Underwater-adhesive microparticle dressing composed of hydrophobically-modified Alaska pollock gelatin for gastrointestinal tract wound healing

Acta Biomaterialia xxx (xxxx) xxx

Contents lists available at ScienceDirect

Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat

Full length article

Underwater-adhesive microparticle dressing composed of hydrophobically-modified Alaska pollock gelatin for gastrointestinal tract wound healing Akihiro Nishiguchi, Yukari Kurihara, Tetsushi Taguchi ⇑ Polymeric Biomaterials Group, Biomaterials Field, Research Center for Functional Materials, National Institute for Materials Science, 1-1 Namiki, Tsukuba, Ibaraki 305-0044, Japan

a r t i c l e

i n f o

Article history: Received 25 March 2019 Received in revised form 24 July 2019 Accepted 21 August 2019 Available online xxxx Keywords: Endoscopic submucosal dissection Wound healing Underwater adhesion Microparticle Hydrophobic interaction

a b s t r a c t Despite the success of minimally-invasive endoscopic submucosal dissection (ESD) for the treatment of early gastrointestinal cancer, additional symptoms after ESD, including contracture, perforation, bleeding, and esophageal stricture remain. Conventional wound dressings were ineffective in preventing stricture because of poor stability of underwater-adhesives on living tissues. Here, we present a microparticlebased wound dressing with underwater adhesive stability for the treatment of gastrointestinal tract wound healing after ESD. Monodisperse microparticles composed of hydrophobically-modified Alaska pollock gelatin were prepared by self-assembly of gelatin in water-ethanol mixed solvents and thermal crosslinking. Hydrophobic modification of gelatin with aliphatic aldehydes increased adhesion strength to gastric and esophageal submucosal tissues through hydrophobic interaction with living tissues and cohesion force. Optimal hydrophobic modification drastically improved underwater stability of microparticles compared to that of non-modified gelatin and formed a thick, integrated hydrogel layer on tissues. Histological observation of rat skin wound healing models showed that hydrophobicallymodified gelatin microparticles decreased the expression levels of a-smooth muscle actin in the dermis layer and could suppress fibrosis and inflammation after ESD. The microparticle wound dressing with high underwater-adhesive stability has enormous therapeutic potential to promote wound healing in the gastrointestinal tract and prevent additional symptoms after ESD. Statement of Significance The goal of this study was to develop wound dressing with strong tissue-adhesive property to living tissues for promoting wound healing after ESD treatment. Monodisperse microparticles composed of hydrophobically-modified Alaska pollock gelatin were prepared by self-assembly of gelatin in waterethanol mixed solvents and thermal crosslinking. Hydrophobic modification of gelatin with aliphatic aldehydes enhanced adhesion strength to gastric and esophageal submucosal tissues through hydrophobic interaction with living tissues and cohesion force. Optimal hydrophobic modification drastically improved underwater stability of microparticles. The in vivo studies were performed to evaluate the ability of this colloidal wound dressing to suppress fibrosis. This new biomaterial has enormous potential to promote wound healing after ESD. Ó 2019 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

1. Introduction Gastrointestinal cancer occurring at the mucosal epithelium is one of the leading causes of death. For instance, esophageal cancer is the eighth most common cancer worldwide, with an estimated 456,000 new cases in 2012 [1]. With the advances in diagnostic ⇑ Corresponding author. E-mail address: [email protected] (T. Taguchi).

and therapeutic technologies, early gastrointestinal cancer can be dissected by endoscopic surgery, called endoscopic submucosal dissection (ESD) [2,3]. Since ESD enables the preservation of the submucosal layer and muscularis propria after the removal of superficial neoplasms, ESD has been widely used as a minimallyinvasive therapy. However, after ESD, additional symptoms, such as severe stricture, often occur due to the inflammation and cicatricial contracture, and the stricture rate after removal of mucosa exceeding three-quarters of esophageal circumference is over

https://doi.org/10.1016/j.actbio.2019.08.040 1742-7061/Ó 2019 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

Please cite this article as: A. Nishiguchi, Y. Kurihara and T. Taguchi, Underwater-adhesive microparticle dressing composed of hydrophobically-modified Alaska pollock gelatin for gastrointestinal tract wound healing, Acta Biomaterialia, https://doi.org/10.1016/j.actbio.2019.08.040

2

A. Nishiguchi et al. / Acta Biomaterialia xxx (xxxx) xxx

90% [4,5]. Although several procedures including endoscopic balloon dilation [6], implantation of a stent [7], and the use of steroids [8] have been used, these methods fail in tissue regeneration and have the risk of causing perforation and bleeding [4]. Wound dressing materials that can physically protect wounds from stimuli are promising for treatment after ESD [9–13], rather than advanced approaches such as cell-based therapy [14], which requires huge costs and preparation time and poses the risk of infection. However, conventional dressing materials such as bioresorbable polymeric meshes [15–17], hydrogels [18], and decellularized constructs [19] easily detach from tissues even with the use of a clip and glue because they lack the design of adhesion to living tissues under wet conditions [20]. Therefore, there is no reliable therapeutic approach to prevent esophageal stricture after ESD [3]. To protect wounds from endoluminal factors and infection after ESD, dressing materials need to be highly tissue-adhesive and stable on submucosal tissues in harsh microenvironments exposed to peristaltic motion and shear stress by liquid. An emerging strategy to improve the underwater adhesive property is based on leveraging physical interaction [21], which is often seen in nature such as in marine organisms [22]. In contrast to chemical reaction-based adhesion, which may cause undesirable toxicity, biomimetic adhesion using physical interactions such as catechol chemistry [23–25], polyelectrolyte complex [26], supramolecular architectures [27,28], and particle adhesives [29,30] is safe, facile for handling, reversible, reconfigurable, and interactive under wet conditions. We previously reported that the use of hydrophobic interaction improved adhesion strength to living tissues such as blood vessels and lungs under wet conditions [31–34]. Recently, we developed a tissue-adhesive colloidal wound dressing composed of hydrophobically-modified gelatin (hm-Gltn) derived from porcine [35]. Microparticles (MPs) prepared by spray drying and thermal crosslinking methods strongly adhered onto gastric submucosal tissues. However, the fabrication process for the particles was complicated and required a specific device, and the particles obtained were polydisperse. Moreover, underwater-adhesive stability to gastrointestinal mucosal tissues has not yet been elucidated. Here, we report a wound dressing of monodisperse MPs with high adhesive stability under wet conditions. Monodisperse MPs were prepared by self-assembly of Alaska pollock-derived Gltn (ApGltn) in water-ethanol mixed solvent and thermal crosslinking. This simple and facile method requires no surfactants and toxic chemicals for particle preparation. We evaluated the adhesion strength of hm-ApGltn MPs to gastric and esophageal submucosal tissues and addressed the effect of alkyl chain length of hydrophobic moieties in hm-ApGltn and thermal crosslinking time on adhesion strength. Moreover, underwater adhesive stability of hmApGltn MPs on esophageal submucosal tissues was assessed. Finally, to address the effect of hm-ApGltn MPs on fibrosis and inflammation, the expression level of a-smooth muscle actin (a SMA) in dermal layers was evaluated using rat full-thickness skin wound models.

