# Fileset

[MDR Effect of Alaska pollock-gelatin sheet.pdf](https://mdr.nims.go.jp/filesets/b07d6845-d5b3-416f-a83c-acd26e573442/download)

## Creator

Ryosuke Tsujisaka, [Taku Suzuki](https://orcid.org/0000-0003-2312-3540), Shinsuke Shibata, Takuji Iwamoto, [Tetsushi Taguchi](https://orcid.org/0000-0003-2541-2530), Masaya Nakamura

## Rights

© The Author(s) 2024. <br>The article is protected by copyright and reuse is restricted to non-commercial and no derivative uses.<br>Tsujisaka R, Suzuki T, Shibata S, Iwamoto T, Taguchi T, Nakamura M. Effect of Alaska pollock-gelatin sheet on repair strength and regeneration of nerve. Journal of Hand Surgery (European Volume). 2024;50(1):76-84. doi:10.1177/17531934241251670[In Copyright](http://rightsstatements.org/vocab/InC/1.0/)

## Other metadata

[Effect of Alaska pollock-gelatin sheet on repair strength and regeneration of nerve](https://mdr.nims.go.jp/datasets/6d3e792c-ac49-4f8a-9628-9c9fd1c5d233)

## Fulltext

Effect of Alaska pollock-gelatin sheet sealant on bonding strength and regeneration of nerve   Ryosuke Tsujisaka, MD1, Taku Suzuki, MD, PhD1, Shinsuke Shibata, MD, PhD2, Takuji Iwamoto, MD, PhD1, Tetsushi Taguchi, PhD3,4, Masaya Nakamura, MD, PhD1   1. Department of Orthopedic Surgery, Keio University School of Medicine 35 Shinano-machi, Shinjuku, Tokyo 160-8582, Japan  2. Electron Microscope Laboratory, Keio University School of Medicine 35 Shinano-machi, Shinjuku, Tokyo 160-8582, Japan  3. Graduate School of Pure and Applied Sciences, University of Tsukuba 1-1-1 Tennodai, Tsukuba, Ibaraki 305-8577, Japan  4. Polymers and Biomaterials Field, Research Center for Functional Materials, National Institute for Materials Science 1-1 Namiki, Tsukuba, Ibaraki 305-0044, Japan    Corresponding authors: Taku Suzuki, MD, PhD  Department of Orthopedic Surgery, Keio University School of Medicine  35 Shinano-machi, Shinjuku-ku, Tokyo 160-8582, Japan Tel: +81-3-5363-3812 Title page (including authors' details)  E-mail: sutaku49@gmail.com  Tetsushi Taguchi, PhD Polymers and Biomaterials Field, Research Center for Functional Materials, National Institute for Materials Science 1-1 Namiki, Tsukuba, Ibaraki 305-0044, Japan  Tel: +81-29-860-4498 Email: TAGUCHI.Tetsushi@nims.go.jp  Keywords: Alaska pollock-gelatin; Alaska pollock-gelatin sealant; Alaska pollock-gelatin sheet; fibrin glue; nerve; sealant   Acknowledgements: We are grateful to Dr. N. Matsumura, N. Nagoshi, and Dr. S. Masuda in Department of Orthopaedic Surgery, Keio University School of Medicine, N. Moritoki in Electron Microscope Laboratory, Keio University School of Medicine, H. Ichimaru, and A. Nishiguchi in Polymers and Biomaterials Field, Research Center for Functional Materials, National Institute for Materials Science for their helpful support. We are grateful to Pro. Y. Kubota and Dr. N. Imanishi for approving the use of the cadavers donated to the Clinical Anatomy Laboratory, Keio University School of Medicine.  mailto:TAGUCHI.Tetsushi@nims.go.jp  Declaration of conflicting interests The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.  Funding This work was supported by the Japan Society for the Promotion of Science (JSPS) KAKENHI (Grant No. 22K09364).  Ethical approval Protocols for the cadaver experiments were approved by the Clinical Anatomy Laboratory with the consent of the families (approval no. 20150385), and animal experiments were approved by the Institutional Animal Care and Use Committee of our institution (approval no. A2022-016).   Informed consent The cadavers used in our study were donated to the Clinical Anatomy Laboratory, Keio University School of Medicine, with the consent of the families.  Contributorship details: All authors contributed to the study conception and design. R.T., T.S., S.S., and T.T. performed experimentation and data acquisition. R.T., T.S. and T.T. searched the scientific literature. R.T., T.S., S.S., and T.T. wrote the first draft of the manuscript. T.I., and M.N. revised the manuscript critically for important content. All authors read and approved the final manuscript.    1 Effect of Alaska pollock-gelatin sheet on repair strength and regeneration of nerve  1  2  3  4  5 ABSTRACT 6 This study aimed to investigate the repair strength and the biocompatibility of Alaska pollock-derived 7 gelatin (ApGltn) sheet for nerve repair. Cadaveric digital nerves were repaired with double suture, 8 single suture + ApGltn sheet, single suture + fibrin glue, single suture, ApGltn sheet, and fibrin, and 9 maximum failure loads were measured (20 nerves each). Rat sciatic nerves were repaired with double 10 suture, single suture + ApGltn sheet, single suture, ApGltn sheet, fibrin glue, and resection (10 nerves 11 each). Macroscopic appearance, muscle weight, and histopathological findings were examined 8 12 weeks postoperatively. The failure load of ApGltn sheet (0.39 N) was significantly higher than that 13 of a fibrin (0.05 N),and that of single suture + ApGltn sheet (1.32 N) was significantly higher than 14 that of a single suture alone (0.97 N). Functional and histological examinations showed similar 15 recovery among sutures, ApGltn, and fibrin groups. ApGltn sheet is useful for clinical application as 16 an alternative to fibrin. 17  18 19 Revised Manuscript  2 INTRODUCTION 20 Acute nerve injury commonly occurs due to upper limb trauma, and primary suture of the nerve 21 is a standard technique for its repair. Some materials, such as fibrin glue (Felix et al., 2013; Rafijah 22 et al., 2013), polyethylene glycol (Bamba et al., 2018; Riley et al., 2015), and laser welding (Barton 23 et al., 2013; Turner et al., 2018), enhance the bonding strength at the repair site (Barton et al., 2014). 24 Among these, fibrin sealant is the most frequently used material at nerve coaptation sites due to its 25 biocompatibility; however, the usefulness of fibrin addition remains controversial due to the lack of 26 bonding strength (Childe et al., 2018; Isaacs et al., 2008; Nishimura et al., 2008; Sameem et al., 2011; 27 Temple et al., 2004; Tse and Ko, 2012). 28 Recently, a novel biocompatible liquid-type sealant composed of Alaska pollock-derived 29 gelatin (ApGltn), partially modified with various alkyl groups and a polyethylene glycol-based 30 crosslinker, was introduced and demonstrated good burst strength when tested on porcine aorta and 31 rat lungs (Mizuno et al., 2017; Taguchi et al., 2016). The liquid-type ApGltn sealant also showed 32 higher bonding strength and an equal effect on nerve regeneration when compared with the fibrin 33 sealant using the digital nerve in cadaveric models and sciatic nerves in rat models. (Masuda et al., 34 2021). Furthermore, Taguchi et al. fabricated tissue-adhesive fibre sheets (ApGltn sheet) based on 35 decyl group-modified ApGltn (C10-ApGltn) by the electrospinning method (Ichimaru et al., 2021). 36 The burst strength, defined as the pressure at which the ApGltn sheets sealing the porcine pleura 37 ruptured as water pressure gradually increased, was 108 times higher than that of commercial 38 polyglycolic acid sheets (Ichimaru et al., 2021). Sheet-type adhesive ApGltn sealant may be clinically 39 Commented [TH1]: You need to make clear in the introduction how this study differs from your previous study:  Masuda S, Suzuki T, Shibata S et al. A novel Alaska pollock gelatin sealant shows higher bonding strength and nerve regeneration comparable to that of fibrin sealant in a cadaveric model and a rat model. Plast Reconstr Surg. 2021, 148: 742e-52e. Commented [鈴木2R1]: Thank you for your valuable comments. These studies (Mizuno et al., 2017; Taguchi et al., 2016) demonstrated the burst strength of ApGltn sealant using porcine aorta and rat lungs. In contrast, our previous study (Masuda et al., 2021) examined the bonding strength of ApGltn sealant utilizing the digital nerve in cadaveric models and the sciatic nerves in rat models. This distinction has been clarified in lines 31-35 on page 2.    3 easier to use for nerve repair than the liquid type because it can be stored at room temperature without 40 requiring any special apparatuses (Ichimaru and Taguchi, 2021). The ApGltn sheet is a promising 41 material for enhancing the bonding strength at the nerve repair site. However, whether it can increase 42 the bonding strength when tension is applied to the ruptured site and there is axonal regeneration 43 remains unclear. Therefore, this study aimed to investigate the bonding strength and biocompatibility 44 of this sheet type of sealant in transected digital nerves in a cadaveric model and sciatic nerves in a 45 rat model (Figure S1). 46  47  48 METHODS 49 In the cadaveric study, all procedures were carried out in accordance with the relevant 50 guidelines and regulations of the Clinical Anatomy Laboratory of our institution. All experimental 51 protocols were approved by the ethics committee (approval no. 20150385). Informed consent was 52 obtained from all participants and/or their legal guardians prior to death. For the animal study, all 53 experimental protocols were approved by Institutional Animal Care and Use Committee of our 54 institution (approval no. A2022-016).  