Ocular surface repair using decellularized porcine conjunctiva

Journal Pre-proof

Ocular surface repair using decellularized porcine conjunctiva Long Zhao , Yanni Jia , Can Zhao , Hua Li , Fuyan Wang , Muchen Dong , Ting Liu , Songmei Zhang , Qingjun Zhou , Weiyun Shi PII: DOI: Reference:

S1742-7061(19)30746-9 https://doi.org/10.1016/j.actbio.2019.11.006 ACTBIO 6437

To appear in:

Acta Biomaterialia

Received date: Revised date: Accepted date:

10 June 2019 28 October 2019 4 November 2019

Please cite this article as: Long Zhao , Yanni Jia , Can Zhao , Hua Li , Fuyan Wang , Muchen Dong , Ting Liu , Songmei Zhang , Qingjun Zhou , Weiyun Shi , Ocular surface repair using decellularized porcine conjunctiva, Acta Biomaterialia (2019), doi: https://doi.org/10.1016/j.actbio.2019.11.006

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Ltd on behalf of Acta Materialia Inc.

Ocular surface repair using decellularized porcine conjunctiva

Authors: Long Zhaoa,1, Yanni Jiab,1, Can Zhaob, Hua Lid, Fuyan Wangd, Muchen Dongb, Ting Liua,c, Songmei Zhanga, Qingjun Zhoua*, Weiyun Shia,b*

a

State Key Laboratory Cultivation Base, Shandong Provincial Key Laboratory

of Ophthalmology, Shandong Eye Institute, Shandong First Medical University & Shandong Academy of Medical Sciences, Qingdao 266000, China b

Shandong Eye Hospital, Shandong Eye Institute, Shandong First Medical

University & Shandong Academy of Medical Sciences, Jinan 250000, China c

Qingdao Eye Hospital, Shandong Eye Institute, Shandong First Medical

University & Shandong Academy of Medical Sciences, Qingdao 266000, China d

Department of Ophthalmology, Qilu Medical College of Shandong University,

Jinan 250000, China *

Corresponding author: Weiyun Shi, M.D., Ph.D.& Qingjun Zhou, Ph.D.

Shandong Provincial Key Laboratory of Ophthalmology, Shandong Eye Institute, Shandong First Medical University & Shandong Academy of Medical Sciences, Qingdao 266000, China Tel: +0086-531-81276002. Fax: +0086-532-85899270 E-mail: [email protected]. [email protected] 1

These authors contributed equally.

1

Abstract The primary functions of the conjunctiva embody ocular surface protection and the maintenance of the tear film equilibrium. Severe conjunctival defects such as symblepharon may impair the integrity of ocular surface and cause loss of visual functions. Here we report the use of a decellularized porcine conjunctiva (DPC) for conjunctival reconstruction in rabbit models and in clinic. Our results show that the major xenoantigens are efficiently removed, while abundant matrix components and integrated microstructures are well preserved in the DPC. These characteristics provide mechanical support and favorable histocompatibility for repairing damaged conjunctiva. The DPC application has demonstrated enhanced transplant stability and improved epithelial regeneration in severe ocular surface damage comparing to those of amniotic membrane (AM), the most frequently applied matrix for ocular surface reconstruction nowadays. In order to test the DPC performance in clinic, three patients with pterygium and one patient with symblepharon underwent transplant with DPC. The grafts in all cases were completely re-epithelized and no graft melt or fibroplasia were observed. These results suggest that the strategy we developed is feasible and effective for conjunctival reconstruction and ocular surface repair.

Keywords Decellularization, conjunctiva, symblepharon, conjunctival reconstruction, ocular surface repair

2

Statement of significance In this study, we adopted an innovative approach to prepare decellularized porcine conjunctiva (DPC). The intricate conjunctiva-specific structures and abundant matrix components were preserved in DPC, which offers favorable mechanical properties for graft. DPC has shown positive effects in ocular surface repair, which has been proven particularly in a rabbit model with severe symblepharon. Reconstructed conjunctiva by DPC exhibited epithelial heterogeneity, extremely resembling that of native conjunctiva. In addition, results from clinical studies were encouraging for pterygium and symblepharon and clinical application of DPC is promising.

3

1. Introduction

The ocular surface consists of continuous conjunctiva and cornea, and injury to either part may result in system-wide dysfunction. Conjunctiva provides immune surveillance and maintains the equilibrium of the tear film [1]. Conjunctival

repair

is

a

prerequisite

for

successful

ocular

surface

reconstruction [2, 3]. Conjunctiva can repair itself spontaneously upon injury [4], however, in cases with extensive conjunctival injuries, such as chemical/thermal burns, ocular cicatricial pemphigoid and Stevens-Johnson syndrome, if left unattended, fornix shortening, symblepharon, corneal opacity are highly potential. Hence, an appropriate biomaterial is required for optimal healing after extensive conjunctival injuries [5, 6]. Amniotic membrane (AM) is currently the most frequently applied substitute for ocular surface reconstruction due to its low immunogenicity, antimicrobial, antiviral, antifibrotic, and antiangiogenic properties [7, 8]. Although AM is a useful adjunct in the conjunctival reconstruction, the therapeutic effect is not stable in long-term observation, especially in cases of large-scale conjunctival defects [9, 10]. With the development of material science,

synthetic

matrices

based

on

collagen

I,

fibrin

and

poly

(lactide-co-glycolide) (PLGA) have also been tested in animal models and showed successful ocular surface reconstruction [11-14]. Nonetheless, it is difficult for artificial materials to fully simulate the mechanical properties and abundant matrix components of natural conjunctiva. The clinical studies of cultivated autologous conjunctival epithelium transplantation have been reported, which can effectively treat large-size conjunctival defects [15, 16]. However, the cultivation of autologous conjunctival epithelium is restricted to specialized laboratories, which increases the cost and complexity of the operation. The source of conjunctival epithelium is also limited, such as in a conjunctival defect involving both eyes. 4

Apart from the basic properties of an epidermal biomaterial, such as stability, resilience, absence of immune antigens and reduced rejection, it’s necessary for an ideal conjunctival substitute to possess similar compositions and structures of native conjunctival extracellular matrix (ECM) [17]. The conjunctival ECM plays a key role in the differentiation of the conjunctival epithelium and the prevention of scar formation [18, 19]. Though autologous conjunctival tissue is the optimum selection clinically for its host-like tissue characteristics and free immunogenicity, it is limited by donor area. Currently, decellularization and extracting native ECM from xenogeneic tissue is a convenient and feasible way. Decellularized materials have been studied in various tissue types, several prominent decellularized materials have been successfully applied in clinic, such as small intestine submucosa, dermal matrix and bovine pericardium [20-22]. However, due to the tissue-specificity of ECM, the protocols of decellularization are difficult to be standardized, and most of the studies on decellularized materials are restricted to in vitro or short-term biocompatibility test in vivo [23, 24]. In the present study, we attempted to find a decellularized conjunctival material to tackle with ocular surface conjunctiva injuries. The physical properties of the decellularized porcine conjunctiva (DPC), including macroscopic and microscopic structures and mechanical properties were evaluated and compared with those of native porcine conjunctiva (NPC). Furthermore, the biological properties of DPC including major ECM components, antigenic substances, cytotoxicity and regeneration potential were also evaluated in vitro. In addition, attempts of DPC application in treatments of a rabbit surgical trauma model and a rabbit symblepharon model had been made. The anatomical, physiological and pathological results of both models were evaluated. Conjunctival epithelium regeneration and goblet cell formation were detected by histological staining and immunohistochemical studies. Finally, we evaluated the therapeutic efficacy of DPC in the patients 5

with pterygium and symblepharon.

