Hydrogels Using a Self-Healing Template - MDPI

materials Communication

An Intriguing Method for Fabricating Arbitrarily Shaped “Matreshka” Hydrogels Using a Self-Healing Template Takeshi Sato 1,2,3 , Koichiro Uto 2 , Takao Aoyagi 2 and Mitsuhiro Ebara 1,2,4, * 1 2

3 4

*

Graduate School of Pure and Applied Science, University of Tsukuba, 1-1-1 Tennodai, Tsukuba, Ibaraki 305-8577, Japan; [email protected] International Center for Materials Nanoarchitectonics (WPI-MANA), National Institute for Materials Science (NIMS), 1-1 Namiki, Tsukuba, Ibaraki 305-0044, Japan; [email protected] (K.U.); [email protected] (T.A.) Japan Society for the Promotion of Science (JSPS), 5-3-1, Kojimachi, Chiyoda-ku, Tokyo 102-0083, Japan Graduate School of Industrial Science and Technology, Tokyo University of Science, 6-3-1 Niijuku, Katsushika-ku, Tokyo 125-8585, Japan Correspondence: [email protected]; Tel.: +81-29-851-3354

Academic Editor: Peter Dubruel Received: 30 September 2016; Accepted: 18 October 2016; Published: 25 October 2016

Abstract: This work describes an intriguing strategy for the creation of arbitrarily shaped hydrogels utilizing a self-healing template (SHT). A SHT was loaded with a photo-crosslinkable monomer, PEG diacrylate (PEGDA), and then ultraviolet light (UV) crosslinked after first shaping. The SHT template was removed by simple washing with water, leaving behind the hydrogel in the desired physical shape. A hierarchical 3D structure such as “Matreshka” boxes were successfully prepared by simply repeating the “self-healing” and “photo-irradiation” processes. We have also explored the potential of the SHT system for the manipulation of cells. Keywords: self-healing material; hydrogel fabrication; biocompatible polymer

1. Introduction Polymeric hydrogels are a widely-studied class of biocompatible soft materials that have attracted increasing attention over the last few decades because of their promising applications in broad fields such as food, cosmetics, and biomaterials [1–3]. Hydrogels are typically prepared by the polymerization of monomers and cross-linkers in a mold. The resulting gel shape readily depends on the shape of the mold, with relatively simple shapes such as discs, cubes, cylinders, or spheres [4–7]. However, constructions of complicated, multicomponent self-standing 3D objects with arbitrary shapes have been a big challenge for soft and water-rich hydrogels. New material engineering approaches have to be considered to construct arbitrary shaped gels. In recent years, several sophisticated techniques have been reported for the fabrication of hydrogels with various 3D structures, including photolithography [8,9] and 3D printing [10,11]. Another approach is to attach different blocks or layers after gel fabrication [12,13]. However, these are time consuming and cost-inefficient, and sometimes difficult to produce. Therefore, a stimuli-responsive polymer-based approach has emerged as an alternative method to manufacture complex 3D structures, due to the polymer’s ability change shape in response to external stimuli [14,15]. The idea is that a 3D object is obtained by folding, bending, or twisting a programmed 2D structure that consists of stimuli-responsive materials [16,17]. Here, we demonstrate the ability of self-healing polymers to act as a template of 3D hydrogel structure with arbitrary shapes (self-healing template; SHT). There has been a growing interest in Materials 2016, 9, 864; doi:10.3390/ma9110864

www.mdpi.com/journal/materials

Materials 2016, 9, 864

2 of 9

Here, we demonstrate the ability of self-healing polymers to act as a template of 3D hydrogel 2 ofin 9 shapes (self-healing template; SHT). There has been a growing interest dynamically-reconstructing or self-healing polymers in recent years because they can undergo automatic healing to repair damage, while stimuli-responsive polymers were designed to function dynamically-reconstructing or self-healing polymers in recent years because they can undergo automatic as passive structures. Intrinsic self-healing polymers [18,19], which have a dynamic dissociation healing to repair damage, while stimuli-responsive polymers were designed to function as passive and re-association of bonds, are especially an increasingly active research area, because they do not structures. Intrinsic self-healing polymers [18,19], which have a dynamic dissociation and re-association require any curing or adhesive agents. Such dynamic bonds can be accomplished through selective of bonds, are especially an increasingly active research area, because they do not require any curing or or specific interactions of two complementary compounds, called host–guest interactions [20,21]. adhesive agents. Such dynamic bonds can be accomplished through selective or specific interactions Recently, metal–ligand interactions [22] have also gained much attention because they are not only of two complementary compounds, called host–guest interactions [20,21]. Recently, metal–ligand thermodynamically stable, but also kinetically labile. Self-healing hydrogels at room temperature, interactions [22] have also gained much attention because they are not only thermodynamically for example, have been demonstrated using ditopic ligands and lanthanide ions [23]. The stable, but also kinetically labile. Self-healing hydrogels at room temperature, for example, have been combination of catechol–Fe3+ bonds have also been known to reconstruct bonds spontaneously [24]. demonstrated using ditopic ligands and lanthanide ions [23]. The combination of catechol–Fe3+ bonds We have recently reported bio-inspired metallo-supramolecular hydrogels using phosphate-terminated have also been known to reconstruct bonds spontaneously [24]. We have recently reported bio-inspired poly(ethylene glycol) (PEG-phos) and various trivalent metal ions, such as Fe3+, V3+, Al3+, Ti3+, and metallo-supramolecular hydrogels using phosphate-terminated poly(ethylene glycol) (PEG-phos) and Ga3+ [25]. In this study, we report a new hydrogel fabrication method utilizing our previously various trivalent metal ions, such as Fe3+ , V3+ , Al3+ , Ti3+ , and Ga3+ [25]. In this study, we report a new developed self-healing hydrogel as a template. hydrogel fabrication method utilizing our previously developed self-healing hydrogel as a template. Materials 2016, 9, 864 structure with arbitrary

