Heterogeneous-structural bilamellar bionic scaffold for integrated regeneration of tarso-conjunctival composite tissue

Xing Huang; Lin Lu; Yi Ding; Mengling Chang; Xiao Liang; Zhaoqi Yuan; Feixue Ding; Peiyi Li; Rui Jin; Xusong Luo

doi:10.1186/s12967-025-07655-0

. 2026 Feb 19;24:420. doi: 10.1186/s12967-025-07655-0

Heterogeneous-structural bilamellar bionic scaffold for integrated regeneration of tarso-conjunctival composite tissue

Xing Huang ^1,^2,^3,^#, Lin Lu ^1,^#, Yi Ding ^1,^#, Mengling Chang ¹, Xiao Liang ¹, Zhaoqi Yuan ¹, Feixue Ding ¹, Peiyi Li ¹, Rui Jin ^1,^✉, Xusong Luo ^1,^3,^✉

PMCID: PMC13020167 PMID: 41715185

Abstract

Background

Defects in the posterior lamella of the eyelid are prevalent in the clinical settings, which comprises the tarsal plate and conjunctiva and has limited regenerative capacity. Currently, homogenous scaffolds are used for repair, but they fail to recapitulate the natural bilayer structure and reepithelialization. In this study, we designed a bilamellar scaffold (bGA-ADM) with a heterogeneous structure to repair such defects and promote the integrated regeneration of tarso-conjunctival composite tissue.

Methods

bGA-ADM was prepared by homogenization, freezing, freeze-drying, heat-vacuum crosslinking, and glutaraldehyde crosslinking, and its physiochemical properties were subsequently characterized. Fibroblasts, endothelial cells, and meibomian gland epithelial cells were seeded in the dermal layer, whereas mucosal epithelial cells were seeded onto the basement membrane layer to validate cytocompatibility and structural heterogeneity. Rabbit ear and eyelid implantation models were used to confirm in vivo biocompatibility and vascularization of bGA-ADM, and an eyelid posterior lamellar defect model was employed to evaluate its repairing efficacy. RNA sequencing was further performed to investigate the potential of bGA-ADM for meibomian gland construction.

Results

bGA-ADM exhibited biomimetic mechanical properties that provided adequate structural support, and allowed adaptive deformation. Consisting of a patterned basement membrane layer and a spongy dermal layer, it supported the stratified growth of fibroblasts, endothelial cells, meibomian gland epithelial cells, and mucosal epithelial cells in vitro. It also enhanced cellular infiltration and vascularization, thereby accelerating tissue regeneration in both ear and eyelid implantation models. When applied to repair the posterior lamellar defects in situ, bGA-ADM adaptively regenerated tissue closely resembling the native tissues, characterized by matrix deposition and conjunctival resurfacing. Furthermore, it fostered a microenvironment that suppressed inflammation in meibomian gland epithelial cells, contributing to subsequent regeneration of the meibomian gland.

Conclusion

bGA-ADM presents a promising candidate for the repair of posterior lamellar eyelid defects, and holds substantial potential for clinical translation.

Graphical Abstract

Supplementary information

The online version contains supplementary material available at 10.1186/s12967-025-07655-0.

Keywords: Tarsal-conjunctival defects, Bilamellar scaffold, Heterogeneous structure, Integrated regeneration, Biomimetic mechanics

Introduction

Defects in the posterior lamella of the eyelid are common in clinical practice, caused by trauma, inflammation, ocular tumors, or congenital diseases. The posterior lamella of the eyelid, a functional bilayer tissue composed of the tarsal plate and conjunctiva, plays a crucial role in supporting the eyelid, protecting the eyeball, and facilitating eye movement [1, 2]. Long-term defects in this region may lead to complications such as eyelid retraction, keratitis, conjunctivitis, and even visual dysfunction [3]. Current clinical strategies for repairing these defects mainly include surgical closure and reconstruction using substitutes. Surgical closure typically involves direct closure for small defects, and tarso-conjunctival grafting for medium-to-large defects. For large defects, reconstruction becomes complex and is associated with a high risk of severe corneal irritation, lagophthalmos on the intact side, secondary reconstructive procedures, and amblyopia [4].