2. Materials and methods 2.1 Ethics All animals were carried out in accordance with the Association for Research in Vision and Ophthalmology (ARVO) Statement for the Use of Animals in Ophthalmic and Vision Research. Animal experimental procedures were approved by the Ethics Committee of Shandong Eye Institute. Human AM and conjunctiva tissue were prepared and preserved by the Eye Bank of Shandong Eye Institute and handled according to the tenets of the Ethics Committee at Shandong Academy of Medical Sciences. The clinical trial was registered in the Chinese Clinical Trial Registry (Tissue Engineering Conjunctiva

for

the

Treatment

of

Pterygium

and

Atretoblepharia,

NCT-02911532) and approved by the Ethics Committee of Shandong Eye Hospital in Jinan, China. Under the informed consent and potential risk explanation, patients were enrolled and implanted with DPC. In the clinical trial, AM was selected as a control. The clinical trial is currently in its preliminary stages and several representative cases (3 patients with pterygium and one with symblepharon) were presented in the study. 2.2 Preparation of biomaterial Three-breed crossbred swine (4-6 months, either gender) without virus infection and medical history were selected. Whole porcine eyes were obtained within 1-3 hrs postmortem and the conjunctivas were isolated. All decellularization steps were carried out with agitation at room temperature in sterile conditions. The native conjunctivas were first treated under 200 MPa high hydrostatic pressure for 3 times, 1 minute each and then processed with 1% (w/v) TrionX-100 (Sigma) for 2 hrs. After careful removal of conjunctival epithelium by epithelium scraper, the samples were cleaned by ultrasound with 20 HZ for 2 minutes, and then incubated in super nuclease (400 U/ml, Sino 6

Biological Inc) and 1% TrionX-100 at 37 ℃ for 2 hrs. Finally, the samples were washed 6 times in PBS, 10 minutes each. The prepared DPCs were sterilized by γ-irradiation (8 kGy, Zhongjin Irradiation, Qingdao, China). AMs were prepared according to the protocol previously described [25]. AMs were treated by EDTA (0.02%, w/v, Sigma) at pH 8.0 for 2 hrs at room temperature. After that, the epithelial cells were scraped off by cell scraper (sigma) and washed thoroughly with PBS. These biomaterials stored at - 20 ℃ until use. 2.3 Quantitative analysis The dry weight of conjunctiva was recorded. The DNA was extracted using GeneJET Genomic DNA purification kit (Thermo Scientific, USA). DNA quantification kit (Invitrogen, USA) was used according to the manufacturer’s instruction. Porcine α-Gal ELISA detection kit (Mlbio, Shanghai, China) and GAG ELISA detection kit (Heng Kang Tiansheng, Beijing, China) were used for quantitation according to the manufacturer´s protocol. Collagen content was determined by Sirius red collagen detection kit (Chondrex, USA). To determine the content of residual nuclease in DPC, a quantitative analysis is performed. The dried DPC was minced and immersed in assay buffer (10mg/ml; 50 mM Tris-HCl, 5mM MgCl2, 1.5 μM BSA, PH 8.0) and agitated at 200rpm for 2 hrs at room temperature. The supernatant was collected and mixed with DNA solution from salmon testes (0.5mg/ml; Sigma).The mixture were incubated at 37 ℃ for 30 minutes. After that, 4% TCA was added and incubated at 4 ℃ for 10 minutes. The supernatants were added into 96-well UV-transparent microplates (Corning) and detected at 260 nm. Nuclease concentrations were determined using a standard curve. All quantitative experiments were repeated at least three times. 2.4 Histological and immunostaining The samples were fixed in 10% (v/v) neutral buffered formalin at room temperature overnight, dehydrated and embedded in paraffin wax. The 7

paraffin-embedded sections were cut to 4 μm thickness. Hematoxylin & eosin (H&E), Periodic Acid-Schiff (PAS) and Masson's trichrome (MT) staining were performed as previously described [26]. Alcian Blue staining was done according to the manufacturer’s instructions (Vector Laboratories, USA). And all the goblet cells from transplant area were counted in PAS and Alcian Blue stained sections (with a total of six sections per eye, 200 × magnification). The lymphocyte cells were assessed in H&E stained slides. Cells were quantitated by counting in four random equal-sized fields per slide. Six slides were selected for each sample as the same multiplier. The protocol of immunohistochemistry (IHC) and immunofluorescence (IF) were performed as previously described [27]. For IF staining, the frozen tissue sections were fixed in 4% paraformaldehyde at room temperature for 15 min. Antibodies used in the study are shown in Table S1. The collagen I antibody reacts with pig and human, but not with rabbit. Lectin from Bandeiraea simplicifolia BS-I Isolectin B4 conjugating FITC was used to detect α-Gal. 2.5 Ultrastructural analysis For Scanning electron microscopy (SEM), tissues were fixed with 2.5% phosphate-buffered glutaraldehyde at 4 °C for 4 hrs, dehydrated in graded ethanol, and dried by using critical point drying. The specimens were mounted on stubs and sputter-coated with gold. Samples were visualized via SEM (VEGA3, TESCAN, CZE). For transmission electron microscopy (TEM), the tissues were fixed in 2.5% phosphate-buffered glutaraldehyde at 4 °C for 4 hrs, post-fixed with 1% osmium tetroxide in 4 °C for 2 hrs. The samples (60 nm thin) were sectioned using an ultramicrotome (UC6, Leica, Wetzlar, Germany) and contrast-stained with 2% aqueous uranyl acetate and 1% aqueous lead citrate. JEM-1400(JEOL, Tokyo, Japan) was used to observe the results and pictures were taken by Olympus iTEM 5.0. After that, the fibril diameter, number, and interfibrillar average distance were analyzed by Nano Measurer 1.2 software. The measure of parameters was repeated at least six times with different TEM 8