2. Methods 2.1. Synthesis of hm-ApGltn Hm-ApGltn was synthesized by reductive amination of the amino group in ApGltn with aliphatic aldehydes according to a previous report [31]. The 20 w/v% ApGltn (Mw = 31,000 Da, amino group: 324 lmol/g, Nitta Gelatin Inc., Japan) was dissolved in ultra-pure water at 50 °C. A 1.5 equivalent of 2-picoline borane (Junsei Chemical Co., Ltd., Japan) and 200 mol% aliphatic aldehydes (C2: acetaldehyde, C4: butanal, C6: hexanal, C8: octanal, C12:

dodecanal, Tokyo Chemical Industry Co., Ltd., Japan) were dissolved in ethanol and slowly added to the above solution, respectively (water:ethanol = 7:3). The reaction proceeded for 17 h at 50 °C. The solution obtained was slowly added to 10-volumes of cold ethanol. The precipitates were then washed with cold ethanol for 1 h at 4 °C followed by filtration with a glass filter. The washing step was repeated three times to remove unreacted aliphatic aldehydes and 2-picoline borane. The products were dried at 25 °C under reduced pressure for 3 days. The degree of substitution of hydrophobic groups was estimated by determining the residual amino groups using 2,4,6,-trinitrobenzensulfonic acid (TNBS, Tokyo Chemical Industry Co., Ltd., Japan) [36]. 2.2. Preparation of hm-ApGltn MPs Hm-ApGltn MPs were prepared by self-assembly of hm-ApGltn using poor solvent as a desolvating reagent. The 5 wt% hm-ApGltn was dissolved in ultra-pure water at 50 °C. After cooling down the solution to room temperature, ethanol as a poor solvent for hmGltn was slowly added to hm-Gltn solution under stirring at room temperature. The turbid solution obtained was cooled at 30 °C for 6 h, and the solvents were evaporated in a freeze-drying machine. The powder of the particles was then thermally-crosslinked at 150 °C for 1–6 h under reduced pressure in an oven to form thermally-crosslinked hm-ApGltn MPs. The morphology of the Gltn particles obtained was observed by scanning electron microscopy (SEM, S-4800 ultrahigh-resolution SEM, HITACHI, Japan). Platinum was sputtered on particles at a 10 nm thickness. The accelerating voltage and working distance were set to 10 kV and 8 mm, respectively. The diameter of the particles was estimated from SEM images using ImageJ software. 2.3. Contact angle measurement To evaluate the surface properties of MPs, contact angle measurement was carried out using an automatic contact angle meter (DM-700, KYOWA, Japan). The gelatin particles (Org, C6, C8, C10, C12) were placed on a glass slide and pressed from the top to form a pellet of particles. A 1 lL droplet of water was added on the pellet, and the contact angle was measured after 5 s. 2.4. X-ray photoelectron spectroscopy (XPS) analysis XPS analysis was performed with X-ray Photoelectron Spectrometer Quantum 2000 (ULVAC-PHI, Inc., Japan) using Al Ka radiation (20 W, 15 kV) at an operating pressure less than 1  10 8 mBar. For the sample preparation, MPs were dispersed in water and casted on a glass substrate to form MP film. The pass energy was set at 20 eV for scans. The obtained spectra were analyses using MultiPak ver.8.2c peak fitting software. A reference charge correction of 284.5 eV for C 1s was used. 2.5. Adhesion strength test Fresh porcine stomach and esophagus (Tokyo Shibaura Zouki, Japan) were employed as a model for the adhesion test. The mucosal membrane was removed by injecting saline into submucosal tissue and cutting with surgical scissors. The tissues composed of submucosa and muscle layers were dissected into 2.5  2.5 cm squares. The adhesion strength and adhesion energy were measured using a Texture Analyzer (TA-XT2i, Stable Microsystems, UK) according to the ASTM F-2258-05, Standard Test Method for Strength Properties of Tissue Adhesives in Tension. First, the tissues were bound to the probe (top) and the hot plate was kept at 37 °C (bottom) with a cyanoacrylate adhesive (Loctite, Henkel, Germany). To remove excess water from the tissues, Kimwipes were

Please cite this article as: A. Nishiguchi, Y. Kurihara and T. Taguchi, Underwater-adhesive microparticle dressing composed of hydrophobically-modified Alaska pollock gelatin for gastrointestinal tract wound healing, Acta Biomaterialia, https://doi.org/10.1016/j.actbio.2019.08.040

A. Nishiguchi et al. / Acta Biomaterialia xxx (xxxx) xxx

placed between two pieces of tissue and pressed at 80 kPa of pressure for 3 min. A total of 100 mg of particles was then placed on the tissue at the bottom. The top probe approached the stage at a 10 mm/min tracking speed and was pressed at 80 kPa of pressure for 3 min. The top probe returned to the original position at a speed of 10 mm/min and then the adhesion strength was measured.

