Investigation of anti-adhesion ability of 8-arm PEGNHS-modified porcine pericardium

In post-adhesion surgery, there is a clinical need for anti-adhesion membranes specifically designed for the liver, given the limited efficacy of current commercial products. To address this demand, we present a membrane suitable for liver surgery applications, fabricated through the modification of decellularized porcine pericardium with 20 KDa hexaglycerol octa (succinimidyloxyglutaryl) polyoxyethylene (8-arm PEGNHS). We also developed an optimized modification procedure to produce a high-performance anti-adhesion barrier. The modified membrane significantly inhibited fibroblast cell adherence while maintaining minimal levels of inflammation. By optimizing the modification ratio, we successfully controlled post-adhesion formation. Notably, the 8-arm PEG-modified pericardium with a molar ratio of 5 exhibited the ability to effectively prevent post-adhesion formation on the liver compared to both the control and Seprafilm®, with a low adhesion score of 0.5 out of 3.0. Histological analysis further confirmed its potential for easy separation. Furthermore, the membrane demonstrated regenerative capabilities, as evidenced by the proliferation of mesothelial cells on its surface, endowing anti-adhesion properties between the abdominal wall and liver. These findings highlight the membrane’s potential as a reliable barrier for repeated liver resection procedures that require the removal of the membrane multiple times.


List of abbreviations
no wound on the liver SP100 wound on the liver GA glutaraldehyde

Introduction
Post-surgery adhesion in the liver poses a significant concern for patients undergoing various procedures such as partial liver resections [1,2], liver transplantation [3,4], abdominal surgery for gonorrheal disease [5], gastrectomy [6], and splenectomy [7].This issue inflicts suffering on patients and presents complications for surgeons during reoperation procedures, contributing to increased blood loss and extended operation times [8,9].Post-operative adhesions result from an abnormal healing process following surgical tissue trauma, characterized by excessive fibrin formation that leads to the development of connective tissue between the injured site and the surrounding area [10,11].
The current solution for this issue is the application of a physical barrier to isolate the injured area from its surroundings, facilitating normal healing processes.A range of barrier options are commercially available, including solid types such as Coseal™ and Gore-Tex®, gel forms like SprayGel®, Intergel®, and Oxiplex®, as well as liquid options such as Adept® and Sepracoat®.These barriers have been widely utilized in various anatomical sites, include the pericardium, peritoneum, and tendons [11][12][13].For liver surgery, Seprafilm®, AdSpray®, and Intreceed® have been utilized, with reported effectiveness and safety.Seprafilm® has demonstrated efficacy in alleviating technical challenges associated with repeat liver sections, reducing adhesion formation to a score of 2 out of 3 [14][15][16][17][18].In the case of AdSpray®, the incidence was not significantly different from the control group, and a high risk of intra-abdominal abscesses was observed [19,20].With Intreceed®, severe adhesions in the liver persisted [17,21,22].These findings highlight the need for effective anti-adhesion materials for use in liver surgery that meet the relevant clinical requirements.
The use of decellularized pericardium as a base material for biomedical applications has gained significant interest due to its shape, availability, and mechanical properties.Decellularized pericardium, especially when prepared using high hydrostatic pressure (HHP), has been reported to preserve the extracellular matrix structure.Decellularized tissue exhibits favorable biocompatibility and induces a minimal immune response [23][24][25][26].In recent research, we introduced the use of the pericardium as the primary material for anti-adhesion applications in heart surgery, achieved by transforming its surface properties from adhesive to anti-adhesive [27][28][29].The application of polyethylene glycol (PEG) for the surface modification of the pericardium has emerged as a promising technique.PEG is characterized by its flexibility, hydrophilicity, biocompatibility, non-toxicity, and the ability to repel proteins and cells [30][31][32][33][34][35][36][37][38][39][40][41].In addition, it has been reported that 8-arm PEG can effectively decrease fibroblast cell adherence [42].
Thus, we hypothesized that an 8-arm PEG-modified pericardium would serve as an effective barrier to prevent adhesion formation between the wound site and its surroundings, thereby facilitating normal wound healing (figure 1).However, the optimal protocol to ensure high efficacy in preventing post-adhesion formation in vivo has not yet been established.
The aim of the current study was to define the optimal conditions for modifying dPPC with 8-arm PEGNHS and to investigate its efficacy in preventing post-adhesion formation following rat liver surgery.Various in vitro parameters, including blood coagulation on the pericardium surface, inflammatory response, and fibroblast cell adherence were investigated.The prevention of post-adhesion formation was assessed using adhesion scores and histological analysis.