images. 2.6 Biomechanical properties The biomechanical properties were studied using a static material testing machine (HZ1007E, Hen Zhun, China) with a 10 KG load cell. The samples were kept moisture by dropping a microscale of PBS in the testing process. The test speed was set at 2 mm/min. Deformation was recorded when the force was greater than 0.098 N. The break point was determined when the force instantly declined by more than 10%. The measurements were analyzed by means of TM2101 (V 5.50) software. All experiments were performed as sextuplicate. 2.7 Cell culture and proliferation assay Human conjunctival epithelial cells (HCECs) were isolated from conjunctiva tissue following protocol previously described [28]. HCECs were seeded in a 24-well plate at a density of 7 × 103 cells/well in DMEM/F12 (Sigma) and 10% FBS (Hyclone) for 6 days. Conditioned culture medium used in experimental group was pretreated. Sterile DPCs were soaked in medium (6 cm2/ml) for 48 hrs at 37 °C, and then removed with a 30 μm filter. Cells were counted every 24 hrs using TC20TM Automated Coulter Counter (BIO-RAD). Each experiment was performed in triplicate. 2.8 Three-dimensional (3D) reconstitution of conjunctiva in vitro The conjunctiva taken from rabbit was cut into approximately 2 mm 3 fragments in cold PBS containing 0.5 mM dithiothreitol (DTT) and incubated in chelating buffer at 4 °C with constant stirring for 10 minutes. Approximately 5 × 105 single cells were then co-cultured with DPC in a 6 cm dish in medium (DMEM/F12 and RPMI1640 [Sigma] as 1:1) supplemented with 10%FBS, 25 ng/ml EGF (PeproTech) and 10 ng/ml FGF-2 (PeproTech) for 14 days at 37 °C with 5% CO2. In the live-dead staining, DPC was dried on a small glass sheet (10 mm × 10 mm) to maintain transparency. Three slides containing the DPC were 9

placed in 12-well plates. The same pieces of glass coated by collagen I (1% v/v, Sigma) were used for control. HECE cells were seeded into 12-well plates at a density of 1 × 105 cells/well. The cells were cultured for 36 hrs, and then stained for 30 minutes with a live or die double dye kit (Abbkine, China) according to the manufacturer's instruction. Three images at different positions of the slides were recorded by a phase-contrast microscope (Nikon Eclipse Ti-U, Tokyo, Japan), and the number of living cells was counted by Photoshop CS 6. 2.9 Application of DPC in rabbit surgical trauma model Young New Zealand white rabbits (either gender, 10-15 weeks, 2-2.5 kg) were maintained in the animal facility of Shandong Eye Institute. Rabbits were anesthetized with intramuscular ketamine hydrochloride (40 mg/kg; Gutian, China) and chlorpromazine hydrochloride (20 mg/kg; Shanghai, China). A rectangular bulbar conjunctival defect of 7.5 mm × 5 mm in the upper temporal part of the right eye adjacent to the limbus was performed. Rabbits were divided into three groups, 32 rabbits per group. The defects were covered with DPC, AM or remained ungrafted (control groups). All transplants were sutured with ten single stich sutures using nylon 10-0. For post-operative treatment, all eyes were given glucocorticosteroids ointment 2 times every day for 5 days, which were then replaced with glucocorticosteroids eye drops, and were gradually cut down over the course of 1 month. The repairment of epithelium lesions were evaluated by fluorescein sodium staining under slit-lamp. Conjunctival hyperaemia was graded using the scoring system: grade 0, no hyperaemia; grade 1, slight hyperaemia; grade 2, moderate hyperaemia; grade 3, severe hyperaemia. 2.10 Application of DPC in rabbit symblepharon model Rabbits were anesthetized and all bulbar conjunctiva and most of the palpebral conjunctiva were excised. Postoperatively, the conjunctival sac was monitored under slit-lamp every other week. After 8 weeks, the severity of 10

symblepharon was evaluated in accordance with a previously reported grading system [29]. Six rabbits were selected as models of severe symblepharon based on the following three parameters. First, the shortest distance through symblepharon from the lid margin to the limbus was shorter than the length of the palpebral conjunctiva (II, III or IV). Second, the width of the symblepharon was more than one-third of the eyelid length (b or c). Third, the inflammatory activity was mild (1+) or moderate (2+). The rabbits were anesthetized and received surgery of conjunctival sac reconstruction with implantation of DPC and AM, respectively. The sutures were removed two weeks after operation. The post-operative medication was consistent with the surgical trauma model. The depth of the conjunctival fornix was measured before and after surgery using a blunt probe with a scale. The relative fornix depth was calculated as the postoperative depth of fornix divided by the depth of baseline. 2.11 Application of DPC in clinical cases All surgeries were performed with peribulbar anesthesia using 2% lidocaine and 0.75% bupivacaine. The pterygium was removed from the cornea by blunt method and detached from its base by sharp scissors. A suitable size DPC graft was used to replace the defective conjunctiva and sutured with 10-0 nylon sutures (The brief surgical procedure is shown in Supplementary

Video).

After

surgery,

a

combination

of

tobramicina

dexametasona eye drops and pranoprofen eye drops were given 4 times daily and tobramicina dexametasona ointment was administered before bedtime. In the symblepharon surgery, the adhesions of conjunctiva to the eye globe were carefully separated, and the bare sclera was exposed. The subconjunctival fibrovascular tissues were removed with minimal damage to conjunctival tissue. DPC graft was fixed by 10-0 nylon sutures in the defect area of conjunctiva. A bandage contact lens was placed at the end of the surgery. Postoperatively, patient received 0.3% levofloxacin, 0.1% fluorometholone eye drops and Vitamin A palmitate eye gel 4 times daily and nightly tobramicina 11

dexametasona ointment. At 2 weeks postoperatively, sutures were removed and fluorescein sodium staining under cobalt blue light was used to examine the integrity of the epithelium. 2.12 Statistical analyses Statistical analyses were performed using GraphPad Prism 6. Each experiment was performed triplicate independently, if not further indicated, and data were presented as mean ± SD. Student’s t test was used in comparing the contents of DPC, AM or native conjunctiva. A value of p < 0.05 was considered as statistically significant.

3. Result

3.1 Architectural feature of DPC In our study, after decellularization, the DPC had no significant difference with isolated NPC in anatomical structure, except the tissue appeared from pink color to white (Fig.1A, E). H&E staining showed no remnant cells in DPC, and the vasculature (black arrow indicated) and basement membrane structure were well-retained (Fig. 1F). SEM and TEM were performed to evaluate the ultrastructure of DPC. Electron microscope analysis of the cross-sectional images of NPC revealed that stromal cells embedded in the ECM fibers (Fig. 1C, D, arrow indicated). After decellularization, the cellular structures were removed. Exposed collagen fibers in DPCs were similar to that in native conjunctiva in diameter, arrangement and interstice (Fig. 1G, H).

12

Fig.1. Histological and ultrastructural comparison of native porcine conjunctiva (NPC) and decellularized porcine conjunctiva (DPC). The photograph of NPC (A) and DPC (E). Scale bar, 10 mm. Histological comparison of native (B) and decellularized conjunctiva (F) by H&E staining. Scale bar, 50 μm. Vessels in NPC and the retained vessels in DPC are indicated by black arrows. Scanning electron microscope (SEM) image of NPC (C) and DPC (G). Transmission electron microscope (TEM) image of NPC (D) and DPC (H). Conjunctival stroma cells are indicated by white arrows. Scale bar is 10 μm in SEM and 2 μm in TEM.

3.2 Components of DPC The expression of collagen I, IV and laminin in DPC were consistent with NPC (Fig. 2A-C) by IF staining. The quantification of remaining collagen and glycosaminoglycan (GAG) confirmed that they were unaltered during decellularization (Fig. 2E, F). Fluorescence staining showed that nucleic acid and α-Gal antigens were completely removed (Fig. 2D). Additionally, DNA and α-Gal antigen were quantitated and significantly reduced in DPC (Fig. 2G, H). The nucleases introduced in the preparation process have also been effectively removed (Table S2).