2.6. Underwater stability test To evaluate the adhesiveness and stability of hm-ApGltn MPs in wet environments like in the esophagus, underwater stability tests of the particles were performed. In tissue adhesion test, compressive pressure was applied to the particles and tissues before measuring tissue strength. However, in ESD procedure, particles were just sprayed to wounds and compressive pressure can not be applied. Therefore, we checked underwater stability of particles on tissues as follows. Porcine-derived esophagus tissues were dissected into 2.5  2.5 cm squares, and the mucosal membranes were removed. The 10 mg of the particles were placed onto submucosal tissue using a spatula without the compression. The particles gradually swelled with tissue fluids of submucosal tissues and became transparent. After 3 min at room temperature, gelation of the particles completed, and hydrogel layer was formed on the tissues. The tissues were then immersed into 30 mL of saline (Otsuka Pharmaceutical, Japan) containing 0.02% sodium azide (Wako, Japan) and incubated at 37 °C in an incubator for 2 or 4 days. The tissues were then fixed in 10% formalin buffer solution, and cross-sections of the tissues were histologically observed using hematoxylin and eosin (HE) staining. The images were scanned using a digital slide scanner (NanoZoomer S210, Hamamatsu Photonics, Japan) and the area of hydrogel layer was quantified using ImageJ.

2.7. Animal experiments Animal experiments using rats were carried out with the approval of the Animal Care and Use Committee of the National Institute for Material Science. Female rats (Wistar, 7 weeks old, n = 3–4) were anesthetized by inhalation of 2.5% isoflurane. Hair on the back skin was shaved with a razor, and the area was disinfected with 70% ethanol. Full-thickness skin wounds were formed on the backs using an 8-mm circular punch. After hemostasis, 10 mg of particles was placed on the wounds and the exposed tissue surface was covered. Wounded areas of control samples were not covered with any dressing materials. The wounds were sealed using a dressing material (Tegaderm, 3 M, USA) to maintain wet conditions and fixed with a rat jacket (BRC Bioresearch center, Japan). The rats were housed in individual cages. The wound areas were macroscopically monitored and measured using ImageJ software. After 7 and 14 days, rats were euthanized by intraperitoneal injection of an overdose of pentobarbital sodium salt (Somnopentyl, Kyoritsu Seiyaku, Japan). Wounded tissues were dissected from the sacrificed rats and fixed in 10% formalin buffer solution for at least 3 days. Cross-sections of the tissues were stained with an a-SMA antibody for histological observation. The images were scanned using a digital slide scanner.

2.8. Statistical analysis All data are expressed as the mean ± SD unless otherwise specified. The values represent the mean ± SD from three independent experiments. A p-value <0.05 was considered statistically significant.

3

3. Results and discussion 3.1. Facile fabrication of hm-ApGltn MPs ApGltn was modified with aliphatic aldehydes possessing different chain lengths ranging from C6 to C12. Amino groups in ApGltn (324 lmol/g) reacted with aldehyde groups via a Schiff base reaction to form a secondary amine in the presence of 2picoline borane. The degree of substitution (D.S.) of aliphatic aldehydes was tuned by changing the feeding ratio of aliphatic aldehydes to synthesize hm-ApGltn with low and high hm-ApGltn (Table S1). Hm-ApGltn MPs were prepared by a simple and facile self-assembly method using poor solvent as a desolvating reagent (Fig. 1a). When ethanol as a poor solvent for ApGltn was slowly added to hm-ApGltn aqueous solution at equal volumes, the solution became turbid to form hm-ApGltn MPs (Fig. 1b). The addition of ethanol dehydrated the hm-ApGltn polymer rich phase, shrank hm-ApGltn polymer, and induced self-assembly to form stabilized emulsion-like droplets [37]. Since the preparation method of Gltn nanoparticles reported previously required a toxic crosslinking reagent [38,39], we chose thermal crosslinking method which is based on the formation of amide bonds by dehydration reaction between amino and carboxy groups in Gltn. The dispersion was treated by freeze-drying (although the solution was not frozen), and the solvent was evaporated. Thermal crosslinking at 150 °C for 1–6 h under reduced pressure was performed to improve the stability of particles in water to obtain a powder of hm-ApGltn MPs at high yield. Scanning electron microscopy (SEM) observation displayed the formation of highly-monodisperse micrometer-sized particles of C10-ApGltn (Fig. 1c). The average diameter was approximately 2 lm (Fig. 1d). The morphology of hm-ApGltn MPs did not change before and after thermal crosslinking. Dried MPs were sprayable with clinically-available air pump devices (Movie S1). This property would be useful for delivering materials to wounds compared to mesh and sheet-type dressing [14]. Optimal preparation conditions were the use of 5% ApGltn solution and ethanol at equal volumes for the particulation in terms of the monodispersity (Fig. S1). Interestingly, hm-ApGltn MPs revealed a unique swelling process. We found that hm-ApGltn MPs spontaneously aggregated during the swelling and formed large hydrogel layers after 1 h of incubation in water (Fig. 2a). The size of aggregates increased after 3 h, and MPs fused together after 24 h. On the other hand, although Org-ApGltn MPs swelled and partly fused together, the interface of MPs was still clear and did not form aggregates after 24 h. These observations indicate that hydrophobic interaction between hmMPs drove the aggregation of hm-MPs in water. Contact angle measurement showed that hydrophobicity of microparticle surfaces increased by modifying aliphatic aldehyde with long alkyl chain length and C10-MPs possessed the highest hydrophobicity among them (Fig. 2b). This result was in good agreement with previous report [35]. More hydrophobic hm-MPs possessing alkyl chain groups on particles surfaces enabled physical interaction between microparticles. This unique property of hm-MPs contributes to protection of wounds by forming a hydrogel layer in situ. To further understand surface property of hm-MPs, XPS analysis was performed. Since MPs were treated with thermal crosslinking at 150 °C in the air, we expect that hydrophobic alkyl chains tend to localize on the particle surfaces. XPS analysis showed that atomic concentration of spectra at 284.5 eV attributed to C–C bonds in C s1 spectra of hm-MPs was 46.3% and higher than that of Org-MPs (43.7%) (Fig. 2c, d). These results indicate that hydrophobic alkyl chains existed on particle surfaces and possibly underwent the surface segregation into the air by thermal treatment. This hydrophobic moieties on MPs allows for robust

Please cite this article as: A. Nishiguchi, Y. Kurihara and T. Taguchi, Underwater-adhesive microparticle dressing composed of hydrophobically-modified Alaska pollock gelatin for gastrointestinal tract wound healing, Acta Biomaterialia, https://doi.org/10.1016/j.actbio.2019.08.040

4

A. Nishiguchi et al. / Acta Biomaterialia xxx (xxxx) xxx

Fig. 1. Preparation of monodispersed hm-ApGltn MPs by simple and facile self-assembly method. (a) Schematic illustration of synthesis of hm-ApGltn and preparation of crosslinked hm-ApGltn MPs. Addition of ethanol to ApGltn solution induced self-assembly of ApGltn to form hm-MPs. After freeze-drying, MPs were thermally-crosslinked at 150 °C. (b) Bright field microscopic image of C10-ApGltn in water and C10-ApGltn in water-ethanol mixed solvent. (c) SEM image and (d) size distribution of hm-ApGltn MPs. Scale bars represent 20 lm for (b) and 10 lm for (c).