Preparation of 8-arm PEG-modified dPPC
The native PPC was sourced from a local slaughterhouse (Tokyo Shibaura Zouki, Tokyo, Japan) and subsequently decellularized using the HHP method.The pericardium was initially pressurized using a cold isostatic pressure machine (Dr CHEF; Kobelco, Hyogo, Japan) at 1000 MPa and 30 • C for 10 min.Then it was sequentially washed while shaking, first with a DNase I solution (0.2 mg ml −1 deoxyribonuclease I (Sigma-Aldrich, St. Louis, MO, USA) with 50 mM magnesium chloride (Ph Eur grade, Merck Tokyo, Japan)) for 7 d, followed by 80% ethanol for 3 d, and finally with 0.1 M citrate buffer for 3 d at 4 • C. The dPPC was modified with 1% 8-arm PEGNHS with a molecular weight of 21 286 Da and a 97% NHS substitution rate (NOF, Tokyo, Japan) in phosphatebuffered saline (PBS, FUJIFILM Wako Pure Chemical Corp.Osaka, Japan) at 4 • C overnight.The modification molar ratio between the NH 2 groups of the dPPC and NHS groups of the 8-arm PEGNHS were 1:2,1:5, and 1:10.Samples were denoted as 8P2, 8P5, and 8P10 according to the modification ratios, where 8 stands for 8-arm PEG, P for dPPC, and 2, 5, or 10 for the NHS modification ratio.

Deoxyribonucleic acid (DNA) content
Native and decellularized freeze-dried PPC (10-20 mg) were digested with lysis buffer (20 mg ml −1 proteinase K, 50 mM Tris-HCl, 1% SDC, 10 mM NaCl, and 20 mM EDTA) at 55 • C overnight.DNA was extracted using phenol/chloroform and precipitated with ethanol and sodium chloride.The DNA content was quantified using the Quanti-iT TM PicoGreen™ dsDNA assay kit (Thermo Fisher Scientific, Tokyo, Japan).Fluorescence intensity was measured at 480 nm excitation and 521 nm emission with a multi-mode microplate reader (Cytation 5; BioTek, Winooski, VT, USA), with results analyzed against a λDNA standard curve.

Water contact angle
The static water contact angle of fresh dPPC, 8P2, 8P5, and 8P10 were analyzed using the air bubble method using a DMo-602 automatic contact angle meter (Kyowa Interface Science Co., Ltd).The sample was held by a ternary system kit, and an air bubble was produced by an inverted needle with a volume of 3 ml.Images and data were captured and analyzed using the interFAce Measurement & Analysis System-Famous program version 7.2.0.

Attenuated total reflection Fourier transform infrared spectroscopy (ATR-FTIR)
ATR-FTIR analysis (Spectrum 100, PerkinElmer Japan, Kanagawa, Japan) was conducted to analyze the chemical bonds between the 8-arm PEGNHS and pericardium.Spectra were recorded between 4000-500 cm −1 with 64 scans and a resolution of 1 cm −1 .Data analysis of three independent replicates was performed using the Spectrum 100 software.

Enzyme degradation
Fresh 2 × 2 cm samples of dPPC, 8P2, 8P5, and 8P10 were immersed in a solution of 0.1% collagenase (5 ml, 250 units mg −1 , 034-22363, FUJIFILM Wako Pure Chemical Corp.Osaka, Japan) supplemented with 5 × 10 −3 M calcium chloride and incubated at 37 • C with constant shaking for 2, 4, 8, and 24 h.As a control, dPPC was immersed in 0.6% GA in PBS at 4 • C overnight.The initial weight of each sample was measured before incubation (W o ), and the remaining weight (W t ) was measured after collagenase treatment.The percentage of remaining weight was calculated using the following equation:

Thrombogenicity
Porcine blood containing 3.24% citric acid was purchased from a local slaughterhouse (Tokyo Shibaura Zouki, Tokyo, Japan), and was used to analyze blood coagulation on samples including dPPC, 8P2, 8P5, and 8P10, as well as controls including cover glass (C218181, Matsunami Glass Ind., Ltd Osaka, Japan) and PTFE (Naflon® PTFE sheet, 7-359-01, Nichias Co., Ltd).A 50 µl droplet of a 1:10 mixture of calcium chloride and blood was applied to the samples and control, then incubated at 37 • C for 2, 4, 6, 8, 10, and 15 min, followed by washing with physiological saline 10 times and subsequent imaging.The imaged samples and 50 µl of blood with calcium chloride were immersed in 2 ml of deionized water for hemolysis at RT overnight.The absorbance of the hemolysis solution was measured at 576 nm using a multi-mode microplate reader (Cytation 5; BioTek, Winooski, VT, USA).The percentage of blood coagulation was calculated using the following formula [43]: where A sample : absorbance of the sample, A deionized water : absorbance of the deionized water, A blood : absorbance of the blood with calcium chloride.

In vivo evaluation 2.10.1. Animals
The experimental protocol was approved by the ethics committee for animal care and use at Tokyo Medical and Dental University (A2023-176A).The 7-weekold male Slc:SD rats (210-230 g weight) were purchased from Japan SLC (Shizuoka, Japan).The rats were stabilized for one week prior to commencing the experimentation.
A 5 cm midline incision was made on the abdomen to expose the liver, which was randomly assigned to one of two groups: no wound and with wound.In the no wound group, 8-arm PEGmodified pericardium at different molar ratios (2 (8P2-SP0, n = 3), 5 (8P5-SP0, n = 3), and 10 (8P10-SP0, n = 3)) were separately implanted onto the liver.

Adhesion analysis
After a 14 d recovery period following surgery, the rats were euthanized using an overdose of sodium pentobarbital.The abdomen was opened in a U-shape fashion.The rats were then divided into two groups for evaluation: adhesion score and histological analysis.Adhesion scoring was conducted by manually separating the implanted membrane from the liver and assigning scores based on the Heydorn et al scoring system [44] as follows: score 0 indicated no adhesion, score 1 indicated light adhesion which was easy to peel off without bleeding, score 2 indicated strong adhesion with minimal or no bleeding upon separation, and score 3 indicated dense adhesion which was difficult to separate by peeling and resulted in bleeding (figure 2).
For histological analysis, liver samples with implanted membranes intact and liver samples with surrounding abdominal walls were extracted from rats.The samples were immersed in 10% formalin neutral buffer solution for subsequent analysis.

Hematoxylin-eosin (H&E) staining
The samples were fixed in 10% formalin neutral buffer solution at RT for 24 h.Subsequently, a sequence of ethanol solutions (70%, 80% 90%, and 100%) followed by xylene were used to dehydrate the samples, after which they were embedded in paraffin.Sections of 4 µm thickness were obtained from the paraffin-embedded samples, then deparaffinized and stained with an H&E solution.Cellular structures were observed using a fluorescence microscope (BZ-X700; Keyence, Osaka, Japan).

Masson's trichrome staining
Paraffin samples (4 µm thickness) were first deparaffinized, followed by staining using a series of Trichome Stain Kits (Modified Masson's) supplied by Scytek Laboratories, Inc., U.S.A.The staining process involved immersion in Bouin's fluid at 60 • C for 60 min, Weigert's iron hematoxylin for 10 min, Biebrich scarlet/acid fuchsin solution for 10 min, phosphomolybdic/phosphotungstic acid solution for 15 min, aniline blue solution for 10 min, and 1% acetic acid solution for 5 min.Subsequently the samples underwent dehydration using a series of alcohol solutions ranging from 70% to 100%, followed by xylene, and finally mounted in synthetic resin.Cellular structures were observed using a fluorescence microscope (BZ-X700; Keyence, Osaka, Japan).