13

Fig.2. Component comparison of NPC and DPC. (A-D) Fluorescence staining of collagen I, collagen IV, laminin and α-Gal in NPC and DPC. Nuclei were stained by DAPI. Scale bar, 50 μm. (E-H) Comparison of NPC and DPC regarding collagen content (E), GAG content (F), DNA content (G) and α-Gal content (H). **** P < 0.0001 (n = 3, unpaired Student’s -test).

3.3 Cytocompatibility of DPC in vitro To determine the potential cytotoxicity of DPC, live-dead scanning was performed to evaluate the cells status on the surface of DPC or collagen I. The cells cultured on the DPC were mostly alive and showed the morphology of epithelial cells (Fig. 3A). Our results showed that the survival rate of HCEC on the DPC was 97 ± 1%, and the survival rate on collagen I was 97.3 ± 0.6% (Fig. 14

3C). The potency of DPC as a scaffold for conjunctival reconstruction was evaluated in vitro by histological study. Upon two weeks of culture, conjunctiva-derived cells formed an epithelial layer on the DPC, which was composed of 1-2 layers of epithelial cells (Fig. 3D). The recellularized DPC displayed a similar structure to the normal epithelium of rabbit bulbar conjunctiva (Fig. 3E). Conjunctival stromal cells were also observed (Fig. S1). The proliferation of epithelial cells was detected in the conditioned culture medium from DPCs. The medium conveyed no obvious proliferation inhibition (Fig. 3F).

Fig.3. Cytocompatibility of DPC in vitro. The live-dead staining imagen of HCEC cultured on DPC (A) and cultured on slides coated by collagen I (B). The dead cells were stained by a red fluorescent dye and living cells were stained with a green fluorescent dye. Scale bar, 50 μm. (C) The ratio of living cells to total cells under different culture conditions. (D) Epithelial cells of rabbit conjunctiva anchored to DPC formed epithelioid structure composed of 1-2 layers of epithelial cells upon 14 days culture. (E) The H&E staining of native bulbar conjunctiva of rabbit. The overview diagram is on display in bottom left corner. Scale bar, 25 μm. (F) Extracts derived from DPC did not show a proliferation inhibition on HCECs compared to normal media control.

3.4 Mechanical property Tensile

testing

experiments

were

performed

to

evaluate

the

biomechanical properties of DPC. Extensibility was represented by peak elongation. The average peak elongation for DPC (109 ± 24%) was almost 15

double that for AM (49 ± 9%), superior to that for NPC (78 ± 16%) and native human conjunctiva (NHC, 87 ± 16%, Fig. 4A). The young's modulus (Ymod) described the elasticity when the material was stretched. The Ymod of DPC was significantly lower than that of AM and higher than that of NPC and no significant difference with that of NHC (Fig. 4B). The dynamic variation of stress-strain was recorded (Fig. 4C). DPC maintained the longer time before rupture under the action of external forces and showed greater stress.

Fig.4. Biomechanical characters of DPC. (A) The peak elongation of DPC was compared with that of AM, NPC and NHC. (B) The Young modulus (Ymod) of DPC was compared with that of AM, NPC and NHC. (C) The dynamic changes of stress were showed in DPC, AM, NHC and NPC. N = 6. Data represented as mean value ±SD. *** P < 0.001, **** P < 0.0001. NS, not significant.

3.5 Evaluation of DPC in vivo 3.5.1 Reparation of rabbit surgical trauma model Rabbit surgical trauma models were repaired by grafting with DPC or AM (Fig. 5A). Our results showed that all the DPC grafts were stable, whereas folding or exfoliating occurred in most of AM graft at the seventh day postoperatively (Fig. 5B, Fig. S2). Two weeks later, DPC and stroma of the recipient rabbit conjunctiva showed a good integration without obvious rejection. Four weeks after surgery, the conjunctival defect was completely repaired, with no significant distinction from the surrounding normal conjunctiva in the DPC group (Fig. 5B). Re-epithelialization was evaluated by fluorescein sodium staining, and epithelial defects were stained yellow (Fig. 5B). Conjunctival defect area was quantitated (Fig. 5C). The DPC and AM 16

groups showed a higher efficiency of epithelial repair than the control group. In order to evaluate the immune stimulation of the grafts, the number of lymphocytes was counted (Fig. 5D). DPC has maintained low antigenic stimulation throughout. The native porcine conjunctiva was transplanted and caused intense xenograft rejection (Fig. S3). Postoperative hyperemia was graded, and the defect treated by DPC was kept at a low score relatively (Fig. 5E). The number of goblet cells was quantified (Fig. 5F). There were relatively more goblet cells in the DPC group at different time points after surgery. The repair of surgical trauma models was evaluated by histological staining. The results showed that DPC has integrated with rabbit conjunctiva at the fourth day postoperatively. Stratified epithelial layers with distinct basal cells were formed on the surface of DPC (Fig. 5G). The AM had a gap with the recipient, which caused epithelial ingrowth in some areas (Fig. 5K, black arrow indicated). At the ninth day postoperatively, the expression of collagen I from swine was weaker, and DPC was replaced by rabbit conjunctival matrix gradually at the edge of the graft (Fig. 5H). The AM melted and many lacunas formed beneath the epithelium (Fig. 5L). Strong PAS-positive staining, representing conjunctival goblet cells, was found in central area of defective at 14th day after surgery. The distribution of goblet cells was denser in DPC (Fig. 5I), compared with spontaneous healing (Fig. 5Q). At 28 days after surgery, conjunctival defects were substantially repaired by histological observation in all three groups (Fig. 5J, N, I). Compared with the control group, the DPC and AM groups showed more similar histologic morphology.

17

Fig.5. Conjunctival reconstruction using DPC in a rabbit surgical trauma model. (A) The working strategy for repairing surgical trauma. (B) Gross images showing the progress of conjunctival repair at various time points. The control was ungrafted. The epithelial defects were stained with fluorescein sodium. (C) Histogram showing percentage variation of residual epithelial defect area during process of repair. (D) A line graph showing the variation in the number of infiltrative lymphocytes. (E) The grading of bulbar hyperaemia was compared in the three groups. (F) The quantification of goblet cells at different time points after operation (C-F) N = 4-8, Data represented as mean value ± SD. * p < 0.05, ** p < 0.01. Immunohistochemical (IHC) staining of collagen I (species-specific for swine or human) showing the status of grafts on the fourth day (G, K) and the ninth day (H, L) postoperatively. The epithelial ingrowth is indicated by black arrow in figure K. The boundary of DPC is indicated by imaginary line in figure H. (O, P) H&E staining revealed a natural repair of the conjunctival defect. PAS staining showed the distribution of goblet cells in the repaired epithelium using DPC (I), AM (M) and control (Q) respectively. Goblet 18

cells are shown by asterisk. The direction of the defect-center point is indicated by the black arrow. The H&E staining results of DPC, AM and control group at 28 days after surgery were shown in figure J, figure N and figure I, respectively. Scale bar, 50 μm.