Fig. 2. Morphological change in water and surface property of hm-MPs. (a) Bright field microscopic images of Org- and hm-ApGltn MPs after the incubation in water. 47C10ApGltn was used for this experiment. (b) Contact angle measurement of MPs (Org, C6, C8, C10, C12). *P < 0.05, **P < 0.01, ***P < 0.01, statistics by one-way analysis of variance (ANOVA) with Tukey’s multiple comparison test (n = 3). High resolution C 1s XPS spectra of (c) Org-MPs and (d) 47C10-MPs. Scale bars represent 50 lm.

adhesion to living tissues containing hydrophobic proteins and the aggregation between particles, which may contribute to the protection of wounds after ESD. 3.2. Effect of derivation and hydrophobicity of Gltn on the size of MPs To clarify the effect of derivation of Gltn on the formation of MPs, porcine Gltn was particulated in the same manner as ApGltn. As described above, we developed a microparticle dressing composed of porcine-derived Gltn by spray drying method. Although porcine Gltn microparticles exhibited strong tissue adhesiveness,

they required 1 h for swelling in swine ESD models. To accelerate swelling speed, we chose ApGltn due to more hydrophilic property. It is reported that ApGltn contains less amount of imino acids such as proline and hydroxyproline, which are correlated with the solubility due to the decrease in intermolecular hydrogen bonding [40]. The imino acid content in ApGltn is about 15.0% per 1000 amino acid residues, which is lower than that in porcine Gltn (about 22.3% per 1000 amino acid residue) [41]. Although Org- and C6ApGltn provide monodisperse MPs, porcine Gltn formed polydisperse, less-spherical MPs even at a similar molecular weight as ApGltn (ApGltn: 31 kDa, Porcine Gltn: 40 kDa) (Fig. S2). The high

Please cite this article as: A. Nishiguchi, Y. Kurihara and T. Taguchi, Underwater-adhesive microparticle dressing composed of hydrophobically-modified Alaska pollock gelatin for gastrointestinal tract wound healing, Acta Biomaterialia, https://doi.org/10.1016/j.actbio.2019.08.040

A. Nishiguchi et al. / Acta Biomaterialia xxx (xxxx) xxx

molecular weight of porcine Gltn (Mw = 100 kDa) formed nonspherical, large aggregates. Since a surfactant that can stabilize the interface between particles and solvents was not used in this method, the solubility of polymers has considerable influence on the formation process of MPs. Porcine Gltn possessing higher intermolecular interaction between collagen-like sequences due to high imino acid content may lose the interfacial stability in waterethanol mixed solvents, which leads to the aggregation. Since ApGltn with low imino acid content possesses high affinity to water compared to porcine Gltn, ApGltn formed monodisperse MPs without surfactants. In the following experiments, ApGltn was used as a component of MPs. One of the key parameters to determine the quality of particles is the balance of the hydrophilic-hydrophobic property. Hydrophobic moieties may localize on the surfaces of MPs since MPs were dispersed in ethanol solution and thermally-crosslinked in air. We prepared MPs composed of Org-, C6-, C8-, C10-, and C12ApGltn with low and high D.S. and evaluated the effect of the hydrophobicity on the size of the MPs (Fig. 3a). In the case of low D.S. of hm-ApGltn, with increasing the length of alkyl chain groups, the dispersity of MPs improved and the average diameter of MPs decreased (Fig. 3b). High D.S. of hm-MPs except C12 showed the same tendency compared to Org-MPs even in C6 and C8. C12-MPs contained larger and smaller MPs due to instability of particle interfaces by too strong of a hydrophobic interaction. 3.3. Effect of alkyl chain length and thermal crosslinking time on adhesion strength to gastric submucosal tissues Stable adhesion to submucosal tissues under wet conditions is the most important functionality to protect the damaged tissues after ESD. We reported that leveraging hydrophobic interaction

5

as a driving force improved adhesion strength to living tissues [31–33] and clarified the effect of hydrophobicity of porcine Gltn MPs on adhesion strength [35]. To achieve strong underwater adhesion to gastrointestinal tissues, optimal chain length modified to ApGltn and thermal crosslinking time were explored. The adhesion test was performed using a porcine-derived gastric tissue according to the ASTM F-2258-05, standard test method of tensile measurement. MPs were placed onto a submucosal tissue and sandwiched with two pieces of tissues at 80 kPa for 3 min. We checked that the adhesin strength without any particles was very low (less than 2 kPa, data not shown). Although low D.S. of hmMPs did not improve adhesion strength compared to Org-MPs, high D.S. of hm-MPs increased adhesion strength with increasing alkyl chain length, and 47C10-MPs showed 2.4-fold increases compared to Org-MPs (Fig. 4a). On the other hand, C12-MPs showed no significant increase of adhesion strength because it was too hydrophobic to swell in water and undergo gelation. These tendencies in the relationship between adhesion strength and D.S. or chain length of alkyl groups were in good agreement with a previous report using porcine Gltn MPs [39]. Although C8-modification displayed the highest adhesion strength in the case of porcine Gltn MPs [35], ApGltn MPs required more hydrophobic, C10modification because ApGltn is more hydrophilic than porcine Gltn. In contrast non-adhesive polymeric meshes [35], highlyadhesive sprayable dressing can be fixed to wounds without the need of suturing and gluing. Moreover, the effect of crosslinking density of MPs on adhesion strength was addressed. The crosslinking density was tuned by thermal crosslinking time. C10-MPs with high D.S. thermallycrosslinked for 3 h showed a significant increase of adhesion strength compared to 1 h of thermal crosslinking (Fig. 4b). Adhesion strength is dependent on both the interfacial adhesion

Fig. 3. Hydrophobicity of Gltn affected the morphology of particles. (a) SEM images and (b) size distribution of MPs composed of Org-, C6-, C8-, C10-, and C12-ApGltn with low and high D.S. 11C6, 13C8, 12C10, and 11C12 as low D.S. of ApGltn and 77C6, 58C8, 47C10, and 48C12 as high D.S. of ApGltn were used for this experiment. The 100 particles were counted from SEM images. Scale bars represent 10 lm.