Immunohistochemical analysis
Mesothelial cells were evaluated using an antipan cytokeratin antibody (AE1/AE3, ab961, Abcam, Cambridge U.K). Deparaffinized 4 µm samples were treated with proteinase K (S3020, Agilent Technologies, Tokyo, Japan) at 37 • C for 10 min for antigen retrieval.Subsequently, the section was permeabilized with 0.5% Triton X-100 in PBS for 10 min at RT, then blocked with 3% hydrogen peroxide in methanol for 10 min at RT, followed by 3% skimmed milk in TPBS for 60 min at RT.The sections were then incubated with the primary antibody (AE1/AE3) at 4 • C overnight, followed by incubation with a secondary antibody (Histofine simple stain MAX-PO (M), HRP, Nichirei, Tokyo, Japan) for 30 min at RT. Subsequently, the samples were stained using Histofine simple stain DAB solution (415 171, Nichirei Biosciences Inc., Tokyo, Japan) for 20 min, counter-stained with Mayer's hematoxylin for 1 min, and observed using a fluorescence microscope (BZ-X700; Keyence, Osaka, Japan).

Statistical analysis
The data are presented as mean ± standard deviation.Multiple treatment comparisons were performed using a one-way ANOVA followed by a post-hoc Tukey's honestly significant difference test.Statistical significance was defined as a p-value below 0.05.Significance levels were indicated as * p < 0.05 and * * p < 0.01.

DNA content
The DNA concentration in the native PPC was 1446.938ng mg −1 , which decreased markedly to 6.306 ng mg −1 in the dPPC following the HHP method (figure 3(a)).This reduction adheres to the recommended DNA residue threshold of below 50 ng mg −1 for dry tissue [45].

Water contact angle
The inverted bubble method is suitable for measuring the water contact angle of hydrophilic materials such as dPCC.In figure 3(b), it can be observed that the air bubble formed was round and adhered to dPPC, 8P2, 8P5, and 8P10.The average contact angle of dPPC was 137 • , which increased to 142 • , 143 • , and 144 • for 8P2, 8P5, and 8P10, respectively (figure 3(c)).These findings suggest that the surface of the pericardium became more hydrophilic following PEG modification.There was no significant difference among the contact angle values for 8P2, 8P5, and 8P10.

Enzymatic degradation
The percentage reduction in sample weight following incubation with collagenase is presented in figure 3(d).The weight of the GA-modified pericardium samples remained at approximately 85% after 24 h, indicating limited degradation by the enzyme.The weight of the dPPC samples exhibited a notable decrease after 2 h of incubation and complete degradation by 8 h.In contrast, the weight of all the 8-arm PEG-modified samples decreased slowly compared to dPPC.After 8 h, their weight remained at more than 50%, decreasing to 40% after 24 h.These findings suggest that the modification of the pericardium with 8-arm PEG can effectively prolong degradation time by the enzyme.

Anti-PEG antibody staining
Images of the anti-PEG antibody staining of cryosections demonstrate that post-modification, PEG effectively covered the entire surface of the pericardium (figure 4).The presence of PEG on the pericardium remained evident throughout the entire 14 d incubation period.

Anti-adhesion effect in vitro
Calcein-AM staining revealed that fibroblast cells adhered to TCPS and dPPC from day 1, with complete coverage observed by day 7 (figure 5(a)).Conversely, the cell morphology on 8P2 and 8P5 remained round until day 7, indicating poor cell adherence.On 8P10, the cells remained round with no proliferation until day 3, after which some cells exhibited signs of proliferation by flattening on day 7. Cell density, as assessed using the WST-8 cell counting kit, is presented in figure 5(b).The number of cells adhered to the modified samples was significantly lower compared to TCPS and dPPC on days 1, 3, and 7.These findings demonstrate the efficacy of 8-arm PEG in preventing fibroblast cell adherence on the pericardium.

Thrombogenicity
Images of blood coagulation and the corresponding blood coagulation rates on the surfaces of the controls (glass and PTFE), dPPC, 8P2, 8P5, and 8P10 are presented in figures 6(a) and (b), respectively.Blood started to coagulate on the glass in 6 min, PTFE in 10 min, dPPC in 8 min, 8P2 in 8 min, 8P5 in 10 min, and 8P10 in 8 min (figure 6(a)).Figure 6(b) highlights the blood coagulation tendency of the dPPC.Notably, with the introduction of the 8-arm PEG at a  molar ratio of 5, the blood coagulation properties of the pericardium improved, exhibiting behavior similar to that of PTFE.However, 8P2 and 8P10 demonstrated limitations in their ability to prevent blood coagulation.