Reconstruction of the posterior lamella of the eyelid using substitutes can maximally preserve the healthy tarsal plate and rapidly restore a normal appearance. Clinically, autologous or allogeneic substitutes [5], such as autologous ear cartilage and allogeneic sclera, are often used. Although these materials possess the advantages of suitable mechanical strength and low immunogenicity, they lack the capacity to support conjunctival regeneration and therefore require additional conjunctival substitutes. Moreover, tissue sources are limited, particularly for allogeneic sclera, which must be preserved in an eye bank [6]. Therefore, various novel scaffolds have been explored in recent years, including biodegradable scaffolds [7], drug-loading scaffolds [8], three-dimensional printed scaffolds [9], and tissue engineering scaffolds [10]. However, due to their homogeneous structure, these scaffolds still face challenges in achieving integrated tarso-conjunctival repair. Additionally, they also tend to exhibit low tissue tolerance and frequently induce chronic inflammatory responses. To date, only one study has reported a biphasic scaffold for tarso-conjunctival repair that successfully achieved re-epithelialization [11]. Nevertheless, this scaffold merely featured two layers with different pore sizes, rather than a biomimetic architecture, and lacked sufficient mechanical strength to support the eyelid. Consequently, there remains a pressing need for a bionic scaffold that provides appropriate mechanical properties, facilitates tissue adaptive regeneration, and supports integrated reconstruction of tarso-conjunctival structure. No existing research has effectively met this critical need.

Acellular dermal matrix (ADM), known for its dual structural characteristics [12], has attracted extensive attention in regenerative medicine. It has been widely used in skin regeneration, breast reconstruction, wound coverage, and rhinoplasty. ADM possesses three key characteristics essential for eyelid reconstruction: (i) It exhibits excellent biocompatibility and low immunogenicity that promote tissue regeneration; (ii) it provides sufficient mechanical strength to support the eyelids; (iii) its basement membrane naturally supports the adhesion, proliferation, and migration of epithelial cells [12]. Furthermore, it offers an abundant supply and is easy to preserve. Despite its advantages, the complication rate of ADM used for posterior lamellar eyelid reconstruction has been reported to reach 39% [13]. Based on both our experience and the literature [4], marginal necrosis and seroma are common postoperative complications, often attributable to insufficient vascularization. In our previous study, a highly porous ADM-derived product was developed through homogenization and thermal-glutaraldehyde crosslinking, which demonstrated rapid vascularization and superior biocompatibility [14]. Therefore, this sponge-like ADM-derived scaffold, combined with an intact basement membrane, shows great potential as an ideal scaffold for regenerating tarso-conjunctival composite tissue, as it can facilitate vascularization and reduce the incidence of complications.

In this study, we developed a bilamellar, thermal-glutaraldehyde crosslinked ADM-derived material (termed bGA-ADM) to mimic native tarso-conjunctival composite tissue as a therapeutic solution for posterior lamellar eyelid defects (Fig. 1). Inspired by the bilayer collagen sponge used in cartilage-bone repair [15], bGA-ADM, entirely derived from ADM, was designed with two structurally heterogeneous layers to enable integrated regeneration of composite tissues. These two layers differ in thickness, fiber alignment, and pore size, rendering the scaffold mechanically suitable, structurally biomimetic, and conducive to hierarchical tissue regeneration. To validate its heterogeneity, fibroblasts, endothelial cells and meibomian gland epithelial cells were seeded on the dermal layer, while mucosal epithelial cells were seeded on the basement membrane layer, demonstrating enhanced cell proliferation and improved functional performance in vitro. Superior cell infiltration and vascularization were observed in the bGA-ADM compared with ADM in both ear and eyelid implantation models, indicating enhanced tissue regeneration. Furthermore, to clarify its practical repair effect, bGA-ADM was applied in a novel eyelid defect model to reconstruct large posterior lamellar defects in situ, resulting in favorable aesthetic restoration and integrated tarso-conjunctival regeneration. Additionally, the potential of bGA-ADM for meibomian gland construction was investigated via RNA sequencing, providing guidance for the future development of glandular analogs with oil secretion function. The results indicated that bGA-ADM fostered an inflammation-suppressive environment through inhibition of the IL17 pathway. In summary, this study introduces a novel heterogeneous material, bGA-ADM, tailored for the integrated reconstruction of posterior lamellar eyelid defects and provides a scientific reference for its future clinical translation of bGA-ADM in the field of eyelid repair.