Please cite this article as: A. Nishiguchi, Y. Kurihara and T. Taguchi, Underwater-adhesive microparticle dressing composed of hydrophobically-modified Alaska pollock gelatin for gastrointestinal tract wound healing, Acta Biomaterialia, https://doi.org/10.1016/j.actbio.2019.08.040

6

A. Nishiguchi et al. / Acta Biomaterialia xxx (xxxx) xxx

Fig. 4. Effect of alkyl chain length and thermal crosslinking time on tissue adhesive property. (a) Adhesion strength of hm-MPs with varied alkyl chain length to gastric submucosal tissues. Porcine gastric mucosa tissue was dissected by injection of saline, and the residual submucosa tissue was fixed onto a probe. MPs were applied onto the tissue surface and the tissues were pressed at 80 kPa for 3 min. The adhesion test was performed according to the ASTM F2258-05 standard method. 11C6, 13C8, 12C10, and 11C12 as low D.S. of ApGltn and 77C6, 58C8, 47C10, and 48C12 as high D.S. of ApGltn were used for this experiment. *P < 0.05, **P < 0.01, Tukey’s multiple comparison test (n = 3). (b) Adhesion strength of hm-MPs treated with different thermal crosslinking time (150 °C, 1–6 h) to gastric submucosal tissues. 77C6, 58C8, 47C10, and 48C12 ApGltn were used for this experiment. *P < 0.05, **P < 0.01, ***P < 0.001, Tukey’s multiple comparison test (n = 3).

strength and the cohesion force. MPs with low crosslinking density had low cohesion force, leading to loss of adhesion strength. We concluded that C10-modificaiton of ApGltn and over 3 h of thermal crosslinking of MPs was an optimal method to achieve strong adhesion strength to gastric submucosal tissues. In the following experiments, MPs composed of high D.S. of hm-Gltn were used (77C6, 58C8, 46C10, 49C12).

enhance interfacial adhesion strength to esophageal submucosal tissues through hydrophobic interaction between hm-MPs and tissues. Moreover, as shown in Fig. 1e, hm-MPs spontaneously fused and formed hydrogel layers, and this behavior based on hydrophobic interaction formed robust film-like structures of hm-MPs and increased cohesion force and adhesion strength. In terms of esophageal submucosal tissues, C10-modified MPs thermal crosslinked for 3 h improved the adhesion as well as gastric tissues.

3.4. Adhesion strength to esophageal submucosal tissue 3.5. Underwater-adhesive stability of particles on submucosa Next, we addressed adhesion strength of hm-ApGltn MPs to an esophageal submucosal tissue. As with gastric tissue, a dressing material on esophageal tissue will be exposed to harsh environments in which large peristaltic motion and shear stress occur, which may lead to the detachment of the dressing. Macroscopic observation of porcine-derived esophageal submucosal tissues during adhesion testing revealed that esophageal tissues treated with C10-MPs were extended when tissues were detached (Fig. 5a). Although there was little difference in maximum adhesion strength, adhesion energy of C10-MPs was the highest among other MPs of ApGltn modified with different alkyl chain lengths (Fig. 5b, c). Especially, C10-MPs thermally-crosslinked for 1 h and 3 h formed sticky film-like structures during the adhesion test (Fig. 5d). The adhesion energy of C10-MPs drastically improved by 3 h of thermal crosslinking and showed a >9-fold increase compared to Org-MPs (Fig. 5e, f and Movie S2). Moreover, there were significant differences in adhesion strength between Org-MPs and C10-MPs prepared by different thermal crosslinking time (Fig. S3). These results suggest that hydrophobic modification can

In addition to tissue adhesiveness, underwater stability of dressing materials is a key property to cover wounds for a longterm period and allow resistance to hydrolysis under wet conditions, shear stress, and peristaltic motion. We evaluated the underwater-adhesive stability of MPs and compared various MPs with different alkyl chain lengths and crosslinking density. MPs were placed on porcine esophageal submucosal tissues without any compression, which attempted to imitate the more practical and actual situation in the ESD procedure (Fig. 6a). After 3 min of incubation to complete gelation, the tissues were immersed into saline (0.02% sodium assize) and incubated at 37 °C for 4 days. HE histological images showed that C10-MPs still remained on submucosal tissues, although Org-MPs already diffused into saline and only a monolayer of MPs remained (Fig. 6b). Hydrogel layers of C10-MPs were formed on tissues by swelling and gelation and were highly stable even in the water environment (Fig. 6c). This result indicates that hydrophobic modification of ApGltn enhances not only adhesion strength to tissues but also underwater stability

Please cite this article as: A. Nishiguchi, Y. Kurihara and T. Taguchi, Underwater-adhesive microparticle dressing composed of hydrophobically-modified Alaska pollock gelatin for gastrointestinal tract wound healing, Acta Biomaterialia, https://doi.org/10.1016/j.actbio.2019.08.040

A. Nishiguchi et al. / Acta Biomaterialia xxx (xxxx) xxx

7

Fig. 5. Tissue adhesiveness to esophageal submucosal tissue. (a) Photos just before detachment of tissues on which Org- and hm-MPs (6 h of thermal crosslinking, MPs-6 h) were employed. (b, c) Force-distance curve and adhesion energy of Org- and hm-ApGltn MPs treated with 6 h of thermal crosslinking. C10-MPs showed the highest adhesion energy to esophageal submucosa. **P < 0.01, *** P < 0.001, Tukey’s multiple comparison test (n = 3). (d) Photos just before detachment of tissues with Org-MPs thermallycrosslinked for 3 h and C10-MPs thermally-crosslinked for 1, 3, and 6 h. (e, f) Force-distance curve and adhesion energy of Org and C10-MPs treated with 1, 3, and 6 h of thermal crosslinking. The 3 h of thermal crosslinking treatment resulted in the highest adhesion energy to esophageal submucosa. ***P < 0.001, Tukey’s multiple comparison test (n = 3). Scale bars represent 1 cm.