2. Results 2. Resultsand andDiscussion Discussion Template (SHT) (SHT) 2.1. Preparation of Self-Healing Template work, arbitrarily shaped hydrogels were fabricated fabricated from photo-crosslinkable photo-crosslinkable In the present work, polymers using usingthe the PEG-phos as SHT. the The SHT.SHT Thewas SHT first using prepared using metal–ligand polymers PEG-phos as the firstwas prepared metal–ligand interactions interactions between trivalent metal ions and four-arm PEG-phos in the presence of a between trivalent metal ions and four-arm PEG-phos in the presence of a photo-crosslinkable photo-crosslinkable solution, in which PEG-diacrylate (PEGDA) and the photo-initiator solution, in which PEG-diacrylate (PEGDA) and the photo-initiator irgacure 2959 wereirgacure added 2959 were (Figure 1A and S1).was Next, the SHT was a certain shapeand by (Figure 1A added and Scheme S1). Next,Scheme the SHT fabricated into fabricated a certain into shape by folding folding and bending (Figure 1B). The arbitrarily-shaped SHT was then exposed to UV irradiation to bending (Figure 1B). The arbitrarily-shaped SHT was then exposed to UV irradiation to crosslink the crosslink the loaded PEGDA and fix shapes the arbitrary (Figurethe 1C). Finally, the gel loaded PEGDA and fix the arbitrary (Figureshapes 1C). Finally, crosslinked gelcrosslinked samples were samples were transferred into water to remove the SHT from the PEGDA gel (Figure 1D). In the transferred into water to remove the SHT from PEGDA gel (Figure 1D). In the present study, 3+, Ti3+, and Ga3+ were examined as 3+ , and present trivalent study, various trivalent metal such as Ga Fe3+3+, V various metal ions, such as Fe3+ ,ions, V3+ , Ti were examined as crosslinkers. All ions crosslinkers.gelated All ions successfully gelated PEG-phos; theaffected gelation time was successfully the PEG-phos; however, the the gelation time washowever, significantly by ion species. 3+ 3+ 3+ significantly affected by ion species. For example, gelation quickly occurred within seconds for Fe For example, gelation quickly occurred within seconds for Fe , Ti , and Ga , while it took 403+s, 3+, while it took 40 s for V3+, as previously reported [25]. These phenomena have been 3+ , asGa Ti3+,Vand for previously reported [25]. These phenomena have been explained by their coulomb potential explained their coulomb rates potential values and water substitution [25].anIn this study, values and by water substitution [25]. In this study, we have chosen V3+rates because optimal periodwe of 3+ because an optimal period of time is required for shaping 3+ have chosen V and manipulating the time is required for shaping and manipulating the hydrogels. In addition, V is relatively transparent hydrogels. In addition, V3+ ismetal relatively to UV compared with other ionstransparent (Figure S1).to UV compared with other metal ions (Figure S1).

Figure 1. 1. Fabrication Figure Fabrication processes processes of of 3D 3D hydrogel hydrogel object object using using self-healing self-healing template template (SHT). (SHT). (A) (A) The The SHT SHT was first prepared using metal–ligand interactions between trivalent metal ions and four-arm PEG-phos was first prepared using metal–ligand interactions between trivalent metal ions and four-arm PEG-phos in the the presence presence of ofPEGDA; PEGDA;(B) (B)The TheSHT SHTwas wasfabricated fabricatedinto intoa acertain certain shape folding and bending in shape byby folding and bending or or assembling each piece; (C) The arbitrarily-shaped SHT was then exposed to UV irradiation to assembling each piece; (C) The arbitrarily-shaped SHT was then exposed to UV irradiation to crosslink crosslink the PEGDA; (D) Finally, PEG-phos and metal ions were extracted from the PEGDA gel. the PEGDA; (D) Finally, PEG-phos and metal ions were extracted from the PEGDA gel.