Fig. 1 — Schematic illustration of the preparation and application of bGA-ADM. bGA-ADM, a heterogenous bilamellar bionic scaffold, was prepared by homogenization, freezing, freeze-drying, heat-vacuum crosslinking, and glutaraldehyde crosslinking. This scaffold facilitates adhesion, proliferation, and functional activity of fibroblasts and endothelial cells within the dermal layer, as well as epithelial cells on the surface of the basement membrane layer. Therefore, it enables adaptive regeneration of tissue resembling the native posterior lamella of the eyelid

Materials and methods

Material preparation

The skin was harvested from a healthy white pig (approximately 50 kg) in accordance with standardized procedures that comply with legal regulations. After removing the superficial epidermis and subcutaneous fat, a split-thickness skin graft with an average thickness of 0.5 mm was prepared. The graft was initially treated with a virus inactivation solution primarily composed of peracetic acid for 2 hours, followed by decellularization using a self-made decellularization solution mainly containing 0.02% (w/v) trypsin at 30–40 °C for 60 minutes. Residual decellularization solution was removed by rinsing with pure water more than five times. The procedure for preparing sheet-like ADM is protected by a patent (CN102580153B). The untreated basement membrane layer was prepared separately as described above, with its thickness was controlled within 0.1–0.2 mm. To prepare the dermal layer, the sheet-like ADM was homogenized to obtain a uniform mixture with particles diameters lesser than 1 mm. The weight ratio was controlled within the range of 0.5–1.5% to ensure a high-porosity scaffold. This homogenate was poured into a mold with the basement membrane layer positioned at the bottom and then frozen at −20 °C overnight. Subsequently, the frozen content was freeze-dried for 48 hours to obtain a non-crosslinked bilamellar scaffold, which was trimmed to a final thickness of 1 mm while preserving the integrity of basement membrane layer. This scaffold then underwent an initial thermal crosslinking step at 120 °C for 24 hours in a vacuum oven (vacuum level 0.6 pa, FZG-4, Changzhou Buqun Drying Equipment Co., Ltd., China), followed by immersing in 0.05 M acetic acid solution containing 0.01% (w/v) glutaraldehyde for another 24 hours at room temperature, resulting in the formation of a bilamellar thermal-glutaraldehyde crosslinked acellular dermal matrix (bGA-ADM). This novel bilamellar scaffold, was entirely derived from the ADM. The bilamellar scaffold was thoroughly washed with pure water over 5 times to remove residual glutaraldehyde. bGA-ADM in a wet state was sterilized by gamma irradiation (25kGy for 3 hours). Alternatively, dried bGA-ADM obtained after freeze-drying could be sterilized using ethylene oxide (800 mg/L for 16 hours). The preparation process received technical support from Jiangsu Unitrump Biomedical Technology Co. (Jiangsu, China).

Characterization of physicochemical properties

To observe the general morphology of the bGA-ADM, photographic documentation of the basement membrane side, dermal side, and lateral side was obtained using a single-lens reflex (SLR) camera (Nikon, Japan). The microstructure of the basement membrane surface, dermal surface, and its longitudinal section (cut up with liquid nitrogen) was further examined using a scanning electron microscope (SEM, ZEISS Gemini 300, Germany). Briefly, the samples were dried using a freezing-drier (YB-FD-1, SHYB Co., Ltd., China) prior to the SEM observation. SEM images were subsequently analyzed using Image J software (United States) to evaluated layer thickness (n = 3, indicating three independent samples per group), angle of fiber arrangement (n = 5), and pore size (n = 24). For fiber orientation analysis (Fig. S1), five angles within each layer were measured per sample. For pore size evaluation, the diameters of ten randomly selected pores within each layer were measured per sample.