55  56 Characteristics and preparation of the sealants 57 Manufacture of ApGltn sheets (Ichimaru et al., 2021) 58   4 C10-ApGltn was synthesized by reductive amination of amino groups in ApGltn with decanal, 59 as previously reported (Mizuno et al., 2017; Taguchi et al., 2016). ApGltn sheets composed of C10-60 ApGltn were fabricated by electrospinning. Briefly, 0.9 g of C10-ApGltn was dissolved in 3 mL of a 61 60% aqueous ethanol solution at 55°C. The solution was loaded into a syringe with an 18 G needle 62 and placed in an electrospinning machine (NANON-03, MECC Co., Ltd., Japan). The solution was 63 then extruded at a rate of 1 mL/h and an electrospinning voltage of 22 kV. The ApGltn sheets were 64 collected on silicone-coated aluminum membranes positioned 15 cm from the needle tip. To improve 65 stability under wet conditions, the obtained sheets were thermally cross-linked at 150°C for 5 h under 66 reduced pressure. Sheet thickness was 500µm, and they were bioresorbable within 4 weeks (Figure 67 S2). The microstructure of the fabricated ApGltn sheets was observed using scanning electron 68 microscopy (SEM; JSM-5600, JEOL Ltd., Japan) after sputtering with platinum for 5 min (Figure 69 S2). Thereafter, the ApGltn sheets were stored at room temperature until further use. 70  71 Fibrin glue  72 Fibrin glue (Beliplast P Combi-set, CSL Behring, PA, USA) used in this study was stored in a 73 refrigerator (4 °C) before use. Fibrinogen powder (40 mg) and coagulation factor XIII (30 IU) were 74 dissolved in Aprotinin solution (500 KIE/0.5 ml). Powder of thrombin concentrate (150 IU) was 75 dissolved in calcium chloride solution (2.94 g/0.5 ml). The fibrinogen and the thrombin solutions 76 were cured by mixing equal volumes of each solution.  77   5  78 Traction force testing using a cadaveric model 79 The primary outcome was the load to failure when traction force was applied to the repaired 80 digital nerves from freshly frozen cadavers. One hundred and twenty digital nerves from six freshly 81 frozen cadavers (mean age: 89 (SD 6) years; three women and three men) were used.  82  83 Surgical procedures for digital nerve repair 84 Digital nerves in a cadaveric model were selected, because our preliminary experiment showed 85 that the sciatic nerve size in mice or rats was too small for clamping to the traction machine (Masuda 86 et al., 2021). All surgical procedures were performed by a single-hand surgeon, according to a 87 previously reported method (Masuda et al., 2021). Radial and ulnar digital nerves (6 cm length) were 88 harvested from all five digits.  89 Each nerve segment was cut transversely at its midpoint. The nerves were repaired using six 90 techniques (20 nerves per group): (a) double suture, (b) single suture + ApGltn sheet, (c) single suture 91 + fibrin sealant, (d) single suture, (e) ApGltn sheet, and (f) fibrin sealant. To compare the failure load 92 using similar size nerves, the same cadavers were used for procedures (a) and (c), for (b) and (d), and 93 for (e) and (f). Digital nerves from the right hand were used for procedures (a), (b), and (e) and those 94 from the left hand for (c), (d), and (f). In procedures (a), (b), (c), and (d), the transected nerve was 95 repaired with a single- or double-epineural suture using an 8-0 monofilament nylon suture (Crownjun 96 KONO, Tokyo, Japan). In procedures (b) and (e), an ApGltn sheet of approximately 10 × 15 mm was 97   6 placed around the nerve repair site (Figure 1). In procedures (c) and (f), approximately 1 ml of fibrin 98 sealant was applied around the repair. The glue sleeve length was five times the width of the nerve 99 itself. 100  101 Biomechanical evaluation 102 Biomechanical strength was tested 5 to 10 min postoperatively, according to a previously 103 described method (Masuda et al., 2021). Approximately 1 cm segments of the proximal and distal 104 nerve ends were clamped to a material testing machine (Table-top Material Tensile Tester; EZ Graph, 105 Shimadzu Corporation, Kyoto, Japan) with a load range of 100 N. Nerves were pulled uniaxially at a 106 rate of 5 mm/min until terminal rupture occurred. The peak recorded load was considered as the 107 maximum failure load.  108  109 Functional evaluation using a rat model 110 The secondary outcome was the functional recovery of the repaired sciatic nerve using a rat 111 model. To prepare a sciatic-nerve injury model, 53 eight-week-old male Wistar rats (Sankyo Labo, 112 Tokyo, Japan) with a mean body weight of 198 (182–224) g at the time of surgery were used. 113  114 Surgical procedures for nerve repair 115   7 All rats were deeply anaesthetized with intraperitoneal ketamine (90 mg/kg; Sankyo, Tokyo, 116 Japan) and xylazine (10 mg/kg; Bayer, Leverkusen, Germany). The left leg was used as the 117 experimental limb, and the right leg was used for sham operation, or as a control. A dorsal 118 longitudinal skin incision was made, and the sciatic nerve was exposed by splitting the gluteal muscle. 119 The nerves were treated with seven surgical interventions: (a) double suture, (b) single suture + 120 ApGltn sheet, (c) single suture, (d) ApGltn sheet, (e) fibrin sealant, (f) resection of the nerve with a 121 5-mm segmental defect, and (g) sham (placebo) operation (10 nerves each for [a] to [f] and three 122 nerves for [g]). Procedures employing suture + fibrin sealant was not performed in the rat model, 123 since the results of biomechanical traction testing showed that there was no significant difference 124 between the procedures employing suture and those employing suture + fibrin sealant. In procedures 125 (a)–(f), the sciatic nerve segment was cut transversely at the midpoint of the nerve. In procedures (a)–126 (c), the transected nerve was repaired with a single- or double-epineural suture using a 9-0 127 monofilament nylon suture (Crownjun KONO, Tokyo, Japan). In procedures (b) and (d), ApGltn 128 sheet of approximately 10 × 5 mm was placed around the nerve division site. In procedures (e), 129 approximately 0.1 ml of fibrin adhesive was placed around the nerve rupture site. When the ApGltn 130 sheet and fibrin sealant were applied to the nerve, a plastic sheet was laid beneath the nerve to avoid 131 adherence of the sealant to the nerve bed. During the sham operations, sciatic nerves were explored 132 without damaging them. After surgery, the treated limbs were not immobilized, and the rats were 133 allowed unrestricted motion. 134   8 To evaluate nerve regeneration, walking track analysis was performed every two weeks until 8 135 weeks after the procedure, when macroscopic examination, muscle weight measurement, and 136 histological examination were conducted. 137  138 Macroscopic examination 139 The sciatic nerve was exposed using a procedure similar to that described previously. The 140 macroscopic appearance of the sciatic nerve, including nerve continuity and ApGltn sheet absorption, 141 was confirmed. Nerve continuity was defined as complete continuity (continuity with normal nerve 142 thickness), incomplete continuity (continuity with nerve narrowing), or complete rupture (visible 143 separation at the coaptation site) (Masuda et al., 2021). 144  145 Muscle weight measurement 146 Both tibialis anterior muscles were harvested and weighed. Muscle recovery was calculated by 147 comparing the weights of the experimental and control limbs.  148  149 Walking track analysis 150 Walking track analysis was performed to evaluate motor function. Rats were placed on a 151 treadmill, and their footprints were scanned using the DigiGait System (Rat Specifics, MA, USA). 152   9 The sciatic functional index (SFI) was calculated according to a previously reported formula (Bain et 153 al., 1989). The zero value of the SFI denotes normal nerve function, and a value of −100 represents 154 total loss of nerve function. 