3.5.2 Reparation of rabbit symblepharon model To evaluate the performance of DPC in repairing severe ocular surface injuries, the severe symblepharon model was constructed in rabbit. And, the representative models were shown in Fig. 6A (left, grade III c 1+) and Fig. 6B (left, grade II c 1+). All bulbar conjunctiva and most of the forniceal conjunctiva were replaced by DPC or AM (Fig. 6A, B showed only a portion of the transplant area due to the limitations of eye position). Two weeks after transplantation of DPC, the symptoms of symblepharon and corneal complications were controlled (Fig. 6C, Fig. S4). After one month, the conjunctival hyperemia and corneal neovascularization were subsiding, and corneal transparency recuperated gradually. Ocular surface symptoms improved further and recurrence of symblepharon was not observed at two months of implantation. In AM groups, the symblepharon was reoccurred at two weeks after surgery (Fig. 6D). Aggressive fibrovascular growth was observed (Fig. S5) and extended over the pupillary zone within two months (Fig. 6D). Percentage of postoperative fornix depth relative to baseline were compared between DPC and AM groups as a quantitative indication of recurrence of symblepharon (Fig. 6E). In DPC treatments, the forniceal depth exhibited a significant increase of 220% at one month and 208% at two months when comparing baseline. By contrast, in the AM groups, fornix shortening returned to preoperative level within two months after implantation.

19

Fig.6. DPC and AM transplantation in rabbit symblepharon models. (A, B) Gross images showing comparison of ocular surface reconstruction before and after surgery. (C, D) Gross images showing the process of postoperative repair. (E) The percentage of fornix depth relative to baseline at various time points in DPC and AM groups. ** p < 0.01.

Regenerated epithelium was evaluated using Alcian Blue staining (Fig. 7A). Several layers of epithelioid cells formed and few goblet cells randomly distributed at the first month postoperatively. Two months after surgery, the reconstructed epithelium was full-fledged resembling normal conjunctival epithelium. The number of goblet cells significantly increased, from 2.4 ± 0.9 cells pre mm to 27 ± 4 cells pre mm (Fig. 7D). Fibrillar arrangements in reconstructed conjunctival stroma were visualized by MT staining (Fig. 7B) and electron microscopy (Fig. 7C, Fig. S6). Serried collagen fibers were observed 20

in the substantia propria at the first month postoperatively. After two months, the collagen fibrils became loose, interfibrillar average distance increased (Fig.7E), and the fibril diameter decreased (Fig.7F), which was similar to that in native conjunctiva. The expression of cytokeratin (CK) was examined by IHC in regenerated epithelium (Fig. 7H). Pan-CK expression was intensively detected. In the early stage, the whole layers of epithelium on DPC exhibited the predominant expression of CK14. However, upon two months, the expression of CK14 was absent in the outer layer of the epithelium but only detected in the cells located in basal layer. Strong expression of CK13 located in the outer layer of the epithelium in perilimbal conjunctiva. The expression pattern of CK in regenerated epithelia was similar to that of native rabbit conjunctiva.

21

Fig.7. The evaluation to DPC graft performance in rabbit symblepharon models. (A) Alcian blue staining showing the distribution of mucopolysaccharides. Goblet cells are shown by asterisk. Scale bar, 25 μm. NRC: native rabbit conjunctiva. (B) MT staining for the visualization of collagen fibrils. Scale bar, 100 μm. (C) TEM images show a cross section of collagen fibrils. Scale bar, 200nm. (D) The number of goblet cells was counted at one and two-month after surgery. (E-G) The conjunctival matrix fibril parameters (diameter, interfibrillar distance, and number) at different time points after transplantation were compared with normal rabbit conjunctiva. * p < 0.05, ** p < 0.01, *** p < 0.001. (H) The 22

expression of CKs in reconstructed conjunctiva and in native rabbit conjunctiva. Scale bar, 25 μm.

3.5.3 Application in clinical cases DPC transplants were performed in three patients with pterygium. After 2 weeks postoperatively, DPC grafts were integrated with autologous conjunctiva and had no apparent melt and hyperemia. The area covered by the DPC was completely epithelialized with staining of sodium fluorescein within two weeks (Fig. 8A). After two months, no significant sign of recurrence was observed (Fig. 8A). In the symblepharon case, the patient’s left eye had already undergone AM transplantations twice, and both procedures were followed by recurrence. A stalk-like adhesion from the medial lower eyelid to the limbus cornea had impaired ocular movement severely (Fig. 8B). After DPC transplantation, a deep inferior fornix formed (Fig. 8B). The DPC covered surfaces showed complete epithelialization at 2 weeks. At one month of implantation, proliferating tissues and fornix shortening were not observed.

23

Fig.8. DPC treatment for pterygium and symblepharon. (A) Slit-lamp photographs and fluorescein staining images showing DPC application in three patients with pterygium. (B) The results of symblepharon repair was showed by slit-lamp photographs at various time points within one month. Fluorescein sodium staining photographs reflecting the recovery of conjunctival epithelium.

4. Discussion

Conjunctiva plays an indispensable role in maintaining the function of ocular surface. Conjunctival damage and scarring can result in ocular discomfort, corneal opacity and even blindness [30, 31]. To prevent severe morbidities, it is crucial to apply suitable conjunctival substitutes. The ideal conjunctival substitute should meet the following criteria. First, it should have a flexible matrix with long-term stability. Second, it should be well tolerated without causing inflammation or stimulating rejection. Third, it can promote the formation of epithelial layers, especially in extensive conjunctival defects. Finally, the formed epithelium should contain goblet cells, which is essential for the tear film and ocular surface stability [32, 33]. The decellularization of xenogeneic tissues or organs has been investigated as a promising strategy to fabricate functional scaffolds for transplantation [24]. Here, we have developed a conjunctival decellularization protocol. The protocol effectively removed cellular components; no cellular structure and only a small amount of DNA and α-Gal antigens was found in the DPC. DNA and α-Gal antigens are considered major inflammatory stimulators for acellular material in xenotransplantation [34, 35]. The complete removal of cell fragments is crucial because residual cellular components in the DPC would lead to an immunological response in vivo [36]. Moreover, nucleases as a potential risk factor, which may impair regeneration and invoke immune response, were removed in the protocol [23]. During the preparation of DPC, another important consideration is to keep intact a reservation of matrix components and unique structure, because 24

tissue development and wound healing depend on the environmental niche [37, 38]. In decellularized conjunctiva that have been reported, long processing periods (more than 24 hrs) were usually used [39, 40], and the native ECM may tend to be damaged [41]. After trying a series of decellularization protocols, we eventually developed a hybrid method and have proved its appropriateness. The surfactant incubation was shortened with high hydrostatic pressure and low frequency ultrasonic. The results from scanning electron microscopy and histological staining presented the undamaged architecture and retention of active ingredients in ECM. To evaluate the biomechanical properties of DPC, mechanical testing was performed. Our study showed that DPC has higher extensibility and tensile strength than those of NPC. It may be caused by the concentration of the GAG in DPC due to decellularization. GAG plays a hydrophilic role in ECM, which is linked to polypeptide core to connect two collagen fibrils and provide the intermolecular force [42, 43]. In addition, a low dose of irradiation may also increase the mechanical strength of DPC [44]. This improved mechanical strength may be critical for the functional implementation of DPC, such as to tolerate surgical suture force, coordinate eye movement and avoid fold of graft [45, 46]. In AM transplants, epithelial ingrowth was observed in the early stage of wound healing (4 days postoperatively), which may be caused by the fold of graft under the movement of eyeball and the continuously close-and-open of eyelids all day. It is well-known that epithelial ingrowth has potential risks, such as graft melting and formation of fibrous scar [47, 48]. Up to now, AM has been the most widely used biological substrate for conjunctival