The three-dimensional surface topography of bGA-ADM was scanned by atomic force microscopy (AFM, SPM-9700, Shimadzu, Japan), and the average surface roughness (Ra) of the basement membrane surface and dermal surface (n = 3) was further calculated using Nanoscope Analysis software (United States).

Residual DNA content is a critical indicator for assessing the extent of biomaterial decellularization and for predicting its potential immunogenicity. DNA isolation was performed according to the manufacturer’s instructions (TIANGEN Biotech, China). DNA was extracted from samples (n = 3) of equal mass and dissolved in 200 μL of Tris-EDTA buffer. After adding 196 μL of double-distilled water to 4 μL of the DNA-containing buffer, the concentration of double-stranded DNA was measured using a photometer (Eppendorf BioPhotometer Plus, Germany).

Fourier-transform infrared spectroscopy (FT-IR, Nicolet Nexus 670 FTIR spectrometer, Thermo Fisher Scientific Inc., United States) was employed to characterize the infrared spectral properties of ADM and bGA-ADM in a dry atmosphere at room temperature, with data collected at 4 cm⁻¹ intervals over a wavelength range of 400–4000 cm^−1.

The water contact angles of the basement membrane layer and dermal layer (n = 3) were measured using an OCA 20 contact angle system (Dataphysics, Germany) with a tilting base to reflect the hydrophilicity of bGA-ADM. Pure water droplets were placed on the surface, and their behavior over time was recorded on video. The water contact angle was calculated by fitting analysis using ImageJ software (United States).

The porosity of the ADM, bGA-ADM, and its spongy dermal layer (n = 3), was measured using the liquid absorption method. A specific mass of the sample (W₀) was weighed, and its volume was measured (V_s). After complete immersion in anhydrous ethanol under vacuum for 30 minutes, the surface liquid was blotted off and the material was reweighed (W_e). ρ presents the density of ethanol. Pore volume (V_p) and porosity were calculated as follows:

The mechanical properties of ADM, bGA-ADM, and native rabbit tarsal plate were assessed using a mechanical analyzer (Model 5542, Instron, United States). bGA-ADM and ADM were equilibrated in PBS at room temperature before testing. Samples were cut into rectangular strips (25 mm in length, 2 mm in width), and their thickness was measured (n = 3). The materials were stretched at a rate of 2 mm/min until fracture, with time, load, and stretch length recorded. The resulting stress-strain curves, ultimate tensile strength (UTS), strain at UTS, and Young’s modulus (calculated from the slope within the 5–15% strain range) were further analyzed using Origin software (United States). The parameters were calculated as follows:

Isolation, culture, and identification of fibroblasts

Human fibroblasts (Fbs) were isolated from five healthy female patients (mean age 29.0 years) undergoing blepharoplasty. Each patient provided written informed consent. Eyelid skin was first washed with PBS, then immersed in a triple-antibiotic solution (penicillin-streptomycin-amphotericin B) for 15 minutes, and rinsed with PBS. Subcutaneous tissue was removed, and the samples were digested overnight at 4 °C using 0.1% (w/v) Dispase II enzyme solution (Roche, China). After removal of the epidermis, the dermal tissue was minced and further digested with 0.1% (w/v) Collagenase NB4 (Serva, Germany) solution for 3 hours. The mixture was filtered through a 40 μm filter, and centrifuged at 1500 rpm for 5 minutes to concentrate cells. Isolated Fbs were cultured in high-glucose Dulbecco’s modified Eagle’s medium (DMEM, Gibco, United States) supplemented with 10% fetal bovine serum (CellMax, China) and 1% streptomycin-penicillin-amphotericin B solution (Gibco, United States), and incubated at 37 °C in a humidified atmosphere containing 5% CO2.