155  156 Histological examination 157 Three out of ten sciatic nerves from each group were examined histologically. To assess the 158 histological changes in the axons and myelin sheaths, specimens from the sciatic nerves 3 mm distal 159 to the repair site in each group were obtained for electron and light microscopy (Kimura et al., 2018; 160 Shibata et al., 2015). Semi-thin sections of 1 μm thickness were stained with 0.1% toluidine blue for 161 7 min and imaged using a BZ-X700 optical microscope (Keyence, Osaka, Japan). Ultrathin axial 162 sections (70 nm thickness) of the sciatic nerve were prepared using an ultramicrotome (Leica UC7, 163 Leica Microsystems GmbH, Wetzlar, Germany) on silicon wafers and stained with uranyl acetate and 164 lead citrate for 10 min. The sections were observed under a SU6600 SEM (Hitachi High-Tech, Tokyo, 165 Japan) by detecting backscattered electrons with an acceleration voltage of 5 kV. For quantitative 166 analysis, axonal density was used to evaluate the regenerated nerve fibres. Axonal density was 167 defined as axonal area/total area, and myelin sheath density was defined as myelin sheath area/total 168 area in a fascicle in each nerve sample (Takagi et al., 2009). For G-ratio quantification, 300 fibres 169 (100 from each nerve) randomly selected from electron microscopy images were used. 170  171 Statistical analysis 172   10 The Shapiro-Wilk test was used to assess the normality of data. Data are presented as mean 173 with standard error. To evaluate intergroup differences, one-way analysis of variance (ANOVA) with 174 Tukey or Games-Howell post hoc comparisons were used for maximum failure load, muscle weight, 175 and axonal density analysis. The Kruskal–Wallis and Dann–Bonferroni tests for post hoc comparisons 176 were used for axon diameter analyses. Two-way repeated ANOVA was used for the comparison 177 between SFI values. Statistical analysis was performed by a hand surgeon who was blinded to clinical 178 information. Post-hoc power analysis was conducted to confirm whether this sample size would be 179 adequate to detect a significant difference, with an alpha of 0.05. Effect sizes are expressed as mean 180 values with standard deviations. The power analysis demonstrated statistical powers of 100%, 97%, 181 98%, and 100% for the cadaveric model, muscle weight, SFI, and G-ratio, respectively. Statistical 182 significance was set at p < 0.05. 183  184  185 RESULTS 186 Failure load using the cadaveric model 187 The maximum failure load for each procedure is shown in figure 2. The maximum failure load 188 of ApGltn sheet was significantly higher than that of a fibrin sealant (0.39 N vs. 0.05 N, p < 0.001). 189 The maximum failure load of single suture + ApGltn sheet was significantly higher than that of a 190 single suture (1.32 N vs. 0.97 N, p = 0.02). There were no significant differences between single 191 suture + fibrin sealant and single suture (0.99 N vs. 0.97 N, p = 0.99). The double suture technique 192   11 (1.65 N) had a higher maximum failure load than the single suture + ApGltn sheet technique, but this 193 was not significant (p = 0.07). The maximum failure load of the ApGltn sheet (0.39 N) and fibrin 194 sealant (0.05 N) was significantly lower than those of the other four procedures (p < 0.001).  195  196 Functional evaluation using the rat model 197 Macroscopic examination 198 All resected sciatic nerves in the double suture, single suture + ApGltn sheet, single suture, and 199 ApGltn sheet groups showed complete nerve continuity without defects. The ApGltn sheets were 200 resorbed and had disappeared by 8 weeks after the initial procedure. Fibrin groups showed complete 201 continuity of the nerves in 9 rats and incomplete continuity in 1 rat. In the resection group, complete 202 rupture was observed in five of ten rats, complete continuity in two, and incomplete continuity in 203 three rats (Figure S3).  204  205 Muscle weight 206  Muscle weight recovery did not significantly differ among the double suture, single suture + 207 ApGltn sheet, single suture, ApGltn sheet, and fibrin techniques (p > 0.05) (Figure 3). However, a 208 significant difference was observed between the resection group and the double suture, single suture 209 + ApGltn sheet, single suture, and ApGltn sheet groups (p < 0.05). 210  211   12 Walking track analysis 212 SFI improved over time; the double suture and single suture + ApGltn sheet groups showed 213 significantly higher SFI values than the ApGltn sheet, single suture, fibrin, and resection groups (p < 214 0.05) (Figure 4). The ApGltn sheet groups showed significantly higher SFI values than the resection 215 group (p < 0.05).  216  217 Histological examination 218 Optical and electron microscopy images of the toluidine blue-stained nerves fibre are shown in 219 figure 5. In axial sections, a larger number of myelinated axons were observed in the double suture, 220 single suture + ApGltn sheet, single suture, ApGltn sheet, and fibrin groups than in the resection 221 group. Quantitative analysis of the axonal density indicated that the double suture and single suture 222 + ApGltn sheet groups had significantly higher values than the fibrin group (Figure 6). No significant 223 differences were observed in axonal density among the double suture, single suture + ApGltn sheet, 224 ApGltn sheet, and single suture groups.  225 The G-ratios are shown in figure S4. Although the double suture group showed thicker myelin 226 regeneration than the single suture + ApGltn sheet group (Figure S4a), the axonal diameters did not 227 significantly differ between the two groups (Figure S5). Similar axonal regrowth patterns were 228 observed in the single suture + ApGltn sheet and single suture groups (Figure S4b) and the single 229 suture and ApGltn sheet groups (Figure S4c); the axonal diameters among the three groups did not 230 significantly differ (Figure S5). The ApGltn sheet group showed thicker myelin regeneration than the 231   13 fibrin group (Figure S4d). The axonal diameters indicated that the double suture, single suture + 232 ApGltn sheet, single suture, and ApGltn sheet groups had significantly higher values than the fibrin 233 group. 234  235  236 DISCUSSION 237 In this study, the bonding strength and functional recovery of transected nerves using ApGltn 238 sheets were compared with traditional sutures and fibrin sealant in cadaveric and rat models. The 239 maximum failure load of the ApGltn sheet was approximately eight times higher than that of the 240 fibrin sealant. Although the bonding strength of the ApGltn sheet was inferior to that of the traditional 241 suture, the addition of the ApGltn sheet to the single suture significantly reinforced the repair strength. 242 Addition of fibrin sealant to the single suture did not significantly reinforce the strength. The ApGltn 243 sheet did not compromise sciatic nerve regeneration compared with the traditional suture. 244 Liquid ApGltn sealant has a stronger breaking strength than fibrin sealant (Masuda et al., 2021; 245 Mizuno et al., 2017; Taguchi et al., 2016; Yamaoka et al., 2019). Taguchi et al. (2016) demonstrated 246 that the burst strength of the ApGltn sealant was 11.6 times higher than that of a fibrin sealant in a 247 burst porcine aorta model (341 vs. 29 mm Hg). Additionally, Masuda et al. (2021) showed that the 248 maximum failure load of liquid-type ApGltn sealant was approximately three times higher than that 249 of the fibrin sealant using a cadaveric digital nerve model (0.22 versus 0.06 N). Although a direct 250 comparison between that study and ours is difficult due to the different digital nerve sizes from 251   14 different specimens employed, the breaking strength of the ApGltn sheet (0.39 N) is higher than that 252 of the liquid-type ApGltn sealant. However, the failure load of the ApGltn sheet was inferior to that 253 of the traditional suture, as also shown in a previous report on liquid ApGltn sealant (Masuda et al., 254 2021).  255 To minimize the effect of size differences, a previous cadaveric model similar to our study 256 compared bilateral digital nerves from the same specimen (Masuda et al., 2021). Addition of liquid 257 ApGltn sealant to the single suture did not significantly increase the maximum failure load (mean 0.1 258 N) compared to the single suture (Masuda et al., 2021); therefore, they concluded that liquid ApGltn 259 sealant with the reported bonding strength cannot be used clinically, and further improvement of the 260 strength in the ApGltn sealant is required. Our study showed that the addition of an ApGltn sheet to 261 a single suture does significantly reinforce the bonding strength (mean 0.35 N). Fibrin glue is 262 sometimes added to the nerve suture site (Isaacs, 2010); however, our results are consistent with the 263 previous findings that the addition of fibrin glues does not significantly increase the maximum failure 264 load (Isaacs et al., 2008). The ApGltn sealant, compared to the fibrin sealant, has been shown to have 265 prolonged adhesive capacity and reduced viral infection risk and cost (Nishimura et al., 2008; Taguchi 266 et al., 2016). Since ApGltn is derived from the waste product of Alaska pollack skin, the cost of 267 materials could be quite low. Therefore, ApGltn sheets could be clinically useful in the future as an 268 alternative to fibrin sealants.  269 In the present study, sciatic nerves repaired with sheet-type ApGltn sealant showed nerve 270 recovery similar to that in the suture group. This result is consistent with that of a previous study 271   15 using liquid ApGltn sealant, showing similar nerve recovery compared to suture and fibrin sealant 272 repair in a rat sciatic nerve model (Masuda et al., 2021). Furthermore, sheet-type ApGltn sealants 273 completely degrade within 21 days without severe inflammation when subcutaneously implanted in 274 the backs of rats (Ichimaru et al., 2021). In an in vitro study, ApGltn sheet had excellent 275 cytocompatibility and efficiently supported the growth of L929 cells (Ichimaru and Taguchi, 2021). 276 Moreover, since it is composed of gelatin, ApGltn sheets do not prevent tissue regeneration.  