3 of 9

2.2. Evaluation of Self-Healing Effect on Mechanical Property of Photo-Crosslinked Hydrogels Figure 2A shows the dynamic fluidic nature of the PEG-phos SHT at room temperature, Materialsby 2016, 9, 864 microscopy. The scratch gradually disappeared with time, suggesting 3 of 9 as observed optical that the SHT behaves as a liquid-like substrate. However, the SHT did not deform its shape during UV 2.2. Evaluation of Self-Healing Effect on Mechanical Property of Photo-Crosslinked Hydrogels irradiation because of the high viscosity of the SHT. To investigate the time-dependent self-healing Figure 2A shows thewere dynamic fluidic together nature of the PEG-phos room temperature, as with process, two pieces of SHT attached (Figure 2B).SHT Oneatof them was stained observed by optical microscopy. The scratch gradually disappeared with time, suggesting that the Fluorescein isothiocyanate-dextran (FITC-dextran; Mw = 3000–5000). After the interfaces came into SHT behaves as a liquid-like substrate. However, the SHT did not deform its shape during UV contact irradiation with eachbecause other, of FITC-dextran started migrate to the other piece of the self-healing SHT. This means the high viscosity of theto SHT. To investigate the time-dependent that diffusion and reformation of metal–ligand interactions successfully occurred at the interface. process, two pieces of SHT were attached together (Figure 2B). One of them was stained with Fluorescein isothiocyanate-dextran Mw =strength 3000–5000). the interfaces into testing To quantify the effect of healing time(FITC-dextran; on the adhesion ofAfter the joint surface,came tensile contactwere with each other, FITC-dextran started migrateslab to the piece ofThe the SHT. This means experiments performed using virgin andtohealed gelother samples. virgin gel samples of that diffusion and reformation of metal–ligand interactions successfully occurred at the interface. SHT containing PEGDA were cut in the middle, and then the two halves were merged together. To quantify the effect of healing time on the adhesion strength of the joint surface, tensile testing After standing for 30, 60, 90, and 180 min, they were exposed to UV light for 10 min; 15 mW·cm−2 experiments were performed using virgin and healed slab gel samples. The virgin gel samples of of intensity. The adhesive measured bythen a loading-to-failure (Figure 2C). SHT containing PEGDAstrength were cutwas in the middle, and the two halves weretensile mergedtest together. The adhesive strength gels withexposed time and reached as−2virgin gel After standing forof 30,PEGDA 60, 90, and 180increased min, they were to UV light forthe 10 same min; 15value mW·cm of adhesive was measured bystrength a loading-to-failure tensile test (Figure sample.intensity. Thus, itThe was foundstrength that the mechanical of a gel was controlled by2C). theThe diffusion strength of PEGDA gels increased withoftime and reached the same of value as virginSHT gel gel in time of adhesive the PEGDA. Figure 2D shows the results an erosion experiment PEG-phos sample. Thus, it was found that the mechanical strength of a gel was controlled by the diffusion water. The SHT swelled rapidly, approaching a plateau after 30 min. Then, the SHT started to lose its time of the PEGDA. Figure 2D shows the results of an erosion experiment of PEG-phos SHT gel in weight steadily forSHT up swelled to 3 h incubation time. At a4 plateau h, the SHT in water. Although water. The rapidly, approaching after completely 30 min. Then,dissolved the SHT started to lose it should more time be extracted from PEGDA hydrogels, this result its take weight steadily forfor up SHT to 3 hto incubation time. At 4 h,crosslinked the SHT completely dissolved in water. Although it should take more time for SHT to between be extracted from crosslinkedcan PEGDA hydrogels, this suggested that the metal–ligand interactions V3+ -PEG-phos be easily dissociated by 3+-PEG-phos can be easily dissociated result suggested that the metal–ligand between dilution. To determine whether the SHTinteractions can be applied forVcreation of multicomponent hydrogels, by dilution. To determine whether the SHT can be applied for creation of multicomponent two types of SHT gels containing PEGDA with different molecular weights (Mn = 3350 and 10,000) hydrogels, two types of SHT gels containing PEGDA with different molecular weights (Mn = 3350 were prepared. The two gels were held in contact with each other and incubated for 180 min and then and 10,000) were prepared. The two gels were held in contact with each other and incubated for 180 exposedmin to UV lightexposed for 10 min. hydrogel washydrogel immersed water for 24 h.for 24 h. and then to UVThe lighthealed for 10 min. The healed was in immersed in water

2. (A) Microscopic observation of the fluidic behavior of SHT (scale bar = 200 μm). Time-lapse Figure Figure 2. (A) Microscopic observation of the fluidic behavior of SHT (scale bar = 200 µm). images show the capability of scratch repair; (B) Two SHT samples were prepared, and each sample Time-lapse images show the capability of scratch repair; (B) Two SHT samples were prepared, and each was cut into two pieces. One of them was colored with Fluorescein isothiocyanate-dextran sample (FITC-dextran) was cut into for twoclarity. pieces. One of them was colored with Fluorescein isothiocyanate-dextran After pressing the fractured surfaces together, they merged into a single (FITC-dextran) forbar clarity. pressing the fractured surfaces together, merged into a single piece (scale = 1 cm);After (C) Effect of self-healing time before UV irradiation on they the adhesive strength of thebar resulting PEGDA gels after irradiationtime (n = 3,before p < 0.05); Erosion behavior SHT (scalestrength piece (scale = 1 cm); (C) Effect of UV self-healing UV(D) irradiation on theofadhesive bar = 1 cm).PEGDA Mass changes at the predetermined (n = 3, p(D) < 0.05). of the resulting gels were afterplotted UV irradiation (n = 3, time p < 0.05); Erosion behavior of SHT (scale bar = 1 cm). Mass changes were plotted at the predetermined time (n = 3, p < 0.05).

Materials 2016, 9, 864 Materials 2016, 9, 864

4 of 9 4 of 9

2.3. Arbitrary Hydrogel Preparation SHT 2.3. Arbitrary Hydrogel Preparationwith withthe theUse Use of SHT As shown Figure 3A, 3A, aaswelling mismatch of theofgels observed. This mismatch induces As shown ininFigure swelling mismatch thewas gels was observed. This mismatch a large internal stress,stress, because the large region region experiences a compression from the from induces a large internal because the swelling large swelling experiences a compression less-swelling region. thethe less-swelling region experiences an extension from from the the the less-swelling region.Oppositely, Oppositely, less-swelling region experiences an extension large-swelling region. However, the adhered gels were not broken, and the healed interface did not large-swelling region. However, the adhered gels were not broken, and the healed interface did not detach. We have also prepared a mosaic hydrogel using three types of SHT gels containing PEGDA detach. We have also prepared a mosaic hydrogel using three types of SHT gels containing PEGDA with different molecular weights (Mn = 3350, 6000, and 10,000). The samples with Mn of 3350 and with6000 different molecular weights (Mn = 3350, 6000, and 10,000). The samples with Mn of 3350 and were stained with blue (methylene blue) and red (methyl red), respectively. The preparation of 6000the were stained with blue (methylene andway red as (methyl red), preparation of mosaic hydrogel was performed in blue) the same in Figure 3A.respectively. The interfacesThe of each piece the mosaic hydrogel was performed in the same way as in Figure 3A. The interfaces of each piece did did not detach by the swelling mismatch (Figure 3B). SEM images also confirm there was no not detach by at thethe swelling (Figure3C). 3B).These SEMresults images also confirm was nocan fracturing fracturing adhesivemismatch interface (Figure indicate that thethere SHT system be utilized to mend two or more 3C). different types of hydrogels create In at the adhesive interface (Figure These results indicate to that the multicomponent SHT system canobjects. be utilized to addition to the 2D multicomponent hydrogels, the SHT system can be used to fabricate mend two or more different types of hydrogels to create multicomponent objects. In addition to the 2D sophisticated 3D objects. the SHT system can be used to fabricate sophisticated 3D objects. multicomponent hydrogels,

