Polyethylene glycol crosslinked decellularized single liver lobe scaffolds with vascular endothelial growth factor promotes angiogenesis in vivo

Jin-S Zhng ,Zhi-Bin Wng ,Zhi-Zhn Li ,Jing-Wn Yng ,Wn-Jing Song ,Yu-Bing Wi ,Jin Mi ,Jin-Gung Wng

a Anatomy Department,School of Basic Medical Sciences,Wenzhou Medical University,Wenzhou 3250 0 0,China

b Institute of Bioscaffold Transplantation and Immunology,Wenzhou Medical University,Wenzhou 3250 0 0,China

c Institute of Hypoxic Medicine,School of Basic Medical Sciences,Wenzhou Medical University,Wenzhou 3250 0 0,China

d Intensive Care Unit,Tongde Hospital of Zhejiang Province,Hangzhou 310012,China

e Department of Geriatric Medicine,The First Affiliated Hospital,Wenzhou Medical University,Wenzhou 3250 0 0,China

f Department of Microbiology and Immunology,Wenzhou Medical University,Wenzhou 3250 0 0,China

g Medical Research Center,Ningbo City First Hospital,Ningbo 3150 0 0,China

h Department of Biochemistry,School of Basic Medical Sciences,Wenzhou Medical University,Wenzhou 3250 0 0,China

Keywords: Decellularization Single liver lobe Polyethylene glycol Angiogenesis Liver tissue engineering

ABSTRACT Background: Improving the mechanical properties and angiogenesis of acellular scaffolds before transplantation is an important challenge facing the development of acellular liver grafts.The present study aimed to evaluate the cytotoxicity and angiogenesis of polyethylene glycol (PEG) crosslinked decellularized single liver lobe scaffolds (DLSs),and establish its suitability as a graft for long-term liver tissue engineering.Methods: Using mercaptoacrylate produced by the Michael addition reaction,DLSs were first modified using N-succinimidyl S-acetylthioacetate (SATA),followed by cross-linking with PEG as well as vascular endothelial growth factor (VEGF).The optimal concentration of agents and time of the individual steps were identified in this procedure through biomechanical testing and morphological analysis.Subsequently,human umbilical vein endothelial cells (HUVECs) were seeded on the PEG crosslinked scaffolds to detect the proliferation and viability of cells.The scaffolds were then transplanted into the subcutaneous tissue of Sprague-Dawley rats to evaluate angiogenesis.In addition,the average number of blood vessels was evaluated in the grafts with or without PEG at days 7,14,and 21 after implantation.Results: The PEG crosslinked DLS maintained their three-dimensional structure and were more translucent after decellularization than native DLS,which presented a denser and more porous network structure.The results for Young’s modulus proved that the mechanical properties of 0.5 PEG crosslinked DLS were the best and close to that of native livers.The PEG-VEGF-DLS could better promote cell proliferation and differentiation of HUVECs compared with the groups without PEG cross-linking.Importantly,the average density of blood vessels was higher in the PEG-VEGF-DLS than that in other groups at days 7,14,and 21 after implantation in vivo.Conclusions: The PEG crosslinked DLS with VEGF could improve the biomechanical properties of native DLS,and most importantly,their lack of cytotoxicity provides a new route to promote the proliferation of cells in vitro and angiogenesis in vivo in liver tissue engineering.

Introduction

Chronic liver diseases are still among the major etiological factors around the world due to the high unmet clinical needs associated with high mortality [1].Orthotopic liver transplantation is the only definitive treatment for end-stage liver disease and liver failure.However,the unsolvable challenges of insufficient number of donors,immune rejection resulting from preoperative high-risk factors,postoperative inflammation,and immune response remain questions [2].Liver tissue engineering represents a valid strategy for recapitulating the hepatic microenvironment and is a promising innovative path to alleviate the shortage of organ donors.Highly bioactive scaffold material with adequate biocompatibility and vascularization is one of the major challenges in liver tissue engineering.Ideal scaffold materials for liver engineering should contain native liver extracellular matrix (ECM) components,maintain the three-dimensional structures of the intrahepatic vascular system,and satisfy mechanical strength requirements [3,4]as well as feature transfer properties [5–8].The construction strategy based on whole-organ decellularization has become one of the most used techniques for preparing biological scaffold materials for organs.The three-dimensional structure of decellularized single liver lobe scaffolds (DLSs),which is formed primarily by the biochemical and mechanical properties of the ECM,greatly affects the capacity of cells for regeneration and provides vascular networks for oxygen and nutrient transport.In addition,the physiological features of the native organ niche also provide the desired hepatic architecture,cell-cell communication,and cell-ECM interaction [9,10],and can be recellularized to obtain a bio-artificial organ intended to mimic the native liver bothinvitroandinvivo[11,12].