277 In addition to the bonding strength, the ApGltn sheet has other advantages over liquid ApGltn 278 sealant. Liquid-type adhesives need to mix two components homogeneously; therefore, they require 279 special apparatuses to rapidly mix the components during surgery (Mizuno et al., 2017). Sheet-type 280 adhesives have been developed for their easy sealing property without any special apparatus. 281 Additionally, the liquid type requires refrigeration before use, whereas the sheet type can be stored at 282 room temperature. Furthermore, liquid-type adhesives are difficult to apply evenly around the entire 283 circumference of the suture site in clinical practice, whereas the sheet-type are more evenly applied. 284 Altogether, the ApGltn sheet is clinically easier to use for nerve repair than the liquid type.  285 Our study had some limitations. The maximum failure load of a repaired nerve could differ 286 according to the digital nerve size. To minimize the effect of size differences, the same cadavers were 287 used for the single suture + ApGltn sheet (b) and single suture procedures (d); the double suture (a) 288 and single suture + fibrin sealant (c) procedures; and ApGltn sheet (e) and fibrin sealant (f) procedures. 289 However, the different cadavers used for the procedures might have affected the results. Nerve 290 regeneration was studied in an animal model only. Therefore, clinical trials are needed to validate the 291   16 application of this sealant in humans. 292 In conclusion, the maximum failure load of the ApGltn sheet was higher than that of the fibrin 293 sealant. Addition of the ApGltn sheet to a single suture significantly reinforces the repair strength of 294 the nerve. Its effect on transected nerve regeneration is similar to that of the traditional suture. 295 Although ApGltn sheet cannot serve as a substitute for sutures, we believe that addition of ApGltn 296 sheets to the nerve suture site will be useful for clinical application in the future as an alternative to 297 fibrin sealant.  298  299  300  301 REFERENCES 302 Bain JR, Mackinnon SE, Hunter DA. Functional evaluation of complete sciatic, peroneal, and 303 posterior tibial nerve lesions in the rat. Plast Reconstr Surg. 1989, 83: 129-38. 304 Bamba R, Riley DC, Kim JS et al. Evaluation of a nerve fusion technique with polyethylene glycol 305 in a delayed setting after nerve injury. J Hand Surg Am. 2018, 43: 82.e1-.e7. 306 Barton M, Morley JW, Stoodley MA et al. Laser-activated adhesive films for sutureless median nerve 307 anastomosis. J Biophotonics. 2013, 6: 938-49. 308 Barton MJ, Morley JW, Stoodley MA, Lauto A, Mahns DA. Nerve repair: Toward a sutureless 309 approach. Neurosurg Rev. 2014, 37: 585-95. 310 Childe JR, Regal S, Schimoler P, Kharlamov A, Miller MC, Tang P. Fibrin glue increases the tensile 311 strength of conduit-assisted primary digital nerve repair. Hand (N Y). 2018, 13: 45-9. 312   17 Felix SP, Pereira Lopes FR, Marques SA, Martinez AM. Comparison between suture and fibrin glue 313 on repair by direct coaptation or tubulization of injured mouse sciatic nerve. Microsurgery. 2013, 33: 314 468-77. 315 Ichimaru H, Mizuno Y, Chen X, Nishiguchi A, Taguchi T. Prevention of pulmonary air leaks using a 316 biodegradable tissue-adhesive fibre sheet based on alaska pollock gelatin modified with decanyl 317 groups. Biomater Sci. 2021, 9: 861-73. 318 Ichimaru H, Taguchi T. Improved tissue adhesion property of a hydrophobically modified alaska 319 pollock derived gelatin sheet by uv treatment. Int J Biol Macromol. 2021, 172: 580-8. 320 Isaacs J. Treatment of acute peripheral nerve injuries: Current concepts. J Hand Surg Am. 2010, 35: 321 491-7. 322 Isaacs JE, McDaniel CO, Owen JR, Wayne JS. Comparative analysis of biomechanical performance 323 of available "nerve glues". J Hand Surg Am. 2008, 33: 893-9. 324 Kimura H, Ouchi T, Shibata S et al. Stem cells purified from human induced pluripotent stem cell-325 derived neural crest-like cells promote peripheral nerve regeneration. Sci Rep. 2018, 8: 10071. 326 Masuda S, Suzuki T, Shibata S et al. A novel Alaska pollock gelatin sealant shows higher bonding 327 strength and nerve regeneration comparable to that of fibrin sealant in a cadaveric model and a rat 328 model. Plast Reconstr Surg. 2021, 148: 742e-52e. 329 Mizuno Y, Mizuta R, Hashizume M, Taguchi T. Enhanced sealing strength of a hydrophobically-330 modified alaska pollock gelatin-based sealant. Biomater Sci. 2017, 5: 982-9. 331 Mizuno Y, Taguchi T. Promotion of cell migration into a hydrophobically modified alaska pollock 332   18 gelatin-based hydrogel. Macromol Biosci. 2019, 19: e1900083. 333 Mizuta R, Taguchi T. Enhanced sealing by hydrophobic modification of alaska pollock-derived 334 gelatin-based surgical sealants for the treatment of pulmonary air leaks. Macromol Biosci. 2017, 17. 335 Nishimura MT, Mazzer N, Barbieri CH, Moro CA. Mechanical resistance of peripheral nerve repair 336 with biological glue and with conventional suture at different postoperative times. J Reconstr 337 Microsurg. 2008, 24: 327-32. 338 Rafijah G, Bowen AJ, Dolores C, Vitali R, Mozaffar T, Gupta R. The effects of adjuvant fibrin sealant 339 on the surgical repair of segmental nerve defects in an animal model. J Hand Surg Am. 2013, 38: 847-340 55. 341 Riley DC, Bittner GD, Mikesh M et al. Polyethylene glycol-fused allografts produce rapid behavioral 342 recovery after ablation of sciatic nerve segments. J Neurosci Res. 2015, 93: 572-83. 343 Sameem M, Wood TJ, Bain JR. A systematic review on the use of fibrin glue for peripheral nerve 344 repair. Plast Reconstr Surg. 2011, 127: 2381-90. 345 Shibata S, Murota Y, Nishimoto Y et al. Immuno-electron microscopy and electron microscopic in 346 situ hybridization for visualizing pirna biogenesis bodies in drosophila ovaries. Methods Mol Biol. 347 2015, 1328: 163-78. 348 Taguchi T, Ryo M, Ito T, Yoshizawa K, Kajiyama M. Robust sealing of blood vessels with cholesteryl 349 group-modified, alaska pollock-derived gelatin-based biodegradable sealant under wet conditions. J 350 Biomed Nanotechnol. 2016, 12: 128-34. 351 Takagi T, Nakamura M, Yamada M et al. Visualization of peripheral nerve degeneration and 352   19 regeneration: Monitoring with diffusion tensor tractography. Neuroimage. 2009, 44: 884-92. 353 Temple CL, Ross DC, Dunning CE, Johnson JA. Resistance to disruption and gapping of peripheral 354 nerve repairs: An in vitro biomechanical assessment of techniques. J Reconstr Microsurg. 2004, 20: 355 645-50. 356 Tse R, Ko JH. Nerve glue for upper extremity reconstruction. Hand Clin. 2012, 28: 529-40. 357 Turner NJ, Johnson SA, Foster LJR, Badylak SF. Sutureless nerve repair with ecm bioscaffolds and 358 laser-activated chitosan adhesive. J Biomed Mater Res B Appl Biomater. 2018, 106: 1698-711. 359 Yamaoka M, Maki N, Wijesinghe A et al. Novel alaska pollock gelatin sealant shows high adhesive 360 quality and conformability. Ann Thorac Surg. 2019, 107: 1656-62. 361  362  363  364  365 FIGURE LEGENDS 366 Figure 1. Resected digital nerves repaired with single suture + ApGltn sheet. An approximately 10 × 367 15 mm sheet was applied around the nerve rupture site. 368  369 Figure 2. The maximum failure load for each procedure. The failure load of the single suture + 370 ApGltn sheet was significantly higher than that of the single suture and that of ApGltn sheet was 371 significantly higher than that of a fibrin sealant. Each box represents the interquartile range of values, 372   20 with the bold line showing the median value. The vertical lines show maximum and minimum values. 373 *p < 0.05, **p < 0.001 374  375 Figure 3. Muscle weight recovery of the tibialis anterior in the experimental limbs compared with 376 that in the control limbs. No significant differences are observed among the procedures employing 377 double sutures, single sutures + ApGltn sheets, single sutures, ApGltn sheets, and fibrin techniques. 378 Each box represents the interquartile range of values, with the bold line showing the median value. 379 The vertical lines show maximum and minimum values. *p < 0.05, **p < 0.001 380  381 Figure 4. Sciatic functional index in each procedure until the completion of 8 postoperative weeks 382 at 4, 6, and 8 weeks, the double suture and single suture + ApGltn sheet groups show significantly 383 higher values than the ApGltn sheet, single suture, fibrin, and resection groups. Data are presented as 384 mean with standard deviation. 385  386 Figure 5. Axial sections of the sciatic nerve: light microscopy with toluidine blue (above) and 387 electron microscopy with uranyl acetate (below) staining. A larger number of myelinated axons are 388 observed in the double suture, single suture + ApGltn sheet, single suture, ApGltn sheet, fibrin groups 389 than in the negative control resection group. 390  391   21 Figure 6. Quantitative analysis of axonal density of the regenerated nerve fibrefibres. The double 392 suture and single suture + ApGltn sheet groups had significantly higher values than the fibrin sealant. 