by hydrophobic interaction-driven aggregation of MPs. On the other hand, C12-MPs did not show high underwater stability as is the case with tissue adhesiveness, indicating the importance of controlling hydrophilic and hydrophobic balance. Interestingly, crosslinking density of MPs substantially affected underwater stability and the structure of hydrogel layers. After applying MPs and incubating in saline for 2 days, C10-MPs thermally-crosslinked for 3 h displayed a thick, dense film-like hydrogel layer (Fig. 6d, e). Each C10-MP completely fused during the incubation. On the other

hand, C10-MPs thermally-crosslinked for 1 h resulted in a porous, loosely-interconnected hydrogel layer, and 6 h of thermal crosslinking maintained original spherical shapes of particles due to high crosslinking density in MPs. Therefore, hm-MPs achieved high underwater-adhesive stability on tissues by optimal hydrophobic interaction [31–35] and the formation of dense hydrogel layers that can protect tissues from endoluminal factors. These results from the underwater stability test showed the same tendency as adhesion strength experiments. The in situ gelling and

Please cite this article as: A. Nishiguchi, Y. Kurihara and T. Taguchi, Underwater-adhesive microparticle dressing composed of hydrophobically-modified Alaska pollock gelatin for gastrointestinal tract wound healing, Acta Biomaterialia, https://doi.org/10.1016/j.actbio.2019.08.040

8

A. Nishiguchi et al. / Acta Biomaterialia xxx (xxxx) xxx

Fig. 6. Underwater-adhesive stability test of hm-MPs on esophageal submucosa tissues. (a) Experimental procedure of underwater stability test. (b) HE images of esophageal submucosal tissues sprayed with Org- and C6-, C8-, C10-, and C12-MPs treated with 6 h of thermal crosslinking. The tissues were immersed into saline with 0.02% sodium azide and incubated at 37 °C for 4 days. (c) The cross-sectional area of the hydrogel layer of MPs (on 1 mm-width of submucosal tissue in cross-section) was measured from HE images. *P < 0.05, **P < 0.01, ***P < 0.001, Tukey’s multiple comparison test (n = 3). (d) HE images of esophageal submucosal tissues sprayed with C10-MPs treated for 1, 3, and 6 h with thermal crosslinking. The tissues were incubated in the saline at 37 °C for 2 days. (e) The area of hydrogel layers composed of C10-MPs. *P < 0.05, **P < 0.01, Tukey’s multiple comparison test (n = 3). Scale bars represent 100 lm for (b) and (d) and 10 lm for insert in (d).

adhesive property of hm-MPs enables the delivery of dressing materials and reduces the physical burden on operators and patients. 3.6. Suppression of fibrosis in skin wound healing model Finally, we addressed whether hm-MPs can suppress the inflammation and fibrosis of wounds in vivo. Full-thickness skin wound healing models of rats were used. Skin wounds (8 mm in diameter) were prepared on the backs of rats. Then, the wounds were covered with 10 mg of MPs. In all samples, the area of wounds decreased and re-epithelialization was almost complete after 14 days (Fig. S4). Immunohistological observation of tissues after 14 days revealed that the expression of a-SMA, which is a differentiation marker of contractile myofibroblasts, decreased when C10- and C12-MPs were applied to the wounds (Fig. 7a). Although the difference of a -SMA expression was small at day 7, C10- and C12-MPs showed significant differences in a -SMA expression in the dermis layers at day 14 compared to control and C6-, C8-MPs (Fig. 7b, c). Fibrosis has a risk for causing scar contracture and stricture even after re-epithelialization is completed. Here, C10-MPs effectively suppressed the fibrosis in the dermis layer, indicating that swelled MPs protect the wound surfaces by forming a hydrogel layer and support the regeneration process of tissues. Since clinically-available polymeric meshes did not suppress fibrosis [35], C10-MPs would be a powerful tool to heal wounds and pre-

vent additional symptoms after ESD. Although there were no significant differences in CD31-positive area (marker of endothelial cells) between samples, we found the tendency that longer hydrophobic alkyl chains increased CD31 expression after 7 days (Fig. S5). This result corresponded with our previous report showing that hydrophobic modification enhanced angiogenesis [33]. After 14 days, only the control sample showed a high CD31 expression level, indicating delayed tissue remodeling. On the other hand, although skin wound healing models used in this study were useful to clarify the effects of MPs in fibrosis and wound healing, skin wounds that were not exposed to body liquid during the healing process are different from the situation of ESD, which occurs under wet conditions. Further investigation is needed using an ESD model formed on gastrointestinal tissue. We expect that molecular modification of ApGltn with distinct hydrophobicity and fabrication technology of MPs will be useful for the treatment to suppress fibrosis in tissues and promote wound healing after ESD. 4. Conclusion In conclusion, we demonstrate the effectiveness of a particlebased wound dressing with underwater adhesive stability. Monodisperse ApGltn-MPs were fabricated using a simple, facile, and biocompatible self-assembly method. Dispersity and size of MPs can be tuned by modified alkyl chain length and preparation conditions. Hydrophobic modification of ApGltn with C10-alkyl

Please cite this article as: A. Nishiguchi, Y. Kurihara and T. Taguchi, Underwater-adhesive microparticle dressing composed of hydrophobically-modified Alaska pollock gelatin for gastrointestinal tract wound healing, Acta Biomaterialia, https://doi.org/10.1016/j.actbio.2019.08.040

A. Nishiguchi et al. / Acta Biomaterialia xxx (xxxx) xxx

9

Fig. 7. Suppression of fibrosis in rat skin wound healing models by hm-MPs. (a) Immunohistological observation of a-SMA-stained dermal layers exposed to Org- and C6-, C8, C10, and C12-MPs for 14 days. The 8 mm of full-thickness skin wounds were prepared on the backs of rats, and the dissected areas were covered with various MPs thermallycrosslinked at 150 °C for 3 h. Quantification of the signal-positive area after (b) 7 days and (c) 14 days. C10- and C12-hMPs effectively suppressed fibrosis as compared to the control, C6-, and C8-MPs. *P < 0.05, **P < 0.01, Tukey’s multiple comparison (n = 4). Scale bars represent 100 lm.

chain groups and 3 h of thermal crosslinking increased adhesion strength of MP dressing to gastric and esophageal mucosal tissues under wet conditions through strong hydrophobic interaction with living tissues and the cohesion force. C10-ApGltn MPs treated with 3 h of thermal crosslinking resulted in the formation of a thick, MPs-fused hydrogel layer in physiological saline. In rat fullthickness skin wound models, 46C10-MPs suppressed fibrosis in the epidermis layers. Sprayable MPs with in situ gelling and adhesive property enable a facile spray-delivery process without the need of a complicated device and coverage of the wounds independent of the surface characteristics of the tissue. Underwateradhesive, monodisperse MPs may serve as a wound dressing material that can suppress the stricture after ESD. Furthermore, in addition to prevention of stricture, this particle-based sprayable wound dressing has therapeutic potential to prevent the risk of wall rupture after the operation and heal inflammatory diseases like ulcerative colitis.

