cell function. To improve gene transfection, we modified the hydrogel surfaces with FN prior to reverse gene transfection, as depicted in Figure 2b. The efficiency shown in Figure 11 is higher than

Figure 11. Effect of different methods of transfection on the expression level in COS-7 cells. Results are presented as the mean ± SD. Asterisks (*) denote significant differences (p < 0.05) calculated using the two-population Student t test, with the expression level resulting from reverse gene transfection as a control.

that of surface-mediated gene transfection without the FN coating. An explanation is that the gene transfection is dependent on the density of immobilized cells, and a lower cell density is believed to be more sensitive to the cytotoxicity of pDNA complexes, which can lead to the reduction of the transfection efficiency.32 On the contrary, improved cell 11830

dx.doi.org/10.1021/la502916j | Langmuir 2014, 30, 11823−11832

Langmuir

Article

applications of hydrogels in regenerative medicine. Adv. Mater. 2014, 26, 85−124. (2) Peppas, N. A.; Hilt, J. Z.; Khademhosseini, A.; Langer, R. Hydrogels in biology and medicine: from molecular principles to bionanotechnology. Adv. Mater. 2006, 18, 1345−1360. (3) Carr, L.; Cheng, G.; Xue, H.; Jiang, S. Y. Engineering the polymer backbone to strengthen nonfouling sulfobetaine hydrogels. Langmuir 2010, 26, 14793−14798. (4) Okumura, Y.; Ito, K. The polyrotaxane gel: a topological gel by figure-of-eight cross-links. Adv. Mater. 2001, 13, 485−487. (5) Gong, J. P.; Katsuyama, Y.; Kurokawa, T.; Osada, Y. Doublenetwork hydrogels with extremely high mechanical strength. Adv. Mater. 2003, 15, 1155−1158. (6) Haraguchi, K.; Ebato, M.; Takehisa, T. Polymer-clay nanocomposites exhibiting abnormal necking phenomena accompanied by extremely large reversible elongations and excellent transparency. Adv. Mater. 2006, 18, 2250−2254. (7) Huang, T.; Xu, H. G.; Jiao, K. X.; Zhu, L. P.; Brown, H. R.; Wang, H. L. A novel hydrogel with high mechanical strength: a macromolecular microsphere composite hydrogel. Adv. Mater. 2007, 19, 1622−1626. (8) Kamata, H.; Akagi, Y.; Kayasuga-Kariya, Y.; Chung, U.; Sakai, T. Nonswellable” hydrogel without mechanical hysteresis. Science 2014, 343, 873−875. (9) Tang, L.; Liu, W.; Liu, G. High-strength hydrogels with integrated functions of H-bonding and thermoresponsive surfacemediated reverse transfection and cell detachment. Adv. Mater. 2010, 22, 2652−2656. (10) Bai, T.; Zhang, P.; Han, Y. J.; Liu, Y.; Liu, W. G.; Zhao, X. L.; Lu, W. Construction of an ultrahigh strength hydrogel with excellent fatigue resistance based on strong dipole−dipole interaction. Soft Matter 2011, 7, 2825−2831. (11) Tang, L.; Yang, Y.; Bai, T.; Liu, W. Robust MeO2MA/vinyl-4,6diamino-1,3,5-triazine copolymer hydrogels-mediated reverse gene transfection and thermo-induced cell detachment. Biomaterials 2011, 32, 1943−1949. (12) Wang, N.; Zhang, J.; Sun, L.; Wang, P.; Liu, W. Gene-modified cell detachment on photoresponsive hydrogels strengthened through hydrogen bonding. Acta Biomater. 2014, 10, 2529−2538. (13) Darnell, M. C.; Sun, J. Y.; Mehta, M.; Johnson, C.; Arany, P. R.; Suo, Z.; Mooney, D. J. Performance and biocompatibility of extremely tough alginate/polyacrylamide hydrogels. Biomaterials 2013, 34, 8042−8048. (14) Sun, J. Y.; Zhao, X.; Illeperuma, W. R.; Chaudhuri, O.; Oh, K. H.; Mooney, D. J.; Vlassak, J. J.; Suo, Z. Highly stretchable and tough hydrogels. Nature 2012, 489, 133−136. (15) Zhang, J.; Wang, N.; Liu, W.; Zhao, X.; Lu, W. Intermolecular hydrogen bonding strategy to fabricate mechanically strong hydrogels with high elasticity and fatigue resistance. Soft Matter 2013, 9, 6331− 6337. (16) Liu, L.; Wang, N.; Han, Y. J.; Li, Y. M.; Liu, W. G. Redoxtriggered self-rolling robust hydrogel tubes for cell encapsulation. Macromol. Rapid Commun. 2014, 35, 344−349. (17) Ben-Tal, N.; Sitkoff, D.; Topol, I. A.; Yang, A. S.; Burt, S. K.; Honig, B. Free energy of amide hydrogen bond formation in vacuum, in water, and in liquid alkane solution. J. Phys. Chem. B 1997, 101, 450−457. (18) Maly, K. E.; Dauphin, C.; Wuest, J. D. Self-assembly of columnar mesophases from diaminotriazines. J. Mater. Chem. 2006, 16, 4695−4700. (19) Ziółkowski, B.; Florea, L.; Theobald, J.; Benito-Lopez, F.; Diamond, D. Self-protonating spiropyran-co-NIPAM-co-acrylic acid hydrogel photoactuators. Soft Matter 2013, 9, 8754−8760. (20) Edahiro, J.; Sumaru, K.; Takagi, T.; Shinbo, T.; Kanamori, T.; Sudoh, M. Analysis of photo-induced hydration of a photochromic poly(N-isopropylacrylamide) -spiropyran copolymer thin layer by quartz crystal microbalance. Eur. Polym. J. 2008, 44, 300−307. (21) Joseph, G.; Pichardo, J.; Chen, G. Reversible photo-/ thermoresponsive structured polymer surfaces modified with a