393 No significant differences were observed among the double suture, single suture + ApGltn sheet, 394 ApGltn sheet, and single suture. Each box represents the interquartile range of values, with the bold 395 line showing the median value. The vertical lines show maximum and minimum values. *p < 0.05, 396 **p < 0.001 397  398 Figure S1. Biomechanical traction testing using transected digital nerves in a cadaveric model and 399 functional testing using transected sciatic nerves in a rat model. 400  401 Figure S2. Photograph of prepared ApGltn sheet (left) and scanning electron microscopy images of 402 ApGltn sheet (right). 403  404 Figure S3. Macroscopic examination of the resected sciatic nerves. All resected sciatic nerves in the 405 double suture, single suture + ApGltn sheet, single suture, ApGltn sheet, and fibrin groups show nerve 406 continuity, while segmental defects are observed in five out of ten rats in the resection group. 407  408 Figure S4. The G-ratio and axon diameter results. The approximate linear regression equation for 409 each group is shown in the graph.  410   22 (a) Double suture and single suture + ApGltn sheet  411 (b) Single suture + ApGltn sheet and single suture  412 (c) Single suture and ApGltn sheet  413 (d) ApGltn sheet and fibrin sealant 414  415 Figure S5. Axonal diameter of the regenerated nerve fibres. The axonal diameters indicated that the 416 double suture, single suture + ApGltn sheet, single suture, and ApGltn sheet groups had significantly 417 higher values than the fibrin group. Each box represents the interquartile range of values, with the 418 bold line showing the median value. The vertical lines show maximum and minimum values. *p < 419 0.05, **p < 0.001 420  421   1 Effect of Alaska pollock-gelatin sheet on repair strength and regeneration of nerve  1  2  3  4  5 ABSTRACT 6 This study aimed to investigate the repair strength and the biocompatibility of Alaska pollock-derived 7 gelatin (ApGltn) sheet for nerve repair. Cadaveric digital nerves were repaired with double suture, 8 single suture + ApGltn sheet, single suture + fibrin glue, single suture, ApGltn sheet, and fibrin, and 9 maximum failure loads were measured (20 nerves each). Rat sciatic nerves were repaired with double 10 suture, single suture + ApGltn sheet, single suture, ApGltn sheet, fibrin glue, and resection (10 nerves 11 each). Macroscopic appearance, muscle weight, and histopathological findings were examined 8 12 weeks postoperatively. The failure load of ApGltn sheet (0.39 N) was significantly higher than that 13 of a fibrin (0.05 N),and that of single suture + ApGltn sheet (1.32 N) was significantly higher than 14 that of a single suture alone (0.97 N). Functional and histological examinations showed similar 15 recovery among sutures, ApGltn, and fibrin groups. ApGltn sheet is useful for clinical application as 16 an alternative to fibrin. 17  18 19 Revised Manuscript Clean  2 INTRODUCTION 20 Acute nerve injury commonly occurs due to upper limb trauma, and primary suture of the nerve 21 is a standard technique for its repair. Some materials, such as fibrin glue (Felix et al., 2013; Rafijah 22 et al., 2013), polyethylene glycol (Bamba et al., 2018; Riley et al., 2015), and laser welding (Barton 23 et al., 2013; Turner et al., 2018), enhance the bonding strength at the repair site (Barton et al., 2014). 24 Among these, fibrin sealant is the most frequently used material at nerve coaptation sites due to its 25 biocompatibility; however, the usefulness of fibrin addition remains controversial due to the lack of 26 bonding strength (Childe et al., 2018; Isaacs et al., 2008; Nishimura et al., 2008; Sameem et al., 2011; 27 Temple et al., 2004; Tse and Ko, 2012). 28 Recently, a novel biocompatible liquid-type sealant composed of Alaska pollock-derived 29 gelatin (ApGltn), partially modified with various alkyl groups and a polyethylene glycol-based 30 crosslinker, was introduced and demonstrated good burst strength when tested on porcine aorta and 31 rat lungs (Mizuno et al., 2017; Taguchi et al., 2016). The liquid-type ApGltn sealant also showed 32 higher bonding strength and an equal effect on nerve regeneration when compared with the fibrin 33 sealant using the digital nerve in cadaveric models and sciatic nerves in rat models. (Masuda et al., 34 2021). Furthermore, Taguchi et al. fabricated tissue-adhesive fibre sheets (ApGltn sheet) based on 35 decyl group-modified ApGltn (C10-ApGltn) by the electrospinning method (Ichimaru et al., 2021). 36 The burst strength, defined as the pressure at which the ApGltn sheets sealing the porcine pleura 37 ruptured as water pressure gradually increased, was 108 times higher than that of commercial 38 polyglycolic acid sheets (Ichimaru et al., 2021). Sheet-type adhesive ApGltn sealant may be clinically 39   3 easier to use for nerve repair than the liquid type because it can be stored at room temperature without 40 requiring any special apparatuses (Ichimaru and Taguchi, 2021). The ApGltn sheet is a promising 41 material for enhancing the bonding strength at the nerve repair site. However, whether it can increase 42 the bonding strength when tension is applied to the ruptured site and there is axonal regeneration 43 remains unclear. Therefore, this study aimed to investigate the bonding strength and biocompatibility 44 of this sheet type of sealant in transected digital nerves in a cadaveric model and sciatic nerves in a 45 rat model (Figure S1). 46  47  48 METHODS 49 In the cadaveric study, all procedures were carried out in accordance with the relevant 50 guidelines and regulations of the Clinical Anatomy Laboratory of our institution. All experimental 51 protocols were approved by the ethics committee (approval no. 20150385). Informed consent was 52 obtained from all participants and/or their legal guardians prior to death. For the animal study, all 53 experimental protocols were approved by Institutional Animal Care and Use Committee of our 54 institution (approval no. A2022-016).  55  56 Characteristics and preparation of the sealants 57 Manufacture of ApGltn sheets (Ichimaru et al., 2021) 58   4 C10-ApGltn was synthesized by reductive amination of amino groups in ApGltn with decanal, 59 as previously reported (Mizuno et al., 2017; Taguchi et al., 2016). ApGltn sheets composed of C10-60 ApGltn were fabricated by electrospinning. Briefly, 0.9 g of C10-ApGltn was dissolved in 3 mL of a 61 60% aqueous ethanol solution at 55°C. The solution was loaded into a syringe with an 18 G needle 62 and placed in an electrospinning machine (NANON-03, MECC Co., Ltd., Japan). The solution was 63 then extruded at a rate of 1 mL/h and an electrospinning voltage of 22 kV. The ApGltn sheets were 64 collected on silicone-coated aluminum membranes positioned 15 cm from the needle tip. To improve 65 stability under wet conditions, the obtained sheets were thermally cross-linked at 150°C for 5 h under 66 reduced pressure. Sheet thickness was 500µm, and they were bioresorbable within 4 weeks (Figure 67 S2). The microstructure of the fabricated ApGltn sheets was observed using scanning electron 68 microscopy (SEM; JSM-5600, JEOL Ltd., Japan) after sputtering with platinum for 5 min (Figure 69 S2). Thereafter, the ApGltn sheets were stored at room temperature until further use. 70  71 Fibrin glue  72 Fibrin glue (Beliplast P Combi-set, CSL Behring, PA, USA) used in this study was stored in a 73 refrigerator (4 °C) before use. Fibrinogen powder (40 mg) and coagulation factor XIII (30 IU) were 74 dissolved in Aprotinin solution (500 KIE/0.5 ml). Powder of thrombin concentrate (150 IU) was 75 dissolved in calcium chloride solution (2.94 g/0.5 ml). The fibrinogen and the thrombin solutions 76 were cured by mixing equal volumes of each solution.  77   5  78 Traction force testing using a cadaveric model 79 The primary outcome was the load to failure when traction force was applied to the repaired 80 digital nerves from freshly frozen cadavers. One hundred and twenty digital nerves from six freshly 81 frozen cadavers (mean age: 89 (SD 6) years; three women and three men) were used.  82  83 Surgical procedures for digital nerve repair 84 Digital nerves in a cadaveric model were selected, because our preliminary experiment showed 85 that the sciatic nerve size in mice or rats was too small for clamping to the traction machine (Masuda 86 et al., 2021). All surgical procedures were performed by a single-hand surgeon, according to a 87 previously reported method (Masuda et al., 2021). Radial and ulnar digital nerves (6 cm length) were 88 harvested from all five digits.  89 Each nerve segment was cut transversely at its midpoint. The nerves were repaired using six 90 techniques (20 nerves per group): (a) double suture, (b) single suture + ApGltn sheet, (c) single suture 91 + fibrin sealant, (d) single suture, (e) ApGltn sheet, and (f) fibrin sealant. To compare the failure load 92 using similar size nerves, the same cadavers were used for procedures (a) and (c), for (b) and (d), and 93 for (e) and (f). Digital nerves from the right hand were used for procedures (a), (b), and (e) and those 94 from the left hand for (c), (d), and (f). In procedures (a), (b), (c), and (d), the transected nerve was 95 repaired with a single- or double-epineural suture using an 8-0 monofilament nylon suture (Crownjun 96 KONO, Tokyo, Japan). In procedures (b) and (e), an ApGltn sheet of approximately 10 × 15 mm was 97   6 placed around the nerve repair site (Figure 1). In procedures (c) and (f), approximately 1 ml of fibrin 98 sealant was applied around the repair. The glue sleeve length was five times the width of the nerve 99 itself. 