Data availability The authors declare that all data supporting the findings of this study are available within the paper and its Supplementary Information.

Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments This study was financially supported in part by the Japan Society for the Promotion of Science (JSPS) KAKENHI (Grant Nos. 16H04524 and 18K14099) and the Project for Translational Research program, Strategic PRomotion for practical application of INnovative medical Technology from the Japan Agency of Medical Research and Development (AMED) under Grant Number JP 18lm0203023. Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.actbio.2019.08.040. References [1] J. Ferlay, I. Soerjomataram, R. Dikshit, S. Eser, C. Mathers, M. Rebelo, D.M. Parkin, D. Forman, F. Bray, Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012, Int. J. Cancer 136 (2015) E359–E386. [2] T. Gotoda, Endoscopic resection of early gastric cancer, Gastric Cancer 10 (2007) 1–11. [3] M. Barret, B. Beye, S. Leblanc, F. Beuvon, S. Chaussade, F. Batteux, F. Prat, Systematic review: the prevention of oesophageal stricture after endoscopic resection, Aliment. Pharmacol. Ther. 42 (2015) 20–39. [4] K. Funakawa, H. Uto, F. Sasaki, Y. Nasu, S. Mawatari, S. Arima, J. Nakazawa, H. Taguchi, S. Hashimoto, S. Kanmura, H. Setoyama, M. Numata, H. Tsubouchi, A. Ido, Effect of endoscopic submucosal dissection for superficial esophageal neoplasms and risk factors for postoperative stricture, Medicine (Baltimore) 94 (2015) e373. [5] S. Ono, M. Fujishiro, K. Niimi, O. Goto, S. Kodashima, N. Yamamichi, M. Omata, Long-term outcomes of endoscopic submucosal dissection for superficial esophageal squamous cell neoplasms, Gastrointest. Endosc. 70 (2009) 860– 866. [6] Y. Ezoe, M. Muto, T. Horimatsu, S. Morita, S.I. Miyamoto, S. Mochizuki, K. Minashi, T. Yano, A. Ohtsu, T. Chiba, Efficacy of preventive endoscopic balloon

Please cite this article as: A. Nishiguchi, Y. Kurihara and T. Taguchi, Underwater-adhesive microparticle dressing composed of hydrophobically-modified Alaska pollock gelatin for gastrointestinal tract wound healing, Acta Biomaterialia, https://doi.org/10.1016/j.actbio.2019.08.040

10

[7]

[8]

[9]

[10] [11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21] [22]

A. Nishiguchi et al. / Acta Biomaterialia xxx (xxxx) xxx dilation for esophageal stricture after endoscopic resection, J. Clin. Gastroenterol. 45 (2011) 222–227. J. Wen, Z. Lu, Y. Yang, Q. Liu, J. Yang, S. Wang, X. Wang, H. Du, J. Meng, H. Wang, E. Linghu, Preventing stricture formation by covered esophageal stent placement after endoscopic submucosal dissection for early esophageal cancer, Dig. Dis. Sci. 59 (2014) 658–663. N. Hanaoka, R. Ishihara, Y. Takeuchi, N. Uedo, K. Higashino, T. Ohta, H. Kanzaki, M. Hanafusa, K. Nagai, F. Matsui, H. Iishi, M. Tatsuta, Y. Ito, Intralesional steroid injection to prevent stricture after endoscopic submucosal dissection for esophageal cancer: a controlled prospective study, Endoscopy 44 (2012) 1007– 1011. K. Kawai, S. Suzuki, Y. Tabata, Y. Ikada, Y. Nishimura, Accelerated tissue regeneration through incorporation of basic fibroblast growth factorimpregnated gelatin microspheres into artificial dermis, Biomaterials 21 (2000) 489–499. W. Czaja, A. Krystynowicz, S. Bielecki, R.M. Brown Jr., Microbial cellulose–the natural power to heal wounds, Biomaterials 27 (2006) 145–151. Y. Matsumoto, Y. Kuroyanagi, Development of a wound dressing composed of hyaluronic acid sponge containing arginine and epidermal growth factor, J. Biomater. Sci. Polym. Ed. 21 (2010) 715–726. T. Kibe, T. Koga, K. Nishihara, T. Fuchigami, T. Yoshimura, T. Taguchi, N. Nakamura, Examination of the early wound healing process under different wound dressing conditions, Oral Surg. Oral Med. Oral Pathol. Oral Radiol. 123 (2017) 310–319. F. Yergoz, N. Hastar, C.E. Cimenci, A.D. Ozkan, T. Tekinay, M.O. Guler, A.B. Tekinay, Heparin mimetic peptide nanofiber gel promotes regeneration of full thickness burn injury, Biomaterials 134 (2017) 117–127. T. Ohki, M. Yamato, M. Ota, R. Takagi, D. Murakami, M. Kondo, R. Sasaki, H. Namiki, T. Okano, M. Yamamoto, Prevention of esophageal stricture after endoscopic submucosal dissection using tissue-engineered cell sheets, Gastroenterology 143 (2012) 582–588. C.I. Kwon, G. Kim, K.H. Ko, Y. Jung, I.K. Chung, S. Jeong, D.H. Lee, S.P. Hong, K.B. Hahm, Bio-sheet graft therapy for artificial gastric ulcer after endoscopic submucosal dissection: an animal feasibility study, Gastrointest. Endosc. 81 (2015) 989–996. K. Takimoto, T. Toyonaga, K. Matsuyama, Endoscopic tissue shielding to prevent delayed perforation associated with endoscopic submucosal dissection for duodenal neoplasms, Endoscopy 44 (2012) E414–E415. K. Takase, M. Aikawa, K. Okada, Y. Watanabe, K. Okamoto, H. Sato, K. Nonaka, S. Yamaguchi, S. Sakuramoto, I. Koyama, M. Miyazawa, Development of novel treatment with a bioabsorbable esophageal patch for benign esophageal stricture, Dis. Esophagus 28 (2015) 728–734. D. Kumar, V.L. Workman, M. O’Brien, J. McLaren, L. White, K. Ragunath, F. Rose, A. Saiani, J.E. Gough, Peptide hydrogels-a tissue engineering strategy for the prevention of oesophageal strictures, Adv. Funct. Mater. 27 (2017) 1702424. A. Nieponice, K. McGrath, I. Qureshi, E.J. Beckman, J.D. Luketich, T.W. Gilbert, S. F. Badylak, An extracellular matrix scaffold for esophageal stricture prevention after circumferential EMR, Gastrointest. Endosc. 69 (2009) 289–296. H. Fukuda, N. Yamaguchi, H. Isomoto, K. Matsushima, H. Minami, Y. Akazawa, K. Ohnita, F. Takeshima, S. Shikuwa, K. Nakao, Polyglycolic acid felt sealing method for prevention of bleeding related to endoscopic submucosal dissection in patients taking antithrombotic agents, Gastroenterol. Res. Pract. 2016 (2016) 1457357. C. Heinzmann, C. Weder, L.M. de Espinosa, Supramolecular polymer adhesives: advanced materials inspired by nature, Chem. Soc. Rev. 45 (2016) 342–358. R.J. Stewart, T.C. Ransom, V. Hlady, Natural underwater adhesives, J. Polym. Sci. B Polym. Phys. 49 (2011) 757–771.