adhesion on an FN-modified hydrogel surface can resist the cytotoxicity to a larger degree. Apart from enhancing cell adhesion, elevating the pDNA concentration in the cell microenvironment is another way to improve gene transfection.33 After the reverse gene transfection was carried out on the FN-modified gel surface for 24 h, another dose of PVDT/pDNA nanocomplexes was added for further transfection for 24 h, which is called sandwich gene transfection.33 The resulting efficiency, as shown in Figure 11, is higher than that achieved with sole FN-modified reverse gene transfection and 3.1 times greater than that of reverse gene transfection without FN modification. The sandwich transfection may allow more of the nanocomplexes to be internalized into the cells, consequently resulting in increased gene expression.

■

CONCLUSIONS A mechanically strong, photosensitive hydrogel was successfully synthesized by the photoinitiated copolymerization of AAm, VDT, and SPAA. The dual hydrogen bonding from DAT− DAT and AAm−AAm resulted in a remarkable enhancement in the tensile and compressive strengths of the hydrogels. The introduction of photoreactive SPAA units enabled the hydrogels to respond to alternate visible−UV light irradiation, bringing about the reversible hydrophobicity−hydrophilicity transition. Accordingly, with the cycled stimulation of visible light and UV irradiation, the cell attachment and detachment could be well repeated. Fibronectin was proven to improve cell adhesion without interfering with light-induced cell detachment. PVDT/pDNA complex nanoparticles could be anchored to the surfaces of hydrogels by hydrogen bonding between diaminotriazine motifs, thereby contributing to efficient reverse gene transfection. Modification with FN combined with sandwich transfection could further enhance the transfection level. Following up with UV irradiation induced the nonharmful release of gene-modified cells for potential tissue engineering applications. The dual hydrogen bonding strategy in conjunction with photoresponsiveness will usher in wide applications of high-strength hydrogels in the biomedical field.

■

ASSOCIATED CONTENT

S Supporting Information *

This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Tel: +86-22-27402487. Fax: +86-22-27404724. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We gratefully acknowledge the support for this work from the National Natural Science Foundation of China (grants 51173129 and 21274105) and National Natural Science Funds for Distinguished Young Scholars (no. 51325305).