100  101 Biomechanical evaluation 102 Biomechanical strength was tested 5 to 10 min postoperatively, according to a previously 103 described method (Masuda et al., 2021). Approximately 1 cm segments of the proximal and distal 104 nerve ends were clamped to a material testing machine (Table-top Material Tensile Tester; EZ Graph, 105 Shimadzu Corporation, Kyoto, Japan) with a load range of 100 N. Nerves were pulled uniaxially at a 106 rate of 5 mm/min until terminal rupture occurred. The peak recorded load was considered as the 107 maximum failure load.  108  109 Functional evaluation using a rat model 110 The secondary outcome was the functional recovery of the repaired sciatic nerve using a rat 111 model. To prepare a sciatic-nerve injury model, 53 eight-week-old male Wistar rats (Sankyo Labo, 112 Tokyo, Japan) with a mean body weight of 198 (182–224) g at the time of surgery were used. 113  114 Surgical procedures for nerve repair 115   7 All rats were deeply anaesthetized with intraperitoneal ketamine (90 mg/kg; Sankyo, Tokyo, 116 Japan) and xylazine (10 mg/kg; Bayer, Leverkusen, Germany). The left leg was used as the 117 experimental limb, and the right leg was used for sham operation, or as a control. A dorsal 118 longitudinal skin incision was made, and the sciatic nerve was exposed by splitting the gluteal muscle. 119 The nerves were treated with seven surgical interventions: (a) double suture, (b) single suture + 120 ApGltn sheet, (c) single suture, (d) ApGltn sheet, (e) fibrin sealant, (f) resection of the nerve with a 121 5-mm segmental defect, and (g) sham (placebo) operation (10 nerves each for [a] to [f] and three 122 nerves for [g]). Procedures employing suture + fibrin sealant was not performed in the rat model, 123 since the results of biomechanical traction testing showed that there was no significant difference 124 between the procedures employing suture and those employing suture + fibrin sealant. In procedures 125 (a)–(f), the sciatic nerve segment was cut transversely at the midpoint of the nerve. In procedures (a)–126 (c), the transected nerve was repaired with a single- or double-epineural suture using a 9-0 127 monofilament nylon suture (Crownjun KONO, Tokyo, Japan). In procedures (b) and (d), ApGltn 128 sheet of approximately 10 × 5 mm was placed around the nerve division site. In procedures (e), 129 approximately 0.1 ml of fibrin adhesive was placed around the nerve rupture site. When the ApGltn 130 sheet and fibrin sealant were applied to the nerve, a plastic sheet was laid beneath the nerve to avoid 131 adherence of the sealant to the nerve bed. During the sham operations, sciatic nerves were explored 132 without damaging them. After surgery, the treated limbs were not immobilized, and the rats were 133 allowed unrestricted motion. 134   8 To evaluate nerve regeneration, walking track analysis was performed every two weeks until 8 135 weeks after the procedure, when macroscopic examination, muscle weight measurement, and 136 histological examination were conducted. 137  138 Macroscopic examination 139 The sciatic nerve was exposed using a procedure similar to that described previously. The 140 macroscopic appearance of the sciatic nerve, including nerve continuity and ApGltn sheet absorption, 141 was confirmed. Nerve continuity was defined as complete continuity (continuity with normal nerve 142 thickness), incomplete continuity (continuity with nerve narrowing), or complete rupture (visible 143 separation at the coaptation site) (Masuda et al., 2021). 144  145 Muscle weight measurement 146 Both tibialis anterior muscles were harvested and weighed. Muscle recovery was calculated by 147 comparing the weights of the experimental and control limbs.  148  149 Walking track analysis 150 Walking track analysis was performed to evaluate motor function. Rats were placed on a 151 treadmill, and their footprints were scanned using the DigiGait System (Rat Specifics, MA, USA). 152   9 The sciatic functional index (SFI) was calculated according to a previously reported formula (Bain et 153 al., 1989). The zero value of the SFI denotes normal nerve function, and a value of −100 represents 154 total loss of nerve function. 155  156 Histological examination 157 Three out of ten sciatic nerves from each group were examined histologically. To assess the 158 histological changes in the axons and myelin sheaths, specimens from the sciatic nerves 3 mm distal 159 to the repair site in each group were obtained for electron and light microscopy (Kimura et al., 2018; 160 Shibata et al., 2015). Semi-thin sections of 1 μm thickness were stained with 0.1% toluidine blue for 161 7 min and imaged using a BZ-X700 optical microscope (Keyence, Osaka, Japan). Ultrathin axial 162 sections (70 nm thickness) of the sciatic nerve were prepared using an ultramicrotome (Leica UC7, 163 Leica Microsystems GmbH, Wetzlar, Germany) on silicon wafers and stained with uranyl acetate and 164 lead citrate for 10 min. The sections were observed under a SU6600 SEM (Hitachi High-Tech, Tokyo, 165 Japan) by detecting backscattered electrons with an acceleration voltage of 5 kV. For quantitative 166 analysis, axonal density was used to evaluate the regenerated nerve fibres. Axonal density was 167 defined as axonal area/total area, and myelin sheath density was defined as myelin sheath area/total 168 area in a fascicle in each nerve sample (Takagi et al., 2009). For G-ratio quantification, 300 fibres 169 (100 from each nerve) randomly selected from electron microscopy images were used. 170  171 Statistical analysis 172   10 The Shapiro-Wilk test was used to assess the normality of data. Data are presented as mean 173 with standard error. To evaluate intergroup differences, one-way analysis of variance (ANOVA) with 174 Tukey or Games-Howell post hoc comparisons were used for maximum failure load, muscle weight, 175 and axonal density analysis. The Kruskal–Wallis and Dann–Bonferroni tests for post hoc comparisons 176 were used for axon diameter analyses. Two-way repeated ANOVA was used for the comparison 177 between SFI values. Statistical analysis was performed by a hand surgeon who was blinded to clinical 178 information. Post-hoc power analysis was conducted to confirm whether this sample size would be 179 adequate to detect a significant difference, with an alpha of 0.05. Effect sizes are expressed as mean 180 values with standard deviations. The power analysis demonstrated statistical powers of 100%, 97%, 181 98%, and 100% for the cadaveric model, muscle weight, SFI, and G-ratio, respectively. Statistical 182 significance was set at p < 0.05. 183  184  185 RESULTS 186 Failure load using the cadaveric model 187 The maximum failure load for each procedure is shown in figure 2. The maximum failure load 188 of ApGltn sheet was significantly higher than that of a fibrin sealant (0.39 N vs. 0.05 N, p < 0.001). 189 The maximum failure load of single suture + ApGltn sheet was significantly higher than that of a 190 single suture (1.32 N vs. 0.97 N, p = 0.02). There were no significant differences between single 191 suture + fibrin sealant and single suture (0.99 N vs. 0.97 N, p = 0.99). The double suture technique 192   11 (1.65 N) had a higher maximum failure load than the single suture + ApGltn sheet technique, but this 193 was not significant (p = 0.07). The maximum failure load of the ApGltn sheet (0.39 N) and fibrin 194 sealant (0.05 N) was significantly lower than those of the other four procedures (p < 0.001).  195  196 Functional evaluation using the rat model 197 Macroscopic examination 198 All resected sciatic nerves in the double suture, single suture + ApGltn sheet, single suture, and 199 ApGltn sheet groups showed complete nerve continuity without defects. The ApGltn sheets were 200 resorbed and had disappeared by 8 weeks after the initial procedure. Fibrin groups showed complete 201 continuity of the nerves in 9 rats and incomplete continuity in 1 rat. In the resection group, complete 202 rupture was observed in five of ten rats, complete continuity in two, and incomplete continuity in 203 three rats (Figure S3).  204  205 Muscle weight 206  Muscle weight recovery did not significantly differ among the double suture, single suture + 207 ApGltn sheet, single suture, ApGltn sheet, and fibrin techniques (p > 0.05) (Figure 3). However, a 208 significant difference was observed between the resection group and the double suture, single suture 209 + ApGltn sheet, single suture, and ApGltn sheet groups (p < 0.05). 210  211   12 Walking track analysis 212 SFI improved over time; the double suture and single suture + ApGltn sheet groups showed 213 significantly higher SFI values than the ApGltn sheet, single suture, fibrin, and resection groups (p < 214 0.05) (Figure 4). The ApGltn sheet groups showed significantly higher SFI values than the resection 215 group (p < 0.05).  