The morphology of Fbs was observed under an optical microscope, and the specific marker vimentin was detected by immunofluorescence staining. Cells were incubated with anti-vimentin antibody (ab92547, Abcam, United States) at a 1:200 dilution, followed by a goat anti rabbit FITC-conjugated secondary antibody (Beyotime, China). Nuclei were counterstained with DAPI (0100–20, Southernbiotech, United States) for cell identification.

Culture and expansion of oral keratinocytes

Immortalized human oral keratinocytes (HOKs) used in this study were denoted by the Department of Stomatology of our hospital. HOKs were cultured in low-glucose DMEM supplemented with 10% fetal bovine serum and 1% streptomycin-penicillin-amphotericin B solution, and were incubated at 37 °C in a humidified atmosphere contained 5% CO2.

Culture and expansion of human umbilical vein endothelial cells

Human umbilical vein endothelial cells (HUVECs) were obtained from the cell bank of the Shanghai Institute of Cell Biology, Chinese Academy of Sciences (Shanghai, China). HUVECs were cultured in low-glucose DMEM supplemented with 10% fetal bovine serum and 1% streptomycin-penicillin-amphotericin B solution, and incubated at 37 °C in a humidified atmosphere containing 5% CO2.

Isolation, culture, and identification of meibomian gland epithelial cells

Meibomian gland epithelial cells (MGECs) were isolated from 10 patients (mean age 19.9 years), undergoing severe ptosis correction surgery. Each patient provided written informed consent. Resected tarsal plates were first washed with PBS, immersed in a triple-antibiotic solution for 15 minutes, and then rinsed with PBS. The connective tissue surrounding the tarsal plate was removed. The samples were digested overnight at 4 °C with a mixture of 0.15% (w/v) Collagenase NB4 (Serva, Germany) and 0.3 U DispaseII enzyme solution, followed by an additional 2-hour digestion at 37 °C. The mixture was centrifuged at 1200 rpm for 8 minutes to concentrate the tissue fragments. These tissue fragments were then further digested with 0.05% (w/v) trypsin (Gibco, United States) at 37 °C for 5 minutes and centrifuged at 1500 rpm for 5 minutes to collect cells. MGECs were cultured in complete keratinocyte medium (KM, Sciencell, United States) and incubated at 37 °C in a humidified atmosphere containing 5% CO2.

The morphological characteristics of MGECs were observed under a light microscope. Cells were identified by LipidTOX^TM staining (1:200, H34477, Invitrogen, United States) and keratin 14 immunofluorescence staining (1:200, ab181595, Abcam, United States).

Toxicity and cytocompatibility of materials

Materials were sized to fit the wells of a 24-well plate. The toxicity of the materials was evaluated indirectly using the extract liquid method. ADM and bGA-ADM were completely immersed in high-glucose medium for 72 hours. The extracts were collected, filtered through a 22-μm filter (Corning, United States), and mixed 1:1 with growth medium to obtain conditioned medium. Compared with growth medium, the proliferation of Fbs (2000 cells per a well in a 96-well plate) in the conditioned medium was assessed on days 1, 3, 5, and 7 using the Cell Counting Kit-8 (CCK-8, Dojindo, Japan). After incubation at 37 °C for 3 hours, the absorbance of the samples was measured at 450 nm using a microplate reader (Thermo, United States). Cytocompatibility was further examined by Live/Dead staining (Yeasen Biotechnology, China). Fbs (1 × 10⁶ cells per piece) were seeded on the dermal layer of ADM and bGA-ADM, and co-cultured for 10 days to evaluate their viability.

Adhesion of varied cells on materials

To enhance cell adhesion efficiency, cell suspensions (100 μL) were carefully and evenly seeded onto the materials, which were then incubated for 4 hours before adding sufficient culture medium (1.5 mL per well in a 24-well plate).