■

REFERENCES

(1) Annabi, N.; Tamayol, A.; Uquillas, J. A.; Akbari, M.; Bertassoni, L. E.; Cha, C.; Camci-Unal, G.; Dokmeci, M. R.; Peppas, N. A.; Khademhosseini, A. 25th anniversary article: rational design and 11831

dx.doi.org/10.1021/la502916j | Langmuir 2014, 30, 11823−11832

Langmuir

Article

spirobenzopyran-containing copolymer for tunable wettability. Analyst 2010, 135, 2303−2308. (22) Dattilo, D.; Armelao, L.; Fois, G.; Mistura, G.; Maggini, M. Wetting properties of flat and porous silicon surfaces coated with a spiropyran. Langmuir 2007, 23, 12945−12950. (23) Samanta, S.; Locklin, J. Formation of photochromic spiropyran polymer brushes via surface-initiated, ring-opening metathesis polymerization: reversible photocontrol of wetting behavior and solvent dependent morphology changes. Langmuir 2008, 24, 9558− 9565. (24) Higuchi, A.; Hamamura, A.; Shindo, Y.; Kitamura, H.; Yoon, B. O.; Mori, T.; Uyama, T.; Umezawa, A. Photon-modulated changes of cell attachments on poly(spiropyran-co-methyl methacrylate) membranes. Biomacromolecules 2004, 5, 1770−1774. (25) Alves, N. M.; Pashkuleva, I.; Reis, R. L.; Mano, J. F. Controlling cell behavior through the design of polymer surfaces. Small 2010, 6, 2208−2220. (26) Engler, A. J.; Chan, M.; Boettiger, D.; Schwarzbauer, J. E. A novel mode of cell detachment from fibrillar fibronectin matrix under shear. J. Cell Sci. 2009, 122, 1647−1653. (27) Gallant, N. D.; Michael, K. E.; Garcia, A. J. Cell adhesion strengthening: contributions of adhesive area, integrin binding, and focal adhesion assembly. Mol. Biol. Cell 2005, 16, 4329−4340. (28) Tamura, A.; Kobayashi, J.; Yamato, M.; Okano, T. Thermally responsive microcarriers with optimal poly(N-isopropylacrylamide) grafted density for facilitating cell adhesion/detachment in suspension culture. Acta Biomater. 2012, 8, 3904−3913. (29) Nagase, K.; Watanabe, M.; Kikuchi, A.; Yamato, M.; Okano, T. Thermo-responsive polymer brushes as intelligent biointerfaces: preparation via ATRP and characterization. Macromol. Biosci. 2011, 11, 400−409. (30) Zhou, Y. M.; Ishikawa, A.; Okahashi, R.; Uchida, K.; Nemoto, Y.; Nakayama, M.; Nakayama, Y. Deposition transfection technology using a DNA complex with a thermoresponsive cationic star polymer. J. Controlled Release 2007, 123, 239−246. (31) Cao, Z. Q.; Liu, W. G.; Liang, D. C.; Guo, G.; Zhang, J. Y. Design of poly(vinyldiaminotriazine)-based nonviral vectors via specific hydrogen bonding with nucleic acid base pairs. Adv. Funct. Mater. 2007, 17, 246−252. (32) Li, C. Y.; Yuan, W.; Jiang, H.; Li, J. S.; Xu, F. J.; Yang, W. T.; Ma, J. PCL film surfaces conjugated with P(DMAEMA)/gelatin complexes for improving cell immobilization and gene transfection. Bioconjugate Chem. 2011, 22, 1842−1851. (33) Yuan, W.; Li, C.; Zhao, C.; Sui, C.; Yang, W. T.; Xu, F. J.; Ma, J. Facilitation of gene transfection and cell adhesion by gelatinfunctionalized PCL film surfaces. Adv. Funct. Mater. 2012, 22, 1835−1842.

11832

dx.doi.org/10.1021/la502916j | Langmuir 2014, 30, 11823−11832