reconstruction

due

to

its

inherent

ability

to

promote

epithelialization [49, 50]. We prove that AM transplantation can repair the conjunctival defect more quickly than natural repair. However, the availability, cost and standardization in the preparation of AM still remain issues [9, 10, 51-53]. AM can degrade quickly in inflammatory environment leading to a 25

decreased chance of epithelialization [40, 54]. In this study, following the degrade of AM, a short-term acute inflammatory response was observed. The reason may be related with the concentrated release of heterogenetic antigens from human AM. Comparatively, DPC transplants showed complete integration into the recipient conjunctiva with a long-term stability and acceptable immunogenicity. The superiorities of DPC may depend on the effective removal of major xenoantigens (DNA and α-gal) and the intact reservation of conjunctiva-specific intricate structures. This study confirms the ability of DPC to promote epithelial repair. The conjunctival ECM protein composition, its specific surface structure and the growth factor composition in DPC may contribute to this effect. Numerous studies have shown that the unique ECM of tissue promotes cellular migration and proliferation [55, 56]. Moreover, the rapid epithelialization may be also related with the readily available conjunctival matrix provided by DPC. It has been reported that re-epithelialization follows fibroblast-guided conjunctival matrix remodeling during the wound healing process [19]. The recovery of goblet cells is an important indicator to the functional rehabilitation of conjunctiva [32, 57]. As previously reported, the longest observation record was 10 days for repairing conjunctival defect in vivo by decellularized conjunctival material, and the center of grafts did not contain goblet cells [40]. In our study, the goblet cells were found in the center of DPC at 2nd week after transplantation. It demonstrated that conjunctival function was restored by DPC. Symblepharon is a severe conjunctival injury disease, which is usually accompanied by chronic inflammation and the loss of limbal stem cells, resulting in serious ocular surface dysfunction, including abnormal corneal surface, fornix shortening, corneal vascularization, and poor epithelial integrity [58, 59]. Although many surgical approaches have been developed, severe symblepharon presents a significant surgical problem [59, 60]. In our research， 26

despite

of

AM

transplantation

and

anti-inflammatory

medications,

symblepharon was proved refractory. Poor prognosis may be caused by the barrier function destruction of limbal area [61, 62]. Moreover, some of cytokines and growth factors inducing scar tissue formation in AM may also be potential causes [63]. In contrast, the transplanted DPC could decrease conjunctival inflammation and eliminate symblepharon as well as improve corneal surface conditions. The conjunctival sac was re-established and the goblet cells were re-populated. The DPC transplant may function as a steady anatomical barrier to restrict fibrous tissue proliferation. In the repair of severe symblepharon models, the fibril gaps appeared more porous two months than one month after surgery. The transformation of matrix fibers may promote the reconstruction of conjunctiva because it accelerated the diffusion of nutrition and direct oxygen delivery [64]. Several studies have corroborated that the ideal decellularized matrix guide site-appropriate cell attachment and differentiation [18, 65, 66]. In our study, the conjunctiva-derived single cells were guided by DPC to the ‘‘right” position, and the recellularization of DPC with an epithelial structure was constructed in vitro. Furthermore, the similar CKs expression patterns were detected in regenerated epithelium in vivo. CK14, as a marker of undifferentiated cells in stratified epithelia, was expressed in the basal cells of conjunctiva. CK13, as a typical marker of mature nonkeratinizing cells, was detected in superficial layers in perilimbal conjunctival epithelium [67, 68]. Interestingly, a dynamic differentiation was showed in the process of conjunctival reconstruction by DPC. At the early stage, full-layer CK14 expression indicated that the conjunctival epithelium was immature. Thus, we proved that DPC can promote cell differentiation in the right direction. The preliminary attempts of DPC application in clinic (three cases of pterygium and one case of symblepharon) were successful. In all cases, DPC showed favorable histocompatibility, promoting re-epithelialization with no 27

obvious inflammatory response and relapse. Although our study has shown the efficiency of DPC in the treatment of pterygium and symblepharon, there are still some limitations. Firstly, our study was limited by the sample size. Second, patients were only followed for two months. Overall, in future studies, we planned to enroll a larger number of cases with longer term follow-up, during which recurrence rate, secretory capacity of goblet cells and potential scar formation will be further evaluated.

5. Conclusions

In this study, we demonstrated an optimized protocol for preparation of decellularized porcine conjunctival matrix. DPC had not only played an active role in reconstruction of conjunctiva, alleviating complications and controlling recurrence in ocular surface injury rabbit model, but also revealed rapid re-epithelization potential in clinic. The prospect of DPC as a suitable material for conjunctival reconstruction in complex clinical situations is promising.

Acknowledgments

We thank all of patients whose generosity made this research possible. We wish to acknowledge Zhen Shi for linguistic advice and Bio-Decell (Beijing) Biology Science ＆ Technology Co.,Ltd. for support of experimental materials. Funding: This work was supported by Taishan Scholar Program [ grant numbers 20161059]; Key Project of National Natural Science Foundation of China [ grant number 81530027].