Three-dimensional structures with adequate mechanical strength play an important role in regulating the adhesion and proliferation of vascular endothelial cells as well as rapidly and effectively promoting angiogenesis.However,the currently used decellularized scaffolds exert negative effects on ECM integrity,significantly decreasing the mechanical strength of ECM,which is attributed to the disruption and reduction in collagen crosslinks.Furthermore,the reduction of cytokines such as vascular endothelial growth factor (VEGF) could also be disadvantageous to the adhesion,growth,and differentiation of cells seeded into decellularized scaffolds [13–15].

Based on the above argument,it is necessary to strengthen the mechanical strength and modify the necessary ingredients for DLS.To overcome the limitations of native DLS,polyethylene glycol (PEG) was used for grafted DLS.PEG is a synthetic,biodegradable and water-soluble polyether that has a wide range of applications in tissue engineering.Because it can be modified by regulating the physical and chemical properties of graft-related materials,it has low immunogenicity and is nontoxic [16].Different types of PEG hydrogels are used as conjugated materials and delivery vehicles [17–19].Generally,different natural ECMs such as fibrinogen and hyaluronic acid can be considered for combination with PEG.In theory,PEG enhances the mechanical strengths of DLS similar to heart valves,electrospun vascular grafts,and decellularized lungs [20–22].

The lack of intact vascular networks and sinusoids leads to thrombosis in the recellularization process,which presents another obstacle to the development of DLS after transplantation.The rapid and thorough vascularization of natural and synthetic scaffolds is a critical part of successful cell-based liver tissue engineering for long-term implant survival and tissue integration.Previous strategies for enhancing vascularization in transplanted biological scaffold materials included the delivery of angiogenic genes,growth factors,and vascular-inductive cell types.The vascular endothelial growth factor (VEGF) induces the proliferation of various cells specific for vascular endothelial cells to initiate angiogenesis through binding to transmembrane VEGF receptors.This is because PEGcoupled polymers not only promote angiogenesis but also carry drugs or bioactive molecules to the injury site [17].Therefore,we hypothesized that PEG crosslinked DLS with VEGF may promote angiogenesisinvivo.

The present study aimed to improve the mechanical properties and angiogenesis of DLSs,which would further facilitate transplantation in liver bioengineering.Therefore,after DLS was modified with N-succinimidyl S-acetylthioacetate (SATA),it was crosslinked with a four-arm PEG with the terminal group of diacrylate in rats through a Michael addition reaction between acrylate and sulfhydryl.Given the cytotoxicity of SATA and the optimal reaction time,PEG concentration and the cytocompatibility of modified DLS were determined in the current studyinvitro.For the better vascularization of the transplant,VEGF was crosslinked with the PEG crosslinked scaffolds.Furthermore,the scaffolds were implanted into the dorsal subcutaneous region of rats to evaluate the vascularization of PEG-VEGF modified DLS (Fig.1).Based on the findings,this methodology has a promising application prospect in tissue engineering.

Fig.1.Schematics of the structures of PEG-VEGF crosslinked DLS and experimental design.A: Schematics of the structures of the crosslinked DLS through Michael addition reaction between SATA and PEG,and the distribution of modified VEGF.B: Schematics of PEG-VEGF-DLS application in vitro and in vivo.VEGF: vascular endothelial growth factor;SATA: N-succinimidyl S-acetylthioacetate;HUVEC: human umbilical vein endothelial cell.