216  217 Histological examination 218 Optical and electron microscopy images of the toluidine blue-stained nerves fibre are shown in 219 figure 5. In axial sections, a larger number of myelinated axons were observed in the double suture, 220 single suture + ApGltn sheet, single suture, ApGltn sheet, and fibrin groups than in the resection 221 group. Quantitative analysis of the axonal density indicated that the double suture and single suture 222 + ApGltn sheet groups had significantly higher values than the fibrin group (Figure 6). No significant 223 differences were observed in axonal density among the double suture, single suture + ApGltn sheet, 224 ApGltn sheet, and single suture groups.  225 The G-ratios are shown in figure S4. Although the double suture group showed thicker myelin 226 regeneration than the single suture + ApGltn sheet group (Figure S4a), the axonal diameters did not 227 significantly differ between the two groups (Figure S5). Similar axonal regrowth patterns were 228 observed in the single suture + ApGltn sheet and single suture groups (Figure S4b) and the single 229 suture and ApGltn sheet groups (Figure S4c); the axonal diameters among the three groups did not 230 significantly differ (Figure S5). The ApGltn sheet group showed thicker myelin regeneration than the 231   13 fibrin group (Figure S4d). The axonal diameters indicated that the double suture, single suture + 232 ApGltn sheet, single suture, and ApGltn sheet groups had significantly higher values than the fibrin 233 group. 234  235  236 DISCUSSION 237 In this study, the bonding strength and functional recovery of transected nerves using ApGltn 238 sheets were compared with traditional sutures and fibrin sealant in cadaveric and rat models. The 239 maximum failure load of the ApGltn sheet was approximately eight times higher than that of the 240 fibrin sealant. Although the bonding strength of the ApGltn sheet was inferior to that of the traditional 241 suture, the addition of the ApGltn sheet to the single suture significantly reinforced the repair strength. 242 Addition of fibrin sealant to the single suture did not significantly reinforce the strength. The ApGltn 243 sheet did not compromise sciatic nerve regeneration compared with the traditional suture. 244 Liquid ApGltn sealant has a stronger breaking strength than fibrin sealant (Masuda et al., 2021; 245 Mizuno et al., 2017; Taguchi et al., 2016; Yamaoka et al., 2019). Taguchi et al. (2016) demonstrated 246 that the burst strength of the ApGltn sealant was 11.6 times higher than that of a fibrin sealant in a 247 burst porcine aorta model (341 vs. 29 mm Hg). Additionally, Masuda et al. (2021) showed that the 248 maximum failure load of liquid-type ApGltn sealant was approximately three times higher than that 249 of the fibrin sealant using a cadaveric digital nerve model (0.22 versus 0.06 N). Although a direct 250 comparison between that study and ours is difficult due to the different digital nerve sizes from 251   14 different specimens employed, the breaking strength of the ApGltn sheet (0.39 N) is higher than that 252 of the liquid-type ApGltn sealant. However, the failure load of the ApGltn sheet was inferior to that 253 of the traditional suture, as also shown in a previous report on liquid ApGltn sealant (Masuda et al., 254 2021).  255 To minimize the effect of size differences, a previous cadaveric model similar to our study 256 compared bilateral digital nerves from the same specimen (Masuda et al., 2021). Addition of liquid 257 ApGltn sealant to the single suture did not significantly increase the maximum failure load (mean 0.1 258 N) compared to the single suture (Masuda et al., 2021); therefore, they concluded that liquid ApGltn 259 sealant with the reported bonding strength cannot be used clinically, and further improvement of the 260 strength in the ApGltn sealant is required. Our study showed that the addition of an ApGltn sheet to 261 a single suture does significantly reinforce the bonding strength (mean 0.35 N). Fibrin glue is 262 sometimes added to the nerve suture site (Isaacs, 2010); however, our results are consistent with the 263 previous findings that the addition of fibrin glues does not significantly increase the maximum failure 264 load (Isaacs et al., 2008). The ApGltn sealant, compared to the fibrin sealant, has been shown to have 265 prolonged adhesive capacity and reduced viral infection risk and cost (Nishimura et al., 2008; Taguchi 266 et al., 2016). Since ApGltn is derived from the waste product of Alaska pollack skin, the cost of 267 materials could be quite low. Therefore, ApGltn sheets could be clinically useful in the future as an 268 alternative to fibrin sealants.  269 In the present study, sciatic nerves repaired with sheet-type ApGltn sealant showed nerve 270 recovery similar to that in the suture group. This result is consistent with that of a previous study 271   15 using liquid ApGltn sealant, showing similar nerve recovery compared to suture and fibrin sealant 272 repair in a rat sciatic nerve model (Masuda et al., 2021). Furthermore, sheet-type ApGltn sealants 273 completely degrade within 21 days without severe inflammation when subcutaneously implanted in 274 the backs of rats (Ichimaru et al., 2021). In an in vitro study, ApGltn sheet had excellent 275 cytocompatibility and efficiently supported the growth of L929 cells (Ichimaru and Taguchi, 2021). 276 Moreover, since it is composed of gelatin, ApGltn sheets do not prevent tissue regeneration.  277 In addition to the bonding strength, the ApGltn sheet has other advantages over liquid ApGltn 278 sealant. Liquid-type adhesives need to mix two components homogeneously; therefore, they require 279 special apparatuses to rapidly mix the components during surgery (Mizuno et al., 2017). Sheet-type 280 adhesives have been developed for their easy sealing property without any special apparatus. 281 Additionally, the liquid type requires refrigeration before use, whereas the sheet type can be stored at 282 room temperature. Furthermore, liquid-type adhesives are difficult to apply evenly around the entire 283 circumference of the suture site in clinical practice, whereas the sheet-type are more evenly applied. 284 Altogether, the ApGltn sheet is clinically easier to use for nerve repair than the liquid type.  285 Our study had some limitations. The maximum failure load of a repaired nerve could differ 286 according to the digital nerve size. To minimize the effect of size differences, the same cadavers were 287 used for the single suture + ApGltn sheet (b) and single suture procedures (d); the double suture (a) 288 and single suture + fibrin sealant (c) procedures; and ApGltn sheet (e) and fibrin sealant (f) procedures. 289 However, the different cadavers used for the procedures might have affected the results. Nerve 290 regeneration was studied in an animal model only. Therefore, clinical trials are needed to validate the 291   16 application of this sealant in humans. 292 In conclusion, the maximum failure load of the ApGltn sheet was higher than that of the fibrin 293 sealant. Addition of the ApGltn sheet to a single suture significantly reinforces the repair strength of 294 the nerve. Its effect on transected nerve regeneration is similar to that of the traditional suture. 295 Although ApGltn sheet cannot serve as a substitute for sutures, we believe that addition of ApGltn 296 sheets to the nerve suture site will be useful for clinical application in the future as an alternative to 297 fibrin sealant.  298  299  300  301 REFERENCES 302 Bain JR, Mackinnon SE, Hunter DA. Functional evaluation of complete sciatic, peroneal, and 303 posterior tibial nerve lesions in the rat. Plast Reconstr Surg. 1989, 83: 129-38. 304 Bamba R, Riley DC, Kim JS et al. Evaluation of a nerve fusion technique with polyethylene glycol 305 in a delayed setting after nerve injury. J Hand Surg Am. 2018, 43: 82.e1-.e7. 306 Barton M, Morley JW, Stoodley MA et al. Laser-activated adhesive films for sutureless median nerve 307 anastomosis. J Biophotonics. 2013, 6: 938-49. 308 Barton MJ, Morley JW, Stoodley MA, Lauto A, Mahns DA. Nerve repair: Toward a sutureless 309 approach. Neurosurg Rev. 2014, 37: 585-95. 310 Childe JR, Regal S, Schimoler P, Kharlamov A, Miller MC, Tang P. Fibrin glue increases the tensile 311 strength of conduit-assisted primary digital nerve repair. Hand (N Y). 2018, 13: 45-9. 312   17 Felix SP, Pereira Lopes FR, Marques SA, Martinez AM. Comparison between suture and fibrin glue 313 on repair by direct coaptation or tubulization of injured mouse sciatic nerve. Microsurgery. 2013, 33: 314 468-77. 315 Ichimaru H, Mizuno Y, Chen X, Nishiguchi A, Taguchi T. Prevention of pulmonary air leaks using a 316 biodegradable tissue-adhesive fibre sheet based on alaska pollock gelatin modified with decanyl 317 groups. Biomater Sci. 2021, 9: 861-73. 