[23] H. Lee, B.P. Lee, P.B. Messersmith, A reversible wet/dry adhesive inspired by mussels and geckos, Nature 448 (2007) 338–341. [24] H. Zhang, L.P. Bre, T. Zhao, Y. Zheng, B. Newland, W. Wang, Mussel-inspired hyperbranched poly(amino ester) polymer as strong wet tissue adhesive, Biomaterials 35 (2014) 711–719. [25] Y. Zhao, Y. Wu, L. Wang, M. Zhang, X. Chen, M. Liu, J. Fan, J. Liu, F. Zhou, Z. Wang, Bio-inspired reversible underwater adhesive, Nat. Commun. 8 (2017) 2218. [26] Q. Zhao, D.W. Lee, B.K. Ahn, S. Seo, Y. Kaufman, J.N. Israelachvili, J.H. Waite, Underwater contact adhesion and microarchitecture in polyelectrolyte complexes actuated by solvent exchange, Nat. Mater. 15 (2016) 407–412. [27] C. Zhong, T. Gurry, A.A. Cheng, J. Downey, Z. Deng, C.M. Stultz, T.K. Lu, Strong underwater adhesives made by self-assembling multi-protein nanofibres, Nat. Nanotechnol. 9 (2014) 858–866. [28] M.A. Gebbie, W. Wei, A.M. Schrader, T.R. Cristiani, H.A. Dobbs, M. Idso, B.F. Chmelka, J.H. Waite, J.N. Israelachvili, Tuning underwater adhesion with cation-pi interactions, Nat. Chem. 9 (2017) 473–479. [29] S. Rose, A. Prevoteau, P. Elziere, D. Hourdet, A. Marcellan, L. Leibler, Nanoparticle solutions as adhesives for gels and biological tissues, Nature 505 (2014) 382–385. [30] Q. Wen, R. Pelton, Design rules for microgel-supported adhesives, Ind. Eng. Chem. Res. 51 (2012) 9564–9570. [31] Y. Mizuno, R. Mizuta, M. Hashizume, T. Taguchi, Enhanced sealing strength of a hydrophobically-modified Alaska pollock gelatin-based sealant, Biomater. Sci. 5 (2017) 982–989. [32] T. Taguchi, R. Mizuta, T. Ito, K. Yoshizawa, M. Kajiyama, Robust sealing of blood vessels with cholesteryl group-modified, Alaska pollock-derived gelatin-based biodegradable sealant under wet conditions, J. Biomed. Nanotechnol. 12 (2016) 128–134. [33] K. Yoshizawa, R. Mizuta, T. Taguchi, Enhanced angiogenesis of growth factorfree porous biodegradable adhesive made with hexanoyl group-modified gelatin, Biomaterials 63 (2015) 14–23. [34] R. Mizuta, T. Taguchi, Enhanced sealing by hydrophobic modification of Alaska pollock-derived gelatin-based surgical sealants for the treatment of pulmonary air leaks, Macromol. Biosci. 17 (2017) 1600349. [35] A. Nishiguchi, F. Sasaki, H. Maeda, M. Kabayama, A. Ido, T. Taguchi, Multifunctional hydrophobized microparticles for accelerated wound healing after endoscopic submucosal dissection, Small 15 (2019) 1901566, https://doi.org/ 10.1002/small.201901566. [36] J. Adler-Nissen, Determination of the degree of hydrolysis of food protein hydrolysates by trinitrobenzenesulfonic acid, J. Agric. Food Chern. 27 (1979) 1256–1262. [37] S. Patra, P. Basak, D.N. Tibarewala, Synthesis of gelatin nano/submicron particles by binary nonsolvent aided coacervation (BNAC) method, Mater. Sci. Eng. C Mater. Biol. Appl. 59 (2016) 310–318. [38] C.J. Coester, K. Langer, H. van Briesen, J. Kreuter, Gelatin nanoparticles by two step desolvation–a new preparation method, surface modifications and cell uptake, J. Microencapsul. 17 (2000) 187–193. [39] S. Azarmi, Y. Huang, H. Chen, S. McQuarrie, D. Abrams, W. Roa, W.H. Finlay, G. G. Miller, R. Löbenberg, Optimization of a two-step desolvation method for preparing gelatin nanoparticles and cell uptake studies in 143B osteosarcoma cancer cells, J. Pharm. Pharmaceut. Sci. 9 (2006) 124–132. [40] T. Ikoma, H. Kobayashi, J. Tanaka, D. Walsh, S. Mann, Int. J. Biol. Macromol. 32 (2003) 199–204. [41] A.A. Karim, R. Bhat, Food Hydrocolloids 23 (2009) 563–576.

Please cite this article as: A. Nishiguchi, Y. Kurihara and T. Taguchi, Underwater-adhesive microparticle dressing composed of hydrophobically-modified Alaska pollock gelatin for gastrointestinal tract wound healing, Acta Biomaterialia, https://doi.org/10.1016/j.actbio.2019.08.040