Methods

Animals and study groups

Forty liver lobes were isolated from healthy Sprague-Dawley rats weighing 180-220 g,and were subjected to perfusion decellularization.The scaffolds were divided into four groups: native scaffolds with or without VEGF (DLS,VEGF-DLS),and PEG crosslinking scaffolds with or without VEGF (PEG-DLS,PEG-VEGF-DLS).Forty-eight healthy male rats weighing 150-180 g were employed as the transplantation recipients.All animals with the certification number of SYXK (Zhejiang) 2019-0 0 09 were obtained from the Animal Experiment Center of Wenzhou Medical University,Wenzhou,China.The animal experimentations in the present study were sacrificed in accordance with the National Institutes of Health Guidelines for Animal Care and the Animal Welfare Act,and was approved by the Animal Ethics Committee of Wenzhou Medical University.

Single liver lobe decellularization

Adult healthy Sprague-Dawley rats were anaesthetized using the intraperitoneal injection of 5% chloral hydrate (0.6 mL/100 g).After the liver was isolated,the left lateral lobe was exposed.The cannulation of the portal vein (8G indwelling needle) was applied as the fluid inlet whereas a cut of inferior vena cava was used as the fluid outlet.The separated single liver lobe was perfused sequentially at a speed of 1 mL/min with heparin PBS for 1 h,1%Triton X-100 for about 10 h,and with 1% penicillin streptomycin mixture PBS for about 12 h to decellularize the liver lobe.

DNA detection

The quantification of DNA was performed by first extracting whole genomic DNA from the tissues using the Universal Genomic DNA Extraction Kit Ver.3.0 (Takara Bio Inc.,Tokyo,Japan).Samples for DNA quantification were then extracted from the decellularized livers,while native rat livers (n=5) were used as controls.The quantity and purity of DNA were determined using a NanoDrop 20 0 0 UV-Vis spectrophotometer (Thermo Scientific,NY,USA).

Sulfhydrylation and optimization of DLS

The prepared DLSs were crosslinked with SATA at different concentrations (0.1,0.5,1,and 2 mg/mL),with pH of 7.4 and at room temperature for different time durations (15,30,60,90,and 120 min).The extent of sulfhydryl group modification was also evaluated in the present study.Sulfhydrylated DLSs at different concentrations were first washed with PBS extensively for the termination of reaction and then infused with 0.5 mol/L hydroxylamine hydrochloride (dissolved in 100 mmol/L PBS,pH 7.4) to deprotect the acetylated sulfhydryl groups.Lastly,the sulfhydrylated DLSs were extensively washed with PBS to remove the excess hydroxylamine hydrochloride,and then was allowed to react with 10 mmol/L DTNB [5,50-dithiobis (2-nitrobenzoic acid) (Sigma,St.Louis,MO,USA)]in Tris-HCl (0.1 mol/L,pH 8.3) for 15 min.The absorbance value of the solution was measured at 412 nm (OD412),which allowed the calculation of 5-thio-2-nitrobenzoic acid anion (TNBA)contents using the standard curves obtained with cysteine and the further deduction of contents of sulfhydryl groups in sulfhydrylated DLS.

Polyethylene glycol cross-linking and morphological analysis

Four-arm-PEG-acrylate was purchased from Tuoyang Biochem Co.Ltd (Shanghai,China).The sulfhydrylated DLS was perfused with enough PEG for different periods of time (the ratio of acryloyl to sulfhydryl was 20:1) [21].The residual content of SATA in PEG crosslinked DLS was measured as described earlier,which reflected the contents of PEG crosslinked to sulfhydrylated DLS at different time in reverse.Then,the PEG crosslinked DLS was determined using a biomechanical test as previously described [23].Finally,the PEG crosslinked DLS was fixed in 4% paraformaldehyde and embedded in paraffin for hematoxylin and eosin (HE) staining as well as immunofluorescence analyses.

Scanning electron microscopy (SEM)

In order to observe the structure of different scaffolds,native liver,normal DLS,and PEG crosslinked DLS were fixed in 2.5% (v/v)glutaraldehyde in PBS at 4 °C for 24 h.After washing extensively with deionized water,the fixed tissue was freeze-dried overnight using a vacuum freeze dryer.Subsequently,the dry samples were coated under vacuum with platinum alloy at a thickness of 25 nm and visualized under scanning electron microscope (S30 0 0-N,Hitachi,Tokyo,Japan).

Analysis of growth factor content

After optimal PEG cross-linking,the DLS was modified with VEGF (20 ng/mL) using a microinjection pump at a speed of 5 mL/h for 1 h.The concentration of VEGF,transforming growth factor-β(TGF-β),platelet derived growth factor (PDGF),hepatocyte growth factor (HGF),connective tissue growth factor (CTGF) and fibroblast growth factor (FGF) (n=3) were measured according to the instructions for commercial ELISA kits (R&D Systems,Minneapolis,MN,USA) as provided by the manufacturer.