Inflammation evaluation
The density of HiBiT-tagged-THP-1 cells on dPPC and 8-arm PEG-dPPC is presented in figure 7(a).The cell density ranged between 3.0 and 3.5 × 4 cells per well in TCPS (LPS+), TCPS (LPS−), and dPPC, which was significantly reduced to 1.0-1.7 × 10 4 cells per well for all 8-arm PEG-dPPC samples.THP-1 cells were stained with calcein-AM and fluorescence images were acquired (figure 7(b)).While THP-1 cells were adhered to the TCPS groups and dPPC, adhesion was reduced on the 8-arm PEG-modified samples.Cell density measurements and image analysis were consistent with each other (figures 7(a) and (b), respectively).

Post-adhesion surgery assessment
All animals survived until the scarification time (14 d post-surgery) without any signs of infection observed macroscopically.
The image presented in figure 8(a) illustrates a representative example of the separation between the implanted samples and the liver, serving as a basis for assigning adhesion scores.The degree of adhesion varied among groups.In the adhesion formation model (SP100), severe adhesion between the abdominal wall and the liver was observed, with blood release upon attempted separation, indicating a severity level of 3. In the dPPC group (dPPC-SP100), strong adhesion to the liver wound was noted, also resulting in blood release during separation, indicating a score of 3. The commercial product Seprafilm® exhibited film residue upon separation, again leading to blood release and a score of 3.
In the absence of a wound (SP0), 8P2 and 8P5 could be separated from the liver without causing damage, receiving a score of 0. However, 8P10 was not separable and resulted in blood release, warranting a score of 3.
In the presence of a wound (SP100), 8P5 could be separated from the liver without damage, receiving a score of 0, whereas 8P2 and 8P10 were not separable and led to blood release, resulting in a score of 3.
The average adhesion scores are presented in figure 8(b).In the SP100 adhesion formation model, adhesion was successfully induced with an average score of 2.5.Seprafilm® exhibited variable results ranging from anti-adhesion effects to adhesion formation, with an average score of 2, highlighting the limitations of this product in liver surgery.Conversely, dPPC resulted in severe adhesion in all cases, with an average score of 3. In the absence of a wound, 8P2-SP0 (average score 1.0) and 8P5-SP0 (average score 0.5) demonstrated potential in reducing adhesion formation, whereas 8P10-SP0 (average score 3.0) exhibited strong adhesion formation.In the presence of a wound, 8P5-SP100 maintained an average score of 0.5, which was not significantly different from 8P5-SP0.However, for 8P2-SP100, the average score increased to 2.5 from 1.0 for 8P2-SP0, indicating increased adhesion formation.Severe adhesion was observed with a score of 3 for 8P10-SP100.Based on these findings, we conclude that 8P5 effectively prevents the formation of adhesions in liver surgery.

Histological analysis of post-adhesion surgery
Adhesion formation and fibrous layer development were observed using H&E staining, while collagen deposition was analyzed using Masson's trichrome staining, visualized in blue (figure 9).Sections were prepared from non-separated samples.In the SP100 group, the abdominal wall was strongly connected to the liver, with plenty of connective tissue and no discernable boundary between the tissues.Regenerative liver and abdominal wall tissues were also observed.Disordered and abundant collagen fibers (blue) indicated the formation of strong and permanent adhesions.
For dPPC-SP100, thick connective tissue with newly generated blood vessels, along with significant collagen deposition, were observed.
In Seprafilm-SP100 samples, no microscopic barrier residue was observed and regeneration of the liver wound was slow.Multinucleated giant cells, slit-like spaces, foam cells, mild collagen deposition with disordered collagen fibers, and new blood vessel formation were also observed.
In the absence of a wound, 8P2-SP0 and 8P5-SP0 exhibited minimal and thin connective tissue, with noticeable spaces between the fibers, indicating ease of separation from the liver.Conversely, in 8P10-SP0 samples, thick connective tissue and disordered collagen fibers were observed, indicating difficulty in separation.
With wounds, both 8P2-SP100 and 8P10-SP100 exhibited strong connective tissue and large blood vessels, indicative of no separability with these samples.8P5-SP100 exhibited similar characteristics to 8P5-SP0, suggesting anti-adhesion prevention in liver surgery.Furthermore, all the 8-arm PEG modified dPPC samples showed cellular infiltration, suggesting regenerative potential.The presence of mesothelial cells was examined by AE1/AE3 immunohistochemical staining (figure 10).In the SP100 group, mesothelial cells were absent between the surface of the abdominal wall and liver due to sandpaper abrasion, resulting in permanent adhesion formation.The Seprafilm-SP100 samples exhibited only a few cells, while none were observed on dPPC-SP100.
Both in the absence and presence of wounds (SP0 and SP100), for the 8P2, 8P5, and 8P10 samples, mesothelial cells were observed and aligned on the surface, indicating that adhesion between the abdominal wall and liver was prevented.This suggests that 8P5 not only facilitated easy separation from the liver but also enabled mesothelial cells to regenerate after 14 d of incubation.