318 Ichimaru H, Taguchi T. Improved tissue adhesion property of a hydrophobically modified alaska 319 pollock derived gelatin sheet by uv treatment. Int J Biol Macromol. 2021, 172: 580-8. 320 Isaacs J. Treatment of acute peripheral nerve injuries: Current concepts. J Hand Surg Am. 2010, 35: 321 491-7. 322 Isaacs JE, McDaniel CO, Owen JR, Wayne JS. Comparative analysis of biomechanical performance 323 of available "nerve glues". J Hand Surg Am. 2008, 33: 893-9. 324 Kimura H, Ouchi T, Shibata S et al. Stem cells purified from human induced pluripotent stem cell-325 derived neural crest-like cells promote peripheral nerve regeneration. Sci Rep. 2018, 8: 10071. 326 Masuda S, Suzuki T, Shibata S et al. A novel Alaska pollock gelatin sealant shows higher bonding 327 strength and nerve regeneration comparable to that of fibrin sealant in a cadaveric model and a rat 328 model. Plast Reconstr Surg. 2021, 148: 742e-52e. 329 Mizuno Y, Mizuta R, Hashizume M, Taguchi T. Enhanced sealing strength of a hydrophobically-330 modified alaska pollock gelatin-based sealant. Biomater Sci. 2017, 5: 982-9. 331 Mizuno Y, Taguchi T. Promotion of cell migration into a hydrophobically modified alaska pollock 332   18 gelatin-based hydrogel. Macromol Biosci. 2019, 19: e1900083. 333 Mizuta R, Taguchi T. Enhanced sealing by hydrophobic modification of alaska pollock-derived 334 gelatin-based surgical sealants for the treatment of pulmonary air leaks. Macromol Biosci. 2017, 17. 335 Nishimura MT, Mazzer N, Barbieri CH, Moro CA. Mechanical resistance of peripheral nerve repair 336 with biological glue and with conventional suture at different postoperative times. J Reconstr 337 Microsurg. 2008, 24: 327-32. 338 Rafijah G, Bowen AJ, Dolores C, Vitali R, Mozaffar T, Gupta R. The effects of adjuvant fibrin sealant 339 on the surgical repair of segmental nerve defects in an animal model. J Hand Surg Am. 2013, 38: 847-340 55. 341 Riley DC, Bittner GD, Mikesh M et al. Polyethylene glycol-fused allografts produce rapid behavioral 342 recovery after ablation of sciatic nerve segments. J Neurosci Res. 2015, 93: 572-83. 343 Sameem M, Wood TJ, Bain JR. A systematic review on the use of fibrin glue for peripheral nerve 344 repair. Plast Reconstr Surg. 2011, 127: 2381-90. 345 Shibata S, Murota Y, Nishimoto Y et al. Immuno-electron microscopy and electron microscopic in 346 situ hybridization for visualizing pirna biogenesis bodies in drosophila ovaries. Methods Mol Biol. 347 2015, 1328: 163-78. 348 Taguchi T, Ryo M, Ito T, Yoshizawa K, Kajiyama M. Robust sealing of blood vessels with cholesteryl 349 group-modified, alaska pollock-derived gelatin-based biodegradable sealant under wet conditions. J 350 Biomed Nanotechnol. 2016, 12: 128-34. 351 Takagi T, Nakamura M, Yamada M et al. Visualization of peripheral nerve degeneration and 352   19 regeneration: Monitoring with diffusion tensor tractography. Neuroimage. 2009, 44: 884-92. 353 Temple CL, Ross DC, Dunning CE, Johnson JA. Resistance to disruption and gapping of peripheral 354 nerve repairs: An in vitro biomechanical assessment of techniques. J Reconstr Microsurg. 2004, 20: 355 645-50. 356 Tse R, Ko JH. Nerve glue for upper extremity reconstruction. Hand Clin. 2012, 28: 529-40. 357 Turner NJ, Johnson SA, Foster LJR, Badylak SF. Sutureless nerve repair with ecm bioscaffolds and 358 laser-activated chitosan adhesive. J Biomed Mater Res B Appl Biomater. 2018, 106: 1698-711. 359 Yamaoka M, Maki N, Wijesinghe A et al. Novel alaska pollock gelatin sealant shows high adhesive 360 quality and conformability. Ann Thorac Surg. 2019, 107: 1656-62. 361  362  363  364  365 FIGURE LEGENDS 366 Figure 1. Resected digital nerves repaired with single suture + ApGltn sheet. An approximately 10 × 367 15 mm sheet was applied around the nerve rupture site. 368  369 Figure 2. The maximum failure load for each procedure. The failure load of the single suture + 370 ApGltn sheet was significantly higher than that of the single suture and that of ApGltn sheet was 371 significantly higher than that of a fibrin sealant. Each box represents the interquartile range of values, 372   20 with the bold line showing the median value. The vertical lines show maximum and minimum values. 373 *p < 0.05, **p < 0.001 374  375 Figure 3. Muscle weight recovery of the tibialis anterior in the experimental limbs compared with 376 that in the control limbs. No significant differences are observed among the procedures employing 377 double sutures, single sutures + ApGltn sheets, single sutures, ApGltn sheets, and fibrin techniques. 378 Each box represents the interquartile range of values, with the bold line showing the median value. 379 The vertical lines show maximum and minimum values. *p < 0.05, **p < 0.001 380  381 Figure 4. Sciatic functional index in each procedure until the completion of 8 postoperative weeks 382 at 4, 6, and 8 weeks, the double suture and single suture + ApGltn sheet groups show significantly 383 higher values than the ApGltn sheet, single suture, fibrin, and resection groups. Data are presented as 384 mean with standard deviation. 385  386 Figure 5. Axial sections of the sciatic nerve: light microscopy with toluidine blue (above) and 387 electron microscopy with uranyl acetate (below) staining. A larger number of myelinated axons are 388 observed in the double suture, single suture + ApGltn sheet, single suture, ApGltn sheet, fibrin groups 389 than in the negative control resection group. 390  391   21 Figure 6. Quantitative analysis of axonal density of the regenerated nerve fibrefibres. The double 392 suture and single suture + ApGltn sheet groups had significantly higher values than the fibrin sealant. 393 No significant differences were observed among the double suture, single suture + ApGltn sheet, 394 ApGltn sheet, and single suture. Each box represents the interquartile range of values, with the bold 395 line showing the median value. The vertical lines show maximum and minimum values. *p < 0.05, 396 **p < 0.001 397  398 Figure S1. Biomechanical traction testing using transected digital nerves in a cadaveric model and 399 functional testing using transected sciatic nerves in a rat model. 400  401 Figure S2. Photograph of prepared ApGltn sheet (left) and scanning electron microscopy images of 402 ApGltn sheet (right). 403  404 Figure S3. Macroscopic examination of the resected sciatic nerves. All resected sciatic nerves in the 405 double suture, single suture + ApGltn sheet, single suture, ApGltn sheet, and fibrin groups show nerve 406 continuity, while segmental defects are observed in five out of ten rats in the resection group. 407  408 Figure S4. The G-ratio and axon diameter results. The approximate linear regression equation for 409 each group is shown in the graph.  410   22 (a) Double suture and single suture + ApGltn sheet  411 (b) Single suture + ApGltn sheet and single suture  412 (c) Single suture and ApGltn sheet  413 (d) ApGltn sheet and fibrin sealant 414  415 Figure S5. Axonal diameter of the regenerated nerve fibres. The axonal diameters indicated that the 416 double suture, single suture + ApGltn sheet, single suture, and ApGltn sheet groups had significantly 417 higher values than the fibrin group. Each box represents the interquartile range of values, with the 418 bold line showing the median value. The vertical lines show maximum and minimum values. *p < 419 0.05, **p < 0.001 420  421 Figure1 Click here to access/download;Figure(s);Figure1.jpghttps://www2.cloud.editorialmanager.com/jhse/download.aspx?id=780612&guid=2489e73e-7e45-495c-9654-ab9c46dcaf91&scheme=1https://www2.cloud.editorialmanager.com/jhse/download.aspx?id=780612&guid=2489e73e-7e45-495c-9654-ab9c46dcaf91&scheme=1Figure2 Click here to access/download;Figure(s);Figure2.jpghttps://www2.cloud.editorialmanager.com/jhse/download.aspx?id=780613&guid=6ab39a1d-9cde-4c2a-b6bf-d11361b1e0bd&scheme=1https://www2.cloud.editorialmanager.com/jhse/download.aspx?id=780613&guid=6ab39a1d-9cde-4c2a-b6bf-d11361b1e0bd&scheme=1Figure3 Click here to access/download;Figure(s);Figure3.jpghttps://www2.cloud.editorialmanager.com/jhse/download.aspx?id=780614&guid=46b0224e-44c5-47ea-b329-eba65d928ef5&scheme=1https://www2.cloud.editorialmanager.com/jhse/download.aspx?id=780614&guid=46b0224e-44c5-47ea-b329-eba65d928ef5&scheme=1Figure4 Click here to access/download;Figure(s);Figure4.jpghttps://www2.cloud.editorialmanager.com/jhse/download.aspx?id=780615&guid=01fc72e1-fad5-4d03-8ea6-66036a316233&scheme=1https://www2.cloud.editorialmanager.com/jhse/download.aspx?id=780615&guid=01fc72e1-fad5-4d03-8ea6-66036a316233&scheme=1Figure5 Click here to access/download;Figure(s);Figure5.jpghttps://www2.cloud.editorialmanager.com/jhse/download.aspx?id=780616&guid=55a5ef37-0fa0-46a7-9039-fd8f8641035a&scheme=1https://www2.cloud.editorialmanager.com/jhse/download.aspx?id=780616&guid=55a5ef37-0fa0-46a7-9039-fd8f8641035a&scheme=1Figure6 Click here to access/download;Figure(s);Figure6.jpghttps://www2.cloud.editorialmanager.com/jhse/download.aspx?id=780617&guid=f2d10258-d8b1-488e-a89f-23b4af7c613e&scheme=1https://www2.cloud.editorialmanager.com/jhse/download.aspx?id=780617&guid=f2d10258-d8b1-488e-a89f-23b4af7c613e&scheme=1