Glycosaminoglycan (GAG) and collagen content

The GAG and total collagen content of frozen native and decellularized livers were determined using rat glycosaminoglycan ELISA kit and total collagen ELISA kit (MLBio,Shanghai,China),respectively.The wet weight of the samples was determined before they were digested and measured according to the protocol provided by manufacturer.The absorbance value of the solution was then measured at 450 nm using a SpectraMax 190 microplate reader (MD,California,USA).

Human umbilical vein endothelial cell (HUVEC) proliferation efficiency in vitro

HUVECs were obtained from PromoCell and cultured in EBM-2 medium (Lonza,Basel,Switzerland) at a humidified atmosphere containing 5% CO 2 and at 37 °C.To investigate cell adhesion and proliferation on the DLS,the scaffolds were cut into 100μm thick sections,adhered to 20 mm cover slips (Wohong Biomedical Equipment Co.,Shanghai,China) and cultured inside a 12-well plate at a density of 5.0 × 104cells.

The scaffolds were sterilized before seeding the cells,using a previously described protocol [24].The viable cells cultured within each group of scaffolds were examined using immunofluorescence with the polyclonal rabbit anti-collagen IV (ab6586) and mouse anti-CD31 antibodies (ab24590) (Abcam,Cambridge,USA) after 36 h of cell seeding.

Cell proliferation in different scaffolds was measured using the Cell-Counting Kit-8 (CCK-8) (Beyotime,Shanghai,China) according to instructions provided by the manufacturer.Cells were then incubated in 10% CCK-8 diluted in normal culture medium at 37 °C until visual color conversion could be observed.The proliferation rates were determined at 24,48,and 72 h after seeding.Simultaneously,the quantitative analysis of apoptotic cells was performed through flow cytometry (BD FACSCanto II,NJ,USA) with Annexin V/FITC and PI apoptosis detection kits (C1062,Beyotime Biotechnology,Shanghai,China) according to the manufacturer’s instructions.The apoptosis-related proteins (Bcl-2,Bax 1:10 0 0,Abcam)were also analyzed using Western blotting.

Evaluation of implanted scaffolds for angiogenesis in vivo

Forty-eight male Sprague-Dawley rats (average body weight 150 g) were used for implantation to investigate the histocompatibility and angiogenesis ability of PEG crosslinked DLS.A section of scaffolds (5 × 8 × 5 mm) from each group was implanted into the dorsolateral body part subcutaneously.Rats were sacrificed at days 3,7,14,and 21 post-implantation.At different time after transplantation,grafts were retrieved and analyzed through HE,immunohistochemical staining (CD31: 1:100,ABclonal,Wuhan,China) and immunofluorescence staining (CD31: 1:100,Abcam;α-SMA: 1:100,Proteintech,Wuhan,China).The CD31 andα-SMA positive results in six fields (× 200 magnification) were quantified by using Image J software (NIH,Bethesda,MD,USA).The maturation index was quantified as the ratio ofα-SMA positive vessels to the total numbers of vessels [25,26].

Statistical analysis

The quantitative results were reported as mean ± standard deviation (SD).Analysis of variance (ANOVA) was also performed to compare the statistical difference levels among the experimental groups.The significance level was set at 0.05 and SPSS 23.0 software (SPSS Inc.,Chicago,IL,USA) was used to carry out the statistical analyses.

Results

Characteristics and optimization of PEG crosslinked DLS

The results showed that the DLS maintained their threedimensional structure after decellularization of a single liver lobe and gradually became homogeneously translucent due to the removal of the cellular contents.The PEG crosslinked scaffold was darker and more translucent.No significant difference was noted in the appearance and shape between PEG crosslinked and normal DLS (Fig.2 A).Furthermore,the rigidity of PEG crosslinked scaffolds was increased.On the other hand,the analysis of DNA content showed that after washing with the Triton X-100,the DNA content decreased from 3297.28 ± 291.57 ng/mg to 127.38 ± 11.79 ng/mg,which confirmed significant DNA depletion (Fig.2 B).This indicated that Triton X-100 was efficient in removing DNA from the liver ECM.