Discussion
Patients who suffer from liver diseases sometimes require multiple surgeries, leading to post-surgery adhesions that are difficult to prevent.Effective anti-adhesion membranes are needed that offer easy separation and size adaptability.In a previous study, we reported the effects of different branched PEGs on dPPC modification.While 8arm PEG-modified pericardium demonstrated significant prevention of fibroblast cell adherence and showed promise as a candidate for future antiadhesion membranes, its efficacy in preventing postadhesion in vivo remains limited.Thus, optimizing the modification ratio to enhance the membrane's anti-adhesion performance in vivo is necessary.This study aimed to explore the impact of three different modification ratios on preventing post-adhesion formation in rat liver surgery.Additionally, in vitro experiments were performed to complement the findings.
Our base material, dPPC, was prepared using the HHP method, which achieved high efficacy in cellular removal, as confirmed by DNA content analysis, indicating low immunogenicity.
After modifying the dPPC with 8-arm PEG, the presence of the PEG molecule throughout the dPPC was confirmed by anti-PEG antibody staining.Even after 14 d of incubation, the PEG molecules remained detectable in the pericardium, highlighting the longevity of the modification.The sustained presence of PEG maintained the hydrophilic surface of the dPPC, as confirmed by water contact angle analysis.Furthermore, the enzymatic degradation properties of the sample were significantly enhanced, indicating that this membrane could serve as an effective barrier during the wound-healing stage without concern for the effects of bodily fluids.
Fibroblast cells did not adhere to the dPPC across all the modification ratios, suggesting that cell signaling was effectively blocked by the presence of the 8-arm PEG molecule [35,38,47].Moreover, the introduction of 8-arm PEG did not induce significant inflammation in the decellularized pericardium, as demonstrated by both the HiBiT assay and histological analysis (8P2-SP0, 8P5-SP0, and 8P10-SP0).
The adhesion formation model can been established through various methods such as mechanical damage, surgical removal, tissue defects, ischemia, and microbe contamination [48].In this study, we chose mechanical damage induced by sandpaper abrasion because of its cost effectiveness and simplicity.We successfully produced an adhesion model between the liver and abdominal wall by subjecting the liver to 100 strokes with sandpaper No. 240, resulting in an average adhesion score of 2.5.To evaluate the efficacy of preventing adhesion formation using this model, we assessed five types of anti-adhesion membranes: Seprafilm®, dPPC, 8P2, 8P5, and 8P10.
Seprafilm® is a bioabsorbable membrane composed of sodium hyaluronate and carboxymethylcellulose.It forms a gel at the surgery site within 24-48 h and degrades within approximately one month.However, it is brittle and sticky, which can complicate surgical procedures.Additionally, it has been associated with potential risks such as hemorrhage and poor wound healing [49,50].Indeed, our own histological analysis revealed signs of delayed wound healing in samples treated with Seprafilm®.These observations align with those of Shimizu et al [51] who reported that while Seprafilm® had an anti-adhesion ability, it varied among animals, as indicated by a wide range of standard deviation.
The dPPC exhibits robust mechanical properties, it is easy to handle, and its size can be adjusted as per the application [42].However, the enhanced biocompatibility of this membrane promotes strong adhesion between the liver, membrane, and abdominal wall.Thus, the modification of dPPC with 8-arm PEG enabled us to control the adhesion and facilitate wound healing.Different modification ratios yielded distinct effects toward achieving this objective.
8P2 exhibited limitations in adhesive prevention and blood coagulation attributed to inadequate PEG coverage on the pericardium.While 8P2-SP0 has a similar anti-adhesion nature to 8P5-SP0, the introduction of a wound leads to moderate adhesion formation.This suggests that the PEG may detach from the pericardium faster in 8P2 compared to 8P5 during the wound healing process, resulting in adhesion formation.
8P10 was unsuitable for anti-adhesion applications due to its high cell adherence and strong adhesion formation once the wound was introduced.In the absence of a wound, severe adhesion between the pericardium and the liver was already evident.This could be attributed to the higher free arm of the PEG presented on the samples, facilitating higher cell attachment and subsequent tissue formation [52][53][54].Upon the introduction of a wound, the woundhealing process begins, leading to the aggregation of various cell types, along with fibrin formation and collagen deposition [10,11].Consequently, this causes greater adhesion; therefore, 8P10 may find utility in alternative biomaterial applications.
8P5 effectively served as a barrier between the liver and abdominal wall.There was no significant difference between 8P5-SP0 and 8P5-SP100, indicating that 8P5 could tolerate the wound healing process (figure 9).It effectively deterred fibroblast cell adherence, allowing the liver wound to heal normally without adhesion formation, and enabling easy separation from the liver [55,56].Throughout the implantation period, the PEG molecules detached and were cleared by bodily fluids, facilitating mesothelial cell adherence and proliferation on the membrane surface [57].This transition endowed the membrane with anti-adhesion properties, preventing adhesion between the membrane and abdominal wall.The natural function of mesothelial cells is creating a non-adhesive and slippery surface [58].Over the incubation period, the 8-arm PEG could detach and be cleared from the 8P5 by bodily fluids.Consequently, 8P5 acquired a function similar to dPPC, demonstrating regenerative capabilities, as evidenced by the infiltration of cells observed in histological sections.These cells likely originated from the bodily fluids that permeated the membrane.As a result, 8P5 is expected to integrate into the body's membrane over the long term once it is implanted.Furthermore, 8P5 demonstrates suitability for the blood environment, as evidenced by the achievement of low blood coagulation on its surface.