28

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

29

References [1] I.K. Gipson, The ocular surface: the challenge to enable and protect vision: the Friedenwald lecture, Invest Ophthalmol Vis Sci. 48 (2007) 4390; 4391-4398. [2] S.C. Tseng, M.A. Di Pascuale, D.T. Liu, Y.Y. Gao, A. Baradaran-Rafii, Intraoperative mitomycin C and amniotic membrane transplantation for fornix reconstruction in severe cicatricial ocular surface diseases, Ophthalmology. 112 (2005) 896-903. [3] E.M. Espana, M. Di Pascuale, M. Grueterich, A. Solomon, S.C. Tseng, Keratolimbal allograft in corneal reconstruction, Eye. 18 (2004) 406-417. [4] R. Arora, D. Mehta, V. Jain, Amniotic membrane transplantation in acute chemical burns, Eye. 19 (2005) 273-278. [5] J. Shimazaki, H.Y. Yang, K. Tsubota, Amniotic membrane transplantation for ocular surface reconstruction in patients with chemical and thermal burns, Ophthalmology. 104 (1997) 2068-2076. [6] N. Tananuvat, T. Martin, The results of amniotic membrane transplantation for primary pterygium compared with conjunctival autograft, Cornea. 23 (2004) 458-463. [7] C.A. Akle, M. Adinolfi, K.I. Welsh, S. Leibowitz, I. McColl, Immunogenicity of human amniotic epithelial cells after transplantation into volunteers, Lancet. 2 (1981) 1003-1005. [8] M. Fernandes, M.S. Sridhar, V.S. Sangwan, G.N. Rao, Amniotic membrane transplantation for ocular surface reconstruction, Cornea. 24 (2005) 643-653. [9] A. Hopkinson, R.S. McIntosh, P.J. Tighe, D.K. James, H.S. Dua, Amniotic membrane for ocular surface reconstruction: donor variations and the effect of handling on TGF-beta content, Invest Ophthalmol Vis Sci. 47 (2006) 4316-4322. [10] H.W. Henderson, J.R. Collin, Mucous membrane grafting, Dev Ophthalmol. 41 (2008) 230-242. [11] J. Witt, M. Borrelli, S. Mertsch, G. Geerling, K. Spaniol, S. Schrader, Evaluation of plastic-compressed collagen for conjunctival repair in a rabbit model, Tissue Eng Part A. 25 (2019) 1084-1095. [12] C. Petsch, U. Schlotzer-Schrehardt, E. Meyer-Blazejewska, M. Frey, F.E. Kruse, B.O. Bachmann, Novel collagen membranes for the reconstruction of the corneal surface, Tissue Eng Part A. 20 (2014) 2378-2389. [13] P. Vinciguerra, E. Albe, P. Rosetta, E. Di Iorio, G. Pellegrini, Custom phototherapeutic keratectomy and autologous fibrin-cultured limbal stem cell autografting: a combined approach, J Refract Surg. 24 (2008) 323-324. [14] S.Y. Lee, J.H. Oh, J.C. Kim, Y.H. Kim, S.H. Kim, J.W. Choi, In vivo conjunctival reconstruction using modified PLGA grafts for decreased scar formation and contraction, Biomaterials. 24 (2003) 5049-5059. [15] N. Scuderi, C. Alfano, G. Paolini, C. Marchese, G. Scuderi, Transplantation of autologous cultivated conjunctival epithelium for the restoration of defects in the ocular surface, Scand J Plast Reconstr Surg Hand Surg. 36 (2002) 340-348. [16] R. Mancino, F. Aiello, S. Ceccarelli, C. Marchese, C. Varesi, C. Nucci, L. Cerulli, Autologous conjunctival epithelium transplantation and scleral patch graft for postlensectomy 30

wound leakage in Marfan syndrome, Eur J Ophthalmol. 22 (2012) 830-833. [17] D.A. Taylor, L.C. Sampaio, Z. Ferdous, A.S. Gobin, L.J. Taite, Decellularized matrices in regenerative medicine, Acta Biomater. 74 (2018) 74-89. [18] F. Gattazzo, A. Urciuolo, P. Bonaldo, Extracellular matrix: a dynamic microenvironment for stem cell niche, Biochim Biophys Acta. 1840 (2014) 2506-2519. [19] P. Bainbridge, Wound healing and the role of fibroblasts, J Wound Care. 22 (2013) 407-408, 410-412. [20] M. Cosentino, A. Kanashiro, A. Vives, J. Sanchez, M.F. Peraza, D. Moreno, J. Perona, V. De Marco, E. Ruiz-Castane, J. Sarquella, Surgical treatment of Peyronie's disease with small intestinal submucosa graft patch, Int J Impot Res. 28 (2016) 106-109. [21] Y.L. Loo, S. Haider, The use of porcine acellular dermal matrix in single-stage, implant-based immediate breast reconstruction: a 2-center retrospective outcome study, Plast Reconstr Surg Glob Open. 6 (2018) e1895. [22] L. D'Ambra, S. Berti, C. Feleppa, P. Magistrelli, P. Bonfante, E. Falco, Use of bovine pericardium graft for abdominal wall reconstruction in contaminated fields, World J Gastrointest Surg. 4 (2012) 171-176. [23] P.M. Crapo, T.W. Gilbert, S.F. Badylak, An overview of tissue and whole organ decellularization processes, Biomaterials. 32 (2011) 3233-3243. [24] A. Porzionato, E. Stocco, S. Barbon, F. Grandi, V. Macchi, R. De Caro, Tissue-engineered grafts from human decellularized extracellular matrices: a systematic review and future perspectives, Int J Mol Sci. 19 (2018). [25] T. Zhang, G.H. Yam, A.K. Riau, R. Poh, J.C. Allen, G.S. Peh, R.W. Beuerman, D.T. Tan, J.S. Mehta, The effect of amniotic membrane de-epithelialization method on its biological properties and ability to promote limbal epithelial cell culture, Investigative ophthalmology & visual science 54(4) (2013) 3072-3081. [26] S.S. Ferreira, F.P.B. Nunes, F.B. Casagrande, J.O. Martins, Insulin modulates cytokine release, collagen and mucus secretion in lung remodeling of allergic diabetic mice, Front Immunol. 8 (2017) 633. [27] L. Zhao, L. Huang, S. Yu, J. Zheng, H. Wang, Y. Zhang, Decellularized tongue tissue as an in vitro model for studying tongue cancer and tongue regeneration, Acta Biomater. 58 (2017) 122-135. [28] L. Tian, M. Qu, Y. Wang, H. Duan, G. Di, L. Xie, Q. Zhou, Inductive differentiation of conjunctival goblet cells by gamma-secretase inhibitor and construction of recombinant conjunctival epithelium, Exp Eye Res. 123 (2014) 37-42. [29] A. Kheirkhah, G. Blanco, V. Casas, Y. Hayashida, V.K. Raju, S.C. Tseng, Surgical strategies for fornix reconstruction based on symblepharon severity, Am J Ophthalmol. 146 (2008) 266-275. [30] M.S. Shapiro, J. Friend, R.A. Thoft, Corneal re-epithelialization from the conjunctiva, Invest Ophthalmol Vis Sci. 21 (1981) 135-142. [31] H.S. Dua, J.V. Forrester, The corneoscleral limbus in human corneal epithelial wound healing, Am J Ophthalmol. 110 (1990) 646-656. [32] S. Schrader, M. Notara, M. Beaconsfield, S.J. Tuft, J.T. Daniels, G. Geerling, Tissue engineering for conjunctival reconstruction: established methods and future outlooks, Curr Eye Res. 34 (2009) 913-924. 31