Fig.2.Characteristics and optimization of PEG crosslinked DLS.A: The macroscopic appearance of decellularized single liver lobe [above: general process of decellularizing single liver lobe;below (left to right): DL S,0.1-PEG-DL S,0.5-PEG-DL S,1-PEG-DL S,2-PEG-DL S].B: Residual DNA quantification (n=5,∗P <0.01).C: The trend of OD412 value of sulfhydryl introduced into the DLS at various concentrations and reaction time.D: HE staining showing the absence of cellular components and nuclear material in decellularized scaffolds,and the high porosity at higher concentration of SATA in PEG crosslinked scaffolds.Scale bars,50 μm.E: The OD412 value of residual sulfhydryl in PEG crosslinked DLS at different reaction time.F: The Young’s modulus of native liver,DLS,and different concentrations of PEG crosslinked DLS (n=6;NS indicates P >0.05,∗∗∗P <0.001,∗∗∗∗P <0.0 0 01).G: Scanning electron microscopy images of the native liver,DL S and 0.5-PEG-DL S.Scale bars,200 μm.H: The amount of total collagen (n=3).I: The content of GAG (n=3).The measured values were normalized against the weight tissue (mg) before and after decellularization.J: Immunofluorescence staining of collagen Ⅰ,collagen Ⅳ,fibronectin,and laminin.The stained sections show intact ECM.Scale bars,50 μm.K: Levels of cytokines in native liver,DLS,and 0.5-PEG-DLS (n=3).L: The VEGF modification in DLS and 0.5-PEG-DLS (NS indicates P >0.05,∗P <0.05,∗∗P <0.01).DLS: decellularized single liver lobe scaffold;PEG: polyethylene glycol;SATA: N-succinimidyl S-acetylthioacetate;GAG: glycosaminoglycan;VEGF: vascular endothelial growth factor;HGF: hepatocyte growth factor;EGF: epidermal growth factor;FGF: fibroblast growth factor;CTGF: connective tissue growth factor;PDGF: platelet derived growth factor;TGF-β: transforming growth factor-β;ECM: extracellular matrix.

The response curves of sulfhydrylation for different concentrations of SATA with different reaction time suggested that the optimal reaction time of SATA for the sulfhydrylation of DLS during the PEG cross-linking process was approximately 1 h (Fig.2 C).Similarly,the OD412 values of residual sulfhydryl in PEG crosslinked DLS extracted under the same concentration levels of PEG suggested that the optimal reaction time for PEG was about 4 h(Fig.2 E).The results for Young’s modulus also proved that the mechanical properties of PEG crosslinked group corresponding to 0.5 mg/mL SATA were not significantly different from the native livers.However,there was evidence that the mechanical properties decreased with increasing PEG concentration (Fig.2 F).

Moreover,no residual cellular components and nuclear material were found in DLS with or without PEG as compared with HEstained native livers.Notably,HE staining indicated that the porosity was increased,and fibers also appeared fused with the increase in the concentration of PEG (Fig.2 D).In addition,based on the aforementioned findings of the present study,it was found that PEG crosslinked DLS had the optimal mechanical properties,and the optimal SATA concentration and reaction time for sulfhydrylation were 0.5 mg/mL and 1 h respectively,whereas the optimal reaction time for PEG cross-linking was 4 h (named as 0.5-PEG-DLS).Therefore,it was established that the PEG concentration that corresponded to a concentration of 0.5 mg/mL SATA was the optimal choice for fabricating DLSs for subsequent research.

The SEM results also revealed the retention of ECM matrix within DLS and that the lack of cells or nuclear debris was maintained.Meanwhile,the 0.5-PEG-DLS exhibited denser and potentially continuous fibers and higher porosity as compared with the control DLS (Fig.2 G).Moreover,the average total collagen and GAG content of the livers after decellularization appeared to be slightly higher per milligram of weight tissue than that of the native DLS (Fig.2 H and I).This may be attributed to the denser structure after cross-linking and the increase of relative contribution of the remaining ECM compounds to the total weight of the sample.Therefore,a direct comparison could not be drawn between pre-and post-decellularization.The immunofluorescence staining for the liver ECM proteins,such as collagen type I/IV,fibronectin and laminin,showed that the DLS preserved the three-dimensional ultrastructure and components of ECM (Fig.2 J).The SEM and immunofluorescence staining of the 0.5-PEG-DLS also indicated that the ECM components were well-preserved and no cells were visible but a denser and porous network structure remained as compared with the decellularized normal liver scaffolds.