Fbs and MGECs, each at a density of 1 × 10⁴ cells per piece, were seeded onto the dermal layer of ADM and bGA-ADM. Fbs were cultured in DMEM medium, whereas MGECs were cultured in KM medium for 3 days. The samples were fixed with 4% paraformaldehyde and stained with tetramethyl rhodamine isothiocyanate (TRITC)-tagged phalloidin (1:200, 40734ES75, Yeasen Biotechnology, China) according to the manufacturer’s introduction to label F-actin, thereby visualizing adhered cells in red. Nuclei were counterstained with DAPI to assist in cell localization. The adhered morphology of MGECs seeded on the dermal surface and HOKs seeded on the basement membrane surface was further examined by SEM. MGECs (5 × 10⁴ cells per piece) and HOKs (2 × 10⁴ cells per piece) were co-cultured with the scaffolds for 10 days and 3 days, respectively.

To assess adhesion and spreading of endothelial cells (ECs), HUVECs (6 × 10⁴ cells) were directly seeded onto the dermal layer of ADM and bGA-ADM, sized to fit the well of a 24-well plate. After co-culture for 10 minutes and 2 hours, respectively, the actin cytoskeleton and nuclei of adhered HUVECs were stained with TRITC-tagged phalloidin (1:200) and DAPI. Cell adhesion and spreading were observed using a fluorescence microscope (Thunder imaging systems, Leica, Germany) and were quantitatively analyzed with Image J software.

Function of varied cells on materials

For proliferation, HUVECs (2.5 × 10⁴ cells) were directly seeded onto the dermal layers of both ADM and bGA-ADM, sized to fit the wells of a 48-well plate. Cells were cultured in a complete medium, and proliferation was assessed using the CCK-8 assay on days 1 and 3. After incubation at 37 °C for 4 hours, the absorbance was measured at 450 nm using a microplate reader (Thermo, United States). For tube formation, HUVECs (5 × 10⁴ cells) were evenly seeded onto the dermal side of Matrigel-coated (356234, Corning, United States) scaffolds, sized to fit the wells of a 24-well plate, and cultured in the serum-free culture medium. After 5 hours of incubation, Calcein-AM staining was performed, and tube-like structure were visualized using a fluorescence microscope. Quantitative analysis, including the number of junctions and total tube length, was performed using the Angiogenesis Analyzer plugin in Image J (United States).

HOKs (2 × 10⁴ cells) were seeded on ADM and bGA-ADM, sized to fit the wells of a 24-well plate, and labeled with Pan-CK antibody, which recognizes multiple keratins. After fixation in 4% paraformaldehyde, samples were incubated with Pan-CK antibody (ab7753, Abcam, United States) at a 1:200 dilution, followed by a goat anti-rabbit FITC-conjugated secondary antibody (Beyotime, China) for visualization. The cytoskeleton was counterstained with TRITC-labeled phalloidin (1:200), and nuclei were stained with DAPI for cell identification. Neutral lipid droplets were stained with LipidTOX^TM (1:200, H34477, Invitrogen, United States) to display intracellular lipid formation. MGECs (1 × 10⁵ cells) were seeded on ADM and bGA-ADM (sized to fit the wells of a 24-well plate), co-cultured for 10 days, and then stained with LipidTOX^TM at a 1:200 dilution.

Quantitative real-time polymerase chain reaction (qRT-PCR)

The expression of matrix deposition-related genes in Fbs was evaluated using qRT-PCR. Fbs (1 × 10⁶ cells) were seeded in ADM and bGA-ADM, sized to fit the wells of a 24-well plate, and co-cultured for 3 days. Total RNA was extracted using TRIzol reagent (Invitrogen, United States), and subsequently reverse-transcribed into cDNA using an RNA purification kit (EZB-RN001-plus, EZBioscience, United States), and a Revers Transcription Kit (A0010CGQ, EZBioscience, United States). Gene expression levels of collagen type I (COL-1