Figure 3. (A) Photographs of two SHT samples containing different molecular weights of PEGDA

Figure 3. (A) Photographs of two SHT samples containing different molecular weights of PEGDA (Mn = 3350 and 10,000). After cutting into two pieces and pressing the fractured surfaces together for (Mn = 3350 and 10,000). After cutting into two pieces and pressing the fractured surfaces together 3 h, PEGDA was crosslinked by UV irradiation; (B) Photographs of a mosaic-type hydrogel. Three for 3SHT h, PEGDA was crosslinked UV irradiation; (B) Photographs a mosaic-type hydrogel. samples containing differentby molecular weights of PEGDA (Mn = 3350,of 6000, and 10,000). After Threeattaching SHT samples containing different molecular weights of PEGDA (Mn = 3350, 6000, and 10,000). these pieces together, PEGDA was crosslinked by UV irradiation (scale bar = 10 mm); (C) AfterSEM attaching these pieces together, PEGDA was crosslinked by UV irradiation (scale bar = 10 mm); image of adhesive interfaces in the mosaic hydrogel (scale bar = 1 mm). (C) SEM image of adhesive interfaces in the mosaic hydrogel (scale bar = 1 mm). We used “paper folding” or “origami” technology to create cubic boxes. First, we prepared the 2D planar figure for a 3D cube using a SHT with PEGDA (Figure 4A). Then, the pre-patterned We used “paper folding” or “origami” technology to create cubic boxes. First, we prepared the figure was folded to make a cubic hydrogel (Figure 4B). After the self-healing process to seal the 2D planar figure for a 3D cube using a SHT with PEGDA (Figure 4A). Then, the pre-patterned figure defect areas, the gel was photo-irradiated. SEM images show that the adhesive interfaces were was tightly folded sealed to make a cubic hydrogel (Figure 4B). After the self-healing process to seal the defect (Figure 4C). To test the air-tightness, the gel box was put into water. The gel box areas,floated the gel photo-irradiated. SEM images that the adhesive interfaces were tightly sealed in was water, and no air or water leakage wasshow observed.

(Figure 4C). To test the air-tightness, the gel box was put into water. The gel box floated in water, 2.4.air Biomedical Applications of Arbitrarily-Shaped Gels. and no or water leakage was observed. To test the applicability of the SHT system, we encapsulated an ant in the box and placed the gel into hexane. After five minutes, the ant was taken out of the box and found to be alive (Figure 4D). We have created structures, such as an nesting To test theadditionally applicability of the hierarchical SHT system, we encapsulated ant in“Matreshka” the box andboxes, placedby the gel repeatingAfter the self-healing and photo-irradiation processes. Figure shows the to photographs of the 4D). into hexane. five minutes, the ant was taken out of the box4Eand found be alive (Figure PEGDA hydrogel boxes before and afterstructures, nesting. Because a facile but“Matreshka” tight fixation boxes, of the adhered We have additionally created hierarchical such as nesting by repeating

2.4. Biomedical Applications of Arbitrarily-Shaped Gels

the self-healing and photo-irradiation processes. Figure 4E shows the photographs of the PEGDA hydrogel boxes before and after nesting. Because a facile but tight fixation of the adhered interfaces is considered to be a key point in the creation of hierarchical 3D objects, this SHT system identifies


5 of 9


5 of 9

interfaces is considered to be a key point in the creation of hierarchical 3D objects, this SHT system identifies a solution Materials 2016, 9, 864 in the field of spatially arranged hydrogel fabrications. Finally, the encapsulated 5 of 9 usinginthe prepared by the SHT system was demonstrated. Suspension ofcells NIH 3T3 acells solution thePEGDA field ofbox spatially arranged hydrogel fabrications. Finally, the encapsulated using interfaces is considered to 1.0 be key in the creation of hierarchical objects, this SHT −1 was fibroblasts with aprepared density of 106 point cells mL placed in theSuspension box.3DThe box was then immersed the PEGDA box by thea×SHT system was demonstrated. of NIH 3T3system fibroblasts identifies a solution in the field of − spatially arranged hydrogel fabrications. Finally, the encapsulated 6 cells 1 was in acidic solution (HCl 0.01 M) mL for 10 min placed (Figurein5A). cells immersed were collected and with a density of 1.0 × 10 the The box. encapsulated The box was then in acidic cells using the PEGDA box prepared by the SHT system was demonstrated. Suspension of NIH 3T3 reseeded(HCl on tissue polystyrene (TCPS) dishes. A live/dead assay of the collected cells solution 0.01 M)culture for 10 min (Figure 5A). The encapsulated cells were collected and reseeded on fibroblasts with a density of 1.0 × 106 cells mL−1 was placed in the box. The box was then immersed showed that almost all the cells were alive, whereas very few cells were alive when they were tissue culture polystyrene (TCPS) dishes. A live/dead assay of the collected cells showed that almost in acidic solution (HCl 0.01 M) for 10 min (Figure 5A). The encapsulated cells were collected and directly injected the HCl solution without the dishes. gel box and Figure S2). all thereseeded cells were alive, whereas very few cells were alive when they 5B were directly injected intocells the HCl on into tissue culture polystyrene (TCPS) A(Figure live/dead assay of the collected An alamar blue (AB) assay was also performed to determine the proliferation of the collected solution without the gel box (Figure 5B and Figure S2). showed that almost all the cells were alive, whereas very few cells were alive when they were cells.An The collected cells were reseeded on TCPS and The alamar blue into (AB) assay alsowithout performed to dishes, determine the proliferation of theobserved. collected cells. directly injected the HCl was solution the gel box (Figure 5Bproliferation and Figure S2).was An alamar blue (AB) was alsothat performed determine thewas proliferation the collected proliferation profile didreseeded not assay differ from of control cells that were directlyofinjected into cell The collected cells were on TCPS dishes, andtoproliferation observed. The proliferation cells. The cellsthat were on medium, TCPS dishes, and proliferation was observed. The culture medium (Dulbecco’s Modified Eagle’s DMEM) without gel box (Figuremedium 5C). As profile did not collected differ from of reseeded control cells that were directly injectedthe into cell culture proliferation profile did not differ from that of control cells that were directly injected into cell expected, HCl treated cells without the gel box did not proliferate the acute of the (Dulbecco’s Modified Eagle’s medium, DMEM) without the gel due box to (Figure 5C).toxicity As expected, culture medium (Dulbecco’s Modified Eagle’s medium, DMEM) without the gel box (Figure 5C). As acidic conditions. Figurethe 5Dgel shows time-dependent in acute pH within the when it was HCl treated cells without box did not proliferatechanges due to the toxicity of gel the box acidic conditions. expected, HCl treated cells without the gel box did not proliferate due to the acute toxicity of the immersed in HCltime-dependent solution. Although the pH gradually decreased to itthe diffusion of in proton Figure 5D shows changes in pH within the gel box due when was immersed HCl acidic conditions. Figure 5D shows time-dependent changes in pH within the gel box when it was ions through the the pHgradually value wasdecreased still 10 min. dueoftoproton solution. Although the pH due to for the diffusion ions through the gel, immersed in gel, HCl solution. Although the maintained pH gradually decreased the diffusion of proton the pH value was still maintained for 10 min. ions through the gel, the pH value was still maintained for 10 min.