In addition,major cytokines,including HGF,TGF-β,FGF,EGF,CTGF,VEGF and PDGF,had no significant difference between the 0.5-PEG-DLS and normal DLS (Fig.2 K).Furthermore,the growth factors were essentially lost with the removal of cells.Therefore,the 0.5-PEG-DLS was modified with VEGF to beneficially promote cell proliferation and angiogenesis (Fig.2 L).

HUVEC proliferation efficiency in modified scaffolds in vitro

HUVECs were culturedinvitroto evaluate the cytocompatibility of PEG and VEGF modified DLS.The immunofluorescence staining results confirmed the localization and proliferation of HUVECs on all the scaffolds after co-culturing for 36 h (Fig.3 A).In general,the cell proliferation on the 0.5 PEG-DLS with or without VEGF was higher than that on the un-crosslinked DLSs.The apoptosis effects of PEG on HUVEC for two days were also analyzed in the current study by flow cytometry assays (Fig.3 B).The percentage of apoptotic HUVECs in the PEG groups with or without VEGF was not significantly different from that of the control group (Fig.3 F).However,the percentage of apoptotic HUVECs was higher in the DLS group with or without VEGF than the control group.In addition,the group with VEGF revealed a higher apoptotic rate,which may be attributed to the effect of VEGF in promoting cell proliferation and differentiation.VEGF is an essential molecule for maintaining the survival and growth of endothelial cell in primary cultures.The CCK-8 assay was also carried out to confirm cell proliferation in each group of DLS (Fig.3 C),and it was found that the PEG crosslinking group had no adverse effects on cell proliferation.

Fig.3.In vitro proliferation of HUVECs on PEG crosslinked DLS.A: Immunofluorescence of HUVECs co-cultured with DLS scaffolds (Scale bars,50 μm).B: Flow cytometry assays showing the level of apoptosis of HUVEC cells in each group.The percentage of apoptotic HUVECs in PEG groups in the presence or absence of VEGF.C: Results of the CCK-8 proliferation assay.0.5-PEG crosslinked DLS did not have adverse effects on cell proliferation.D and E: Expression of apoptosis-related proteins as determined by Western blotting.F: Percentage of apoptotic HUVEC in each group (NS indicates P >0.05;∗∗P <0.01).HUVEC: human umbilical vein endothelial cell;PEG: polyethylene glycol;DLS: decellularized single liver lobe scaffold;VEGF: vascular endothelial growth factor.

The results of the Western blotting analysis of apoptosis-related proteins were also consistent with those of flow cytometry.The PEG crosslinked scaffolds had better anti-apoptotic ability (Fig.3 D and E).PEG-VEGF-DLS had a better biocompatibility for HUVECs growth and proliferation.

Angiogenesis after subcutaneous implantation in rat model

Most DLS groups without PEG cross-linking developed purulent infection,and granulation tissue gradually formed during the different stages.At day 7 post-implantation,neovascular cord-like structure appeared in the interior of the VEGF-DL S,0.5-PEG-DL S,and 0.5-PEG-VEGF-DLS groups,especially in the VEGF-modified group.Whereas,there was still relatively severe inflammation,and the formation of vascular cord structure was not significant in the DLS group.Notably,there was a significant increase in the number of new blood vessels in all groups at days 14 and 21 postimplantation (Fig.4).

In order to accurately quantify the development of vascular structures,neovascularization was identified as the presence of endothelial cell clusters using a CD31 marker through immunochemical staining (Fig.5 A),and the degree of neovascularization was counted accordingly.There was a significant increase in the number of blood vessels in the 0.5-PEG-DLS and 0.5-PEG-VEGF-DLS relative to those without PEG crosslinked DLS at days 7,14,and 21.Importantly,it was noted that the implantation of 0.5-PEG-VEGFDLS induced a significant increase in the number of capillaries at days 14 and 21 (Fig.5 B).Furthermore,to determine the percentage of mature vessels,the maturation indexes were calculated based on the CD31 andα-SMA double immunofluorescence staining results obtained at day 21.The maturation index of PEG-VEGF-DLS group was the highest among all the groups (Fig.6).Based on these results,the DLS modified with PEG and VEGF was effective for promoting angiogenesis and maturation.