[33] I.V. Yannas, Emerging rules for inducing organ regeneration, Biomaterials. 34 (2013) 321-330. [34] U. Galili, The alpha-gal epitope and the anti-Gal antibody in xenotransplantation and in cancer immunotherapy, Immunol Cell Biol. 83 (2005) 674-686. [35] K.H. Hussein, K.M. Park, K.S. Kang, H.M. Woo, Biocompatibility evaluation of tissue-engineered decellularized scaffolds for biomedical application, Mater Sci Eng C Mater Biol Appl. 67 (2016) 766-778. [36] B.N. Brown, J.E. Valentin, A.M. Stewart-Akers, G.P. McCabe, S.F. Badylak, Macrophage phenotype and remodeling outcomes in response to biologic scaffolds with and without a cellular component, Biomaterials. 30 (2009) 1482-1491. [37] J. De Waele, K. Reekmans, J. Daans, H. Goossens, Z. Berneman, P. Ponsaerts, 3D culture of murine neural stem cells on decellularized mouse brain sections, Biomaterials. 41 (2015) 122-131. [38] M. Walraven, B. Hinz, Therapeutic approaches to control tissue repair and fibrosis: Extracellular matrix as a game changer, Matrix Biol. 71-72 (2018) 205-224. [39] S. Kasbekar, S.B. Kaye, R.L. Williams, R.M.K. Stewart, S. Leow-Dyke, P. Rooney, Development of decellularized conjunctiva as a substrate for the ex vivo expansion of conjunctival epithelium, J Tissue Eng Regen Med. 12 (2018) e973-e982. [40] J. Witt, S. Mertsch, M. Borrelli, J. Dietrich, G. Geerling, S. Schrader, K. Spaniol, Decellularised conjunctiva for ocular surface reconstruction, Acta Biomater. 67 (2018) 259-269. [41] A. Gilpin, Y. Yang, Decellularization strategies for regenerative medicine: from processing techniques to applications, Biomed Res Int. 2017 (2017) 9831534. [42] J. Halper, M. Kjaer, Basic components of connective tissues and extracellular matrix: elastin, fibrillin, fibulins, fibrinogen, fibronectin, laminin, tenascins and thrombospondins, Adv Exp Med Biol. 802 (2014) 31-47. [43] Y. Bi, P. Patra, M. Faezipour, Structure of collagen-glycosaminoglycan matrix and the influence to its integrity and stability, Conf Proc IEEE Eng Med Biol Soc. 2014 (2014) 3949-3952. [44] W.Q. Sun, P. Leung, Calorimetric study of extracellular tissue matrix degradation and instability after gamma irradiation, Acta Biomater. 4 (2008) 817-826. [45] D.M. Adelman, J.C. Selber, C.E. Butler, Bovine versus porcine acellular dermal matrix: a comparison of mechanical properties, Plast Reconstr Surg Glob Open. 2 (2014) e155. [46] G.A. Monteiro, N.L. Rodriguez, A.I. Delossantos, C.T. Wagner, Short-term in vivo biological and mechanical remodeling of porcine acellular dermal matrices, J Tissue Eng. 4 (2013). [47] T.R. Kramer, V. Chuckpaiwong, D.G. Dawson, N. L'Hernault, H.E. Grossniklaus, H.F. Edelhauser, Pathologic findings in postmortem corneas after successful laser in situ keratomileusis, Cornea. 24 (2005) 92-102. [48] D.R. Hardten, M.M. Fahmy, G.K. Vora, J.P. Berdahl, T. Kim, Fibrin adhesive in conjunction with epithelial ingrowth removal after laser in situ keratomileusis: long-term results, J Cataract Refract Surg. 41 (2015) 1400-1405. [49] S. Barabino, M. Rolando, G. Bentivoglio, C. Mingari, S. Zanardi, R. Bellomo, G. Calabria, Role of amniotic membrane transplantation for conjunctival reconstruction in ocular-cicatricial 32

pemphigoid, Ophthalmology. 110 (2003) 474-480. [50] A. Solomon, E.M. Espana, S.C. Tseng, Amniotic membrane transplantation for reconstruction of the conjunctival fornices, Ophthalmology. 110 (2003) 93-100. [51] I. Rahman, D.G. Said, V.S. Maharajan, H.S. Dua, Amniotic membrane in ophthalmology: indications and limitations, Eye. 23 (2009) 1954-1961. [52] Y. Feng, M. Borrelli, S. Reichl, S. Schrader, G. Geerling, Review of alternative carrier materials for ocular surface reconstruction, Curr Eye Res. 39 (2014) 541-552. [53] C.J. Connon, J. Doutch, B. Chen, A. Hopkinson, J.S. Mehta, T. Nakamura, S. Kinoshita, K.M. Meek, The variation in transparency of amniotic membrane used in ocular surface regeneration, Br J Ophthalmol. 94 (2010) 1057-1061. [54] H. Zhao, M. Qu, Y. Wang, Z. Wang, W. Shi, Xenogeneic acellular conjunctiva matrix as a scaffold of tissue-engineered corneal epithelium, PLoS One. 9 (2014) e111846. [55] E. Vorotnikova, D. McIntosh, A. Dewilde, J. Zhang, J.E. Reing, L. Zhang, K. Cordero, K. Bedelbaeva, D. Gourevitch, E. Heber-Katz, S.F. Badylak, S.J. Braunhut, Extracellular matrix-derived products modulate endothelial and progenitor cell migration and proliferation in vitro and stimulate regenerative healing in vivo, Matrix Biol. 29 (2010) 690-700. [56] R.A. Allen, L.M. Seltz, H. Jiang, R.T. Kasick, T.L. Sellaro, S.F. Badylak, J.B. Ogilvie, Adrenal extracellular matrix scaffolds support adrenocortical cell proliferation and function in vitro, Tissue Eng Part A. 16 (2010) 3363-3374. [57] I.K. Gipson, Goblet cells of the conjunctiva: A review of recent findings, Prog Retin Eye Res. 54 (2016) 49-63. [58] M. Farid, N. Lee, Ocular surface reconstruction with keratolimbal allograft for the treatment of severe or recurrent symblepharon, Cornea. 34 (2015) 632-636. [59] J. Cheng, H. Zhai, J. Wang, H. Duan, Q. Zhou, Long-term outcome of allogeneic cultivated limbal epithelial transplantation for symblepharon caused by severe ocular burns, BMC Ophthalmol. 17 (2017) 8. [60] T. Li, Y. Shao, Q. Lin, D. Zhang, Reversed skin graft combining with lip mucosa transplantation in treating recurrent severe symblepharon: A case report, Medicine (Baltimore). 97 (2018) e12168. [61] Q. Le, J. Xu, S.X. Deng, The diagnosis of limbal stem cell deficiency, Ocul Surf. 16 (2018) 58-69. [62] S. Ahmad, Concise review: limbal stem cell deficiency, dysfunction, and distress, Stem Cells Transl Med. 1 (2012) 110-115. [63] H.S. Dua, V.S. Maharajan, A. Hopkinson, Controversies and limitations of amniotic membrane in ophthalmic surgery, Cornea External Eye Dis. (2006) 21–33. [64] D. Huang, B. Xu, X. Yang, B. Xu, J. Zhao, Conjunctival structural and functional reconstruction using acellular bovine pericardium graft (Normal GEN(R)) in rabbits, Graefes Arch Clin Exp Ophthalmol. 254 (2016) 773-783. [65] J. Cortiella, J. Niles, A. Cantu, A. Brettler, A. Pham, G. Vargas, S. Winston, J. Wang, S. Walls, J.E. Nichols, Influence of acellular natural lung matrix on murine embryonic stem cell differentiation and tissue formation, Tissue Eng Part A. 16 (2010) 2565-2580. [66] N.C. Cheng, B.T. Estes, H.A. Awad, F. Guilak, Chondrogenic differentiation of adipose-derived adult stem cells by a porous scaffold derived from native articular cartilage extracellular matrix, Tissue Eng Part A. 15 (2009) 231-241. 33

[67] S. Merjava, A. Neuwirth, M. Tanzerova, K. Jirsova, The spectrum of cytokeratins expressed in the adult human cornea, limbus and perilimbal conjunctiva, Histol Histopathol. 26 (2011) 323-331. [68] A. Matsuda, Y. Tagawa, H. Matsuda, Cytokeratin and proliferative cell nuclear antigen expression in superior limbic keratoconjunctivitis, Curr Eye Res. 15 (1996) 1033-1038.

34

Graphical abstract

35