Figure 4. A (A) A photograph a 2Dplanar planarfigure figure prepared prepared from forfor a 3D3D gel gel cube (scale bar =bar 1 =1 Figure 4.4.(A) of of a 2D fromSHT SHT cube (scale Figure (A) photograph A photograph of a 2D planar figure prepared froma SHT for a 3D gel cube cm); (B) A photograph of a cubic hydrogel box prepared by folding the pre-patterned 2D figure cm); (B) of a cubic hydrogel box prepared folding by thefolding pre-patterned 2D figure (scale barA= 1photograph cm); (B) A photograph of a cubic hydrogel boxby prepared the pre-patterned (scale bar = 1 cm); (C) SEM images of the adhered interface of the cubic hydrogel box (scale bar = 1 (scale bar = 1 cm); (C) SEM images of the adhered interface of the cubic hydrogel box (scale bar = 1 2D figure (scale bar =(D) 1 cm); (C) SEM the adhered interface of the(top). cubic mm and 50 mm); Photographs of images the cubicofhydrogel box floated in hexane Anhydrogel ant was box mm and 50 mm); (D) Photographs of the cubic hydrogel box floated in hexane (top). An ant was (scaleencapsulated bar = 1 mm in and mm); of the floated hexane the50 box. After(D) fivePhotographs minutes, the ant wascubic taken hydrogel out of the box box and was in found to be(top). encapsulated in the(scale box.bar After five minutes, antofwas taken out of thebefore box and was to be An ant was(bottom) encapsulated in the five the minutes, the was taken of the box andfound was found alive = 1box. cm);After (E) Photographs the ant hydrogel boxesout and after nesting alive (bottom) (scale bar = 1 cm); (E) Photographs of the hydrogel boxes before and after nesting to be (scale alive (bottom) (scale bar = 1 cm); (E) Photographs of the hydrogel boxes before and after nesting bar = 1 cm). Each box was stained with blue (methylene blue) or red (methyl red) for clarity. (scale bar bar = = 11 cm). (methyl red) red) for for clarity. clarity. (scale cm). Each Each box box was was stained stained with with blue blue (methylene (methylene blue) blue) or or red red (methyl

Figure 5. Cont.

Materials 2016, 9, 864 Materials 2016, 9, 864

6 of 9 6 of 9

Figure Figure5.5.(A) (A)Digital Digitalphotographs photographsofofNIH NIH3T3 3T3cells cellsininDulbecco’s Dulbecco’sModified ModifiedEagle’s Eagle’smedium medium(DMEM), (DMEM), HCl HCl (0.01 (0.01 M), M),and and in in aa gel gel box box floating floating in in HCl HCl (0.01 (0.01 M); M); (B) (B) Live Live and and dead dead cell cell assay assay after after each each treatment treatment(red; (red;dead, dead,green; green;alive) alive)(scale (scalebars bars==200 200μm). µm). The Theencapsulated encapsulatedcells cellswere werecollected collectedand and reseededon ontissue tissueculture culturepolystyrene polystyrene(TCPS) (TCPS) dishes; Alamar blue assay to assess reseeded dishes; (C)(C) Alamar blue assay waswas usedused to assess the the proliferation of the collected after each treatment. Data arepresented presentedasasmean mean ±±standard standard proliferation of the collected cellscells after each treatment. Data are deviation(n (n==22per percondition); condition);(D) (D)Time-dependent Time-dependentpH pHchanges changesof ofthe thecell cellsuspended suspendedHCl HClsolution solution deviation without(open) (open)and andwith with(closed) (closed)gel gelbox. box.The Thegel gelboxes boxeswere wereself-healed self-healedfor for00(circle), (circle),60 60(triangle), (triangle), without and180 180(square) (square)min minprior priorto toUV UVcrosslink. crosslink. and