Fig.6.Immunofluorescence analysis to estimate the maturation index of regenerated vessels in subcutaneous implantation at day 21.A: Representative images of endothelial cells (ECs) with CD31 (green) and smooth muscle cells (SMCs) with α-SMA (red) after immunofluorescence staining at day 21 post-implantation.Scale bars,50 μm.The PEGVEGF crosslinked scaffolds improved the number of small blood vessels and the maturation index.B: Quantification of CD31 positive cells.C: Quantification of α-SMA positive cells.D: Quantification of maturation index (%).The maturation index was based on the comparison of the percentage of α-SMA positive cells area in the total area of CD31 positive cells respectively.∗P <0.05,∗∗P <0.01.PEG: polyethylene glycol;VEGF: vascular endothelial growth factor.

Discussion

Decellularized scaffolds that derived from native tissues or organs have been widely investigated and acknowledged as advanced functional biomaterials for regenerative medicine as well as tissue engineering applications.Accordingly,numerous decellularization protocols have been developed to obtain DLS,while the perfusion decellularization method has been the most selected method for this purpose [27–31].The present study used Sprague-Dawley rats because they are one of the most widely used experimental animal models.Although the rat liver consists of six lobes,the different vascular diameters that correspond to each lobe of the rat’s liver yield different volumes of perfusate due to changes in the perfusion pressure during the process of decellularization.Thus we decellularized the single liver lobe using the reagent buffer and washing protocol by inserting a catheter into the intrahepatic branch of the portal vein on the left lateral liver lobe.The operation retaining only one vessel provided sufficient material for one liver lobe perfusion-decellularization and the rapid preparation of DLS.

The cell nuclei in tissue organs were removed through the intravascular injection of enzymatic (DNase) solutions such as Triton X-100 and/or sodium dodecyl sulfate (SDS).It was found that a significant amount of GAG and collagen were removed during perfusion with SDS as compared with Triton X-100 [28,32].In particular,SDS has a high affinity for proteins,which increases the difficulty of their removal from tissue after decellularization.In comparison,a lower concentration of Triton X-100 could generate a larger collagen content and improve ECM protein structure as well as thermal stability [28].In the present study,1% Triton X-100 solution was used to prepare DLS and evaluate the DNA,collagen,and GAG contents in DLS,as well as for the histology evaluation.The results showed that 1% Triton X-100 could effectively remove the nucleic acids,as shown by the detection results of DNA content and the absence of fluorescence from all DLS stained with DAPI.The DLS preserved the specific vasculatures,while it had no significant negative influence on the content of collagen and GAG in DLS.

The main role of tissue engineered scaffolds is to provide suitable three-dimensional microenvironments and mechanical support for cell growth and function [33].However,the decellularization processes are usually quite harsh not only toward the cell components containing DNA but also to the proteins and other ECM components.A large portion of bioactive molecules that were removed during liver perfusion decellularization caused damage to the microarchitecture,decreased the uniaxial and biaxial strength,and the maximum tangential stiffness of ECM,and thus limited the functional recovery.

In order to further increase the mechanical strength and improve the three-dimensional structures of DLS in the current study,a series of four-arm PEG was integrated into DLS by the perfusion method in the presence of SATA,a protein modification agent,and sulfhydryl groups were added to proteins [21,34].By comparing different concentrations of PEG,the application of 0.5 mg/mL SATA with PEG (0.5-PEG-DLS) significantly improved the mechanical properties and there was no significant difference from the native livers,which retained a denser and porous network structure similar to results of previous PEG crosslinked lungs and electrospun [20,22].Our SEM results also showed that 0.5-PEG-DLS formed a denser space network structure with increased porosity.Notably,PEG crosslinked group made the spatial structure of DLS more stable and maintain similar mechanical properties as the natural liver.