Conclusion
In this study, the optimal modification procedure for dPPC using an 8-arm PEG molecule was established, achieving a modification ratio of 5. Through this procedure, we successfully developed a cell-repelling surface while maintaining a low inflammatory response.A post-adhesion model between the liver and abdominal wall was successfully established using mechanical damage with sandpaper.The 8P5 sample emerged as a highly effective membrane for controlling postadhesion formation in liver surgery, with a low adhesion score of 0.5 out of 3.0.The ease of separation of 8P5 from the liver highlights its potential for facilitating repeated liver resections.In the future, it would be valuable to explore the application of this material in larger animal models, as this could pave the way for clinical studies with enhanced translatability.
from the Japan Society for the Promotion of Science, Yoshimi Memorial TMP Grant of the Japanese Society for Artificial Organs, Leading Advanced Projects for Medical Innovation from the Japan Agency for Medical Research and Development (AMED-LEAP) (JP21gm0810008 to A K), and Regulatory Science Projects for Medical Innovation from the Japan Agency for Medical Research and Development (AMED-RS) (JP22mk0101219 to A K).

Figure 2 .
Figure 2. Schematic illustrating the adhesion score system between the liver and pericardium.Score 0 indicates no adhesion; score 1 indicates light adhesion, easy to peel off without bleeding; score 2 indicates strong adhesion with minimal/no bleeding upon separation; score 3 indicates dense adhesion, difficult to separate, bleeding.

Figure 9 .
Figure 9. H&E and Masson's trichrome staining of non-peeling samples after 14 d implantation on rat livers.SP0: no wound, SP100: wound, L: liver, AW: abdominal wall, P: pericardium, SF: Seprafilm®, black dashed box: the site of adhesion formation where the level of adhesion increases proportionally with the thickness of the line, white dashed box: the site of collagen deposition site.2, 5, 10: NHS ratio.Scale bars: 100 µm.