3.3.Methods Methods 3.1.Materials Materials 3.1. Four-arm poly(ethylene poly(ethyleneglycol) glycol)(4-arm (4-arm PEG) = 40,000) and linear PEG= 6000) (Mn =were 6000) Four-arm PEG) (Mn(Mn = 40,000) and linear PEG (Mn were provided by NOF Co., Ltd. (Tokyo, Japan) and purified by precipitations in hexane. provided by NOF Co., Ltd. (Tokyo, Japan) and purified by precipitations in hexane. 2-Hydroxy-40 -(2-hydroxyehoxy)-2-methylpropiophenone(irgacure (irgacure 2959) 2959) and andlinear linearPEG PEG(Mn (Mn==3350 3350 2-Hydroxy-4′-(2-hydroxyehoxy)-2-methylpropiophenone and 10,000) 10,000) were Co., LLC. (St.(St. Louis, MO,MO, USA). Tetrahydrofuran (THF) and were purchased purchasedfrom fromSigma-Aldrich Sigma-Aldrich Co., LLC. Louis, USA). Tetrahydrofuran ultradehydrated, diisopropylamine, titanium(III) chloride solution iron(III) chloride hexahydrate, (THF) ultradehydrated, diisopropylamine, titanium(III) chloride(20%), solution (20%), iron(III) chloride and phosphoryl were purchased frompurchased Wako Purefrom Chemical Industries Ltd. (Osaka, Japan)Ltd. and hexahydrate, andchloride phosphoryl chloride were Wako Pure Chemical Industries used as Japan) received. Acryloyl wasAcryloyl purchased from Tokyo Co., Ltd. (Tokyo, (Osaka, and used aschloride received. chloride was Chemical purchasedIndustry from Tokyo Chemical Japan) and used as(Tokyo, received.Japan) Vanadium(III) hexahydrate was purchased Thermo Fisher Industry Co., Ltd. and usedchloride as received. Vanadium(III) chloridefrom hexahydrate was Scientific Chemicals Inc. (Waltham, MA, USA)Chemicals and used asInc. received. purchased from Thermo Fisher Scientific (Waltham, MA, USA) and used as received. 3.2. Polymer Synthesis

3.2. Polymer Synthesis of terminal phosphorylated four-arm PEG (4-arm PEG-phos) was carried out The preparation as follows. Four-armof PEG (Mn =phosphorylated 40,000) with hydroxyl endPEG group was dissolved 300carried mL of THF The preparation terminal four-arm (4-arm PEG-phos)inwas out (1.67% w/v). POCl was dissolved in 200 mL of super dehydrated THF (5% v/v), and the as follows. Four-arm3 PEG (Mn = 40,000) with hydroxyl end group was dissolved in 300 mL solution of THF was kept at ca. 0 ◦3 Cwas with an iced bath. solution was thenTHF added the POCl solution. (1.67% w/v). POCl dissolved in 200The mLPEG of super dehydrated (5%into v/v), and the3 solution Diisopropyl amine was also added to the PEG and POCl mixture to remove the generated HCl. 3 was kept at ca. 0 °C with an iced bath. The PEG solution was then added into the POCl3 solution. The mixture was then stirred at room temperature for 24 h. After the reaction, THF was totally Diisopropyl amine was also added to the PEG and POCl3 mixture to remove the generated HCl. evaporated rotary evaporator, and the residue for was24 dissolved 200reaction, mL of water. The mixtureusing was athen stirred at room temperature h. Afterinthe THFThis wasaqueous totally solution was then dialyzed for 3 days against water using a dialysis membrane (molecular weight cut evaporated using a rotary evaporator, and the residue was dissolved in 200 mL of water. This off = 3500). The dialyzed aqueous solution was against then lyophilized to obtain phosphate-terminated PEG aqueous solution was then dialyzed for 3 days water using a dialysis membrane (molecular as a white powder. weight cut off = 3500). The dialyzed aqueous solution was then lyophilized to obtain

phosphate-terminated PEG as a white powder. 3.3. Preparation of Poly(ethylene glycol) Diacrylate (PEGDA) 3.3. Preparation of Poly(ethylene glycol) Diacrylate (PEGDA) Linear poly(ethylene glycol) (Mn = 3350, 6000, or 10,000, 1 eq. mole amount) was dissolved in 50 mL of THF. Acryloyl chloride of 132.5 eq. molar relative to the terminal group of poly(ethylene Linear poly(ethylene glycol) (Mn = 3350, 6000, or 10,000, 1 eq. mole amount) was dissolved in glycol) was diluted with dichloromethane. The acryloyl chloride was slowly added into the PEG 50 mL of THF. Acryloyl chloride of 132.5 eq. molar relative to the terminal group of poly(ethylene glycol) was diluted with dichloromethane. The acryloyl chloride was slowly added into the PEG solution by cooling with an ice bath. The reaction solution was gently bubbled by N2 gas overnight


7 of 9

solution by cooling with an ice bath. The reaction solution was gently bubbled by N2 gas overnight to remove HCl from the solution. After the reaction, the excess amount of acryloyl chloride was removed by precipitating into diethyl ether three times. A white precipitate was collected by vacuum filtration and then dried in vacuo. 3.4. Hydrogel Preparation Using Self-Healing Templates (SHTs) The 4-arm PEG-phos and PEGDA were dissolved in an irgacure 2959 (10 wt %) solution at concentrations of 41.3 g/L and 125 g/L, respectively. A 0.3 M vanadium chloride aqueous solution was prepared, and PEGs solution (800 µL) and vanadium chloride solution (50 µL) were mixed together. The mixed solution was immediately poured into a mold that was made from a glass slide (50 mm × 50 mm) with silicone rubber of 1 mm thickness. The solution was then incubated for a few minutes to cross-link the self-healing hydrogel network. 3.5. Optical Analysis of Metal Ion Solutions Aqueous solutions of titanium(III) chloride, iron(III) chloride hexahydrate, and vanadium(III) chloride hexahydrate at a concentration of 8.3 × 10−3 M were prepared. Absorbance of these solutions was characterized from 300 to 600 nm using a Jasco V-650 spectrophotometer (Jasco Co., Tokyo, Japan). 3.6. Tensile Tests Dumbbell-shaped SHTs were cut in half using a razor blade. Two pieces were pushed together so that their surfaces came into contact with each other. After standing for a predetermined time (0, 30, 60, 90, and 180 min), the samples were exposed to UV irradiation (using Optical Modulex SX-U1251HQ, Ushio, Tokyo, Japan) to cross-link the PEGDA gel. The gel was immersed in water for 1 day to dissolve the self-healing network. The gel samples were subjected to tensile tests utilizing a tensile testing machine (EZ-S 500N, Shimadzu, Kyoto, Japan). These samples underwent tensile tests at 3 mm/min until the gel specimens fractured. 3.7. Preparation of Arbitrarily-Shaped Hydrogels Prepared SHTs were cut, folded, and then attached to form a predetermined shape. The interfaces underwent the self-healing process for an appropriate amount of time, typically 60 min. The self-healed SHTs were exposed to UV light (15 mW·cm−2 ) for 10 min in order to form photo cross-linked hydrogel networks. Photo cross-linked networks were then immersed in water for at least one day to remove V3+ ions. 3.8. Cell Culture Suspension of NIH 3T3 fibroblasts with a density of 1.0 × 106 cells mL−1 were put in a box (1 cm3 )-shaped gel with an injection pump, and then the gel was placed in 4 mL of HCl (0.01 M) for 10 min. The encapsulated cells were collected and reseeded on tissue culture polystyrene (TCPS) dishes in Dulbecco’s Modified Eagle’s medium (DMEM) in the presence of 10% fetal bovine serum (FBS) at 37 ◦ C for 72 h. A live/dead assay was performed to determine the number of viable and non-viable cells. The collected cells were treated with 500 µL of 2 µM calcein AM (positive) and 4 µM EthD-1 (negative) solution for 30 min at room temperature, and then observed by fluorescence microscopy (IX71, Olympus, Tokyo, Japan). Calcein AM and EthD-1 produced green and red fluorescence at 488 nm and 543 nm, respectively. For the evaluation of cell viability, the cells underwent an alamar blue (AB) assay after 3 h, 24 h, 48 h, and 72 h. As negative and positive control, the same type of cells were treated by 4 mL of 0.01 M HCl for 10 min and DMEM containing 10% FBS, respectively. The collected cells were treated with an AB solution. The supernatants were placed into a 96-well micro plate, and the absorbance was measured at 590 nm using a micro-plate reader (Bio-Rad Laboratories).


8 of 9

4. Conclusions A self-healing template-based approach was successfully demonstrated for the creation of arbitrarily-shaped hydrogels. By using this technique, multicomponent 2D gels were successfully prepared. In addition, it has also been applied to fabricate sophisticated 3D objects, such as “Matreshka” boxes. The prepared hydrogels showed tight sealing of the adhesive interfaces without the use of sutures. This approach will provide a robust and facile method for the manipulation and delivery of living cells as well as the formation of tissues mimicking native tissue constructs. Supplementary Materials: The following are available online at www.mdpi.com/1996-1944/9/11/864/s1. Scheme S1 shows schematic illustration for synthesis of 4-arm PEG-phos and PEGDA. Figure S1 indicates the result of absorbance spectra of FeCl3 , TiCl3 and VCl3 aqueous solution (8.3 × 10−3 M each). FeCl3 solution shows the absorbance wavelength of 300–400 nm. This wavelength range is utilized for UV crosslinking of PEGDA gel. Figure S2 shows the result of Cell viabilities obtained from live/dead assays. Acknowledgments: This work was supported by a Grant-in-Aid for JSPS Fellows KAKENHI (Grant Number 265612). We are thankful to RIKEN-cell bank for providing the cells. Author Contributions: Takeshi Sato, Takao Aoyagi and Mitsuhiro Ebara conceived and developed the hydrogel fabrication methods; Koichiro Uto performed cell experiments; all authors analyzed and discussed the data; all authors wrote the paper. Conflicts of Interest: The authors declare that they have no conflict of interest.

References 1. 2. 3. 4.

5. 6. 7. 8. 9.

10. 11.

12. 13. 14.

Shewan, H.M.; Stokes, J.R. Review of techniques to manufacture micro-hydrogel particles for the food industry and their applications. J. Food Eng. 2013, 119, 781–792. [CrossRef] George, M.; Abraham, T.E. pH sensitive alginate-guar gum hydrogel for the controlled delivery of protein drugs. Int. J. Pharm. 2007, 335, 123–129. [CrossRef] [PubMed] Lee, K.Y.; Mooney, D.J. Hydrogels for Tissue Engineering. Chem. Rev. 2001, 101, 1869–1879. [CrossRef] [PubMed] Makino, K.; Hiyoshi, J.; Ohshima, H. Effects of thermosensitivity of poly (N-isopropylacrylamide) hydrogel upon the duration of a lag phase at the beginning of drug release from the hydrogel. Colloids Surf. B Biointerfaces 2001, 20, 341–346. [CrossRef] Odake, S.; Hatae, K.; Shimada, A.; Iibuchi, S. Apparent Diffusion Coefficient of Sodum Chloride in Cubical Agar Gel. Agric. Biol. Chem. 1990, 54, 2811–2817. Wang, T.; Zheng, S.; Sun, W.; Liu, X.; Fu, S.; Tong, Z. Notch insensitive and self-healing PNIPAm–PAM–clay nanocomposite hydrogels. Soft Matter 2014, 10, 3506–3512. [CrossRef] [PubMed] Park, T.G.; Hoffman, A.S. Immobilization and characterization of β-galactosidase in thermally reversible hydrogel beads. J. Biomed. Mater. Res. 1990, 24, 21–38. [CrossRef] [PubMed] Chan, V.; Zorlutuna, P.; Jeong, J.H.; Kong, H.; Bashir, R. Three-dimensional photopatterning of hydrogels using stereolithography for long-term cell encapsulation. Lab Chip 2010, 10, 2062–2070. [CrossRef] [PubMed] Hwang, D.K.; Oakey, J.; Toner, M.; Arthur, J.A.; Anseth, K.S.; Lee, S.; Zeiger, A.; Van Vliet, K.J.; Doyle, P.S. Stop-Flow Lithography for the Production of Shape-Evolving Degradable Microgel Particles. J. Am. Chem. Soc. 2009, 131, 4499–4504. [CrossRef] [PubMed] Pataky, K.; Braschler, T.; Negro, A.; Renaud, P.; Lutolf, M.P.; Brugger, J. Microdrop Printing of Hydrogel Bioinks into 3D Tissue-Like Geometries. Adv. Mater. 2012, 24, 391–396. [CrossRef] [PubMed] Li, C.; Faulkner-Jones, A.; Dun, A.R.; Jin, J.; Chen, P.; Xing, Y.; Yang, Z.; Li, Z.; Shu, W.; Liu, D. Rapid Formation of a Supramolecular Polypeptide–DNA Hydrogel for In Situ Three-Dimensional Multilayer Bioprinting. Angew. Chem. Int. Ed. 2015, 54, 3959–3961. [CrossRef] [PubMed] Asoh, T.-A.; Kikuchi, A. Electrophoretic adhesion of stimuli-responsive hydrogels. Chem. Commun. 2010, 46, 7793–7795. [CrossRef] [PubMed] Asoh, T.-A.; Takaishi, K.; Kikuchi, A. Adhesion of poly(vinyl alcohol) hydrogels by the electrophoretic manipulation of phenylboronic acid copolymers. J. Mater. Chem. B 2015, 3, 6740–6745. [CrossRef] Ebara, M.; Uto, K.; Idota, N.; Hoffman, J.M.; Aoyagi, T. The taming of the cell: Shape-memory nanopatterns direct cell orientation. Int. J. Nanomed. 2014, 9, 117–126. [CrossRef] [PubMed]


15. 16.

17.

18. 19.

20. 21.

22. 23.

24.

25.

9 of 9

Ganta, S.; Devalapally, H.; Shahiwala, A.; Amiji, M. A review of stimuli-responsive nanocarriers for drug and gene delivery. J. Control. Release 2008, 126, 187–204. [CrossRef] [PubMed] Stroganov, V.; Al-Hussein, M.; Sommer, J.-U.; Janke, A.; Zakharchenko, S.; Ionov, L. Reversible Thermosensitive Biodegradable Polymeric Actuators Based on Confined Crystallization. Nano Lett. 2015, 15, 1786–1790. [CrossRef] [PubMed] Zhang, X.; Pint, C.L.; Lee, M.H.; Schubert, B.E.; Jamshidi, A.; Takei, K.; Ko, H.; Gillies, A.; Bardhan, R.; Urban, J.J.; et al. Optically- and Thermally-Responsive Programmable Materials Based on Carbon NanotubeHydrogel Polymer Composites. Nano Lett. 2011, 11, 3239–3244. [CrossRef] [PubMed] Zeng, C.; Seino, H.; Ren, J.; Hatanaka, K.; Yoshie, N. Bio-Based Furan Polymers with Self-Healing Ability. Macromolecules 2013, 46, 1794–1802. [CrossRef] Imato, K.; Nishihara, M.; Kanehara, T.; Amamoto, Y.; Takahara, A.; Otsuka, H. Self-Healing of Chemical Gels Cross-Linked by Diarylbibenzofuranone-Based Trigger-Free Dynamic Covalent Bonds at Room Temperature. Angew. Chem. Int. Ed. 2012, 51, 1138–1142. [CrossRef] [PubMed] Harada, A.; Kobayashi, R.; Takashima, Y.; Hashizume, A.; Yamaguchi, H. Macroscopic self-assembly through molecular recognition. Nat. Chem. 2011, 3, 34–37. [CrossRef] [PubMed] Kakuta, T.; Takashima, Y.; Nakahata, M.; Otsubo, M.; Yamaguchi, H.; Harada, A. Preorganized Hydrogel: Self-Healing Properties of Supramolecular Hydrogels Formed by Polymerization of Host-Guest-Monomers that Contain Cyclodextrins and Hydrophobic Guest Groups. Adv. Mater. 2013, 25, 2849–2853. [CrossRef] [PubMed] Basak, S.; Nanda, J.; Banerjee, A. Multi-stimuli responsive self-healing metallo-hydrogels: Tuning of the gel recovery property. Chem. Commun. 2014, 50, 2356–2359. [CrossRef] [PubMed] Yuan, J.; Fang, X.; Zhang, L.; Hong, G.; Liu, Y.; Zheng, Q.; Xu, Y.; Ruan, Y.; Weng, W.; Xia, H.; et al. Multi-responsive self-healing metallo-supramolecular gels based on “click” ligand. J. Mater. Chem. 2012, 22, 11515–11522. [CrossRef] Holten-Andersen, N.; Harrington, M.J.; Birkedal, H.; Lee, B.P.; Messersmith, P.B.; Lee, K.Y.C.; Waite, J.H. pH-induced metal-ligand cross-links inspired by mussel yield self-healing polymer networks with near-covalent elastic moduli. Proc. Natl. Acad. Sci. USA 2011, 108, 2651–2655. [CrossRef] [PubMed] Sato, T.; Ebara, M.; Tanaka, S.; Asoh, T.-A.; Kikuchi, A.; Aoyagi, T. Rapid self-healable poly(ethylene glycol) hydrogels formed by selective metal-phosphate interactions. Phys. Chem. Chem. Phys. 2013, 15, 10628–10635. [CrossRef] [PubMed] © 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).