Normal angiogenesis occurs when existing vasculature begins to dilate to sprout new capillaries through migration,proliferation,and maturation of endothelial cells as well as endothelial progenitor cells.Proliferation must occur following the migration of endothelial cells into this newly formed space.However,VEGF,which is a main player in angiogenesis,is capable of provoking the migration and proliferation of endothelial cells since endothelial cells are normally quiescent [35].Moreover,the promising activity of VEGF is hampered mainly by its short half-life.Also the VEGF is significantly decreased in DLS.Therefore,in this study,we established a stable and functional vessel network,and VEGF solutions were perfused into 0.5-PEG-DLS to supplement defects of insufficient vessels.In addition,appropriate VEGF concentration was also important.Therefore,we selected a concentration similar to that in natural liver because high concentration also has undesirable effects.PEG emer ged as the best candidate for protein modification because of its highly hydrophilic characteristic [36].

It has been reported that gelatin and PEG based hydrogels provide a powerful cell culture platform for tissue engineering [37,38].Although grafting the scaffolds with PEG may compensate the microenvironments and provide mechanical support,it can also sustain the delivery of VEGF from controlled release systems and protect it from degradation [17].However,this technique may also negatively impact the repopulation of DLS.Therefore,the present study evaluated the bioactivity and cell-supporting properties of DLS.The efficacy of the system was demonstrated in co-cultured DLS with HUVECsinvitro,and it exerted no adverse effects on the viability of HUVECs.The VEGFs were encapsulated within the PEG-DLS as a transplantation vector to evaluate the effect of PEG crosslinked DLS on angiogenesis.There was evidence that a release system performed positively as compared with VEGF-free PEG-DLS for vascularization in a subcutaneous embedding modelinvivo.Although most of the intracellular materials that contain antigens eliciting host inflammatory response had been removed,the response was present even after the transplantation of DLS.Large amounts of inflammatory cell infiltrated around the DLS group after transplantation and were eventually replaced by the granulation tissue.

Although the present study found a negative state of transplanted PEG crosslinked DLS grafts,it was noted that inflammatory cells in the PEG-DLS were uniformly dispersed and abscesses were rarely seen post-transplantation.In addition,the number of blood vessels of PEG crosslinked group was significantly more effectively distributed when crosslinked with VEGF as compared with the normal DLS groups at days 7,14 and 21 after implantation.We observed that the PEG-VEGF had the capacity to significantly enhance angiogenesis and maturation at day 21.It is estimated that the presence of PEG in the DLS allows for the storage and release of VEGF that can also enable ECM scaffolds to serve as a reservoir for the delivery of factors to cells within the matrix in controlled concentrations and gradients.The PEG crosslinked DLS may also be used as an optimization strategy to deliver VEGF or other growth factor in the field of liver regeneration.

Based on the findings,this study represents a critical step in the development of functional recellularized liver scaffolds,which can be used not only for transplantation but also for drug screening and disease-modeling studiesinvitro.Our optimization results show that the application of PEG crosslinked DLS with VEGF as mixed grafts provides a naturally derived three-dimensional microenvironment structure and better mechanical properties for cell growth and proliferation as well as promotes angiogenesisinvivo.Nonetheless,the underlying angiogenesis mechanism should still be investigated.Future research should also focus on orthotropic transplantation,and the long-term effectiveness of PEG grafted decellularized organs in larger animal models should be evaluatedinvivo.

Acknowledgments

We thank Drs.Mao-Chao Ding,Jun-Jun Xu,Wen Xu and Huai-Rui Cui for the experimental guidance and assistance with this study.

CRediTauthorshipcontributionstatement

Jian-SeZhang:Data curation,Formal analysis,Methodology,Validation,Writing– original draft.Zhi-BinWang:Data curation,Formal analysis,Methodology,Validation,Writing– original draft.Zhi-ZhenLai:Data curation,Formal analysis,Methodology,Validation,Writing– original draft.Jing-WenYang:Investigation.Wen-JingSong:Investigation.Yu-BingWei:Investigation.JinMei:Supervision.Jian-GuangWang:Conceptualization,Funding acquisition,Project administration,Writing– review &editing.

Funding

This study was supported by grants from Natural Science Foundation of Zhejiang Province (LY20H180011),National Natural Science Foundation of China (81970653) and Medical and Health Science and Technology project of Zhejiang (2016KYA061).

Ethicalapproval

This study was executed in accordance to the National Institutes of Health guidelines for animal care and Animal Welfare Act and was approved by the Animal Ethics Committee of Wenzhou Medical University (2021-0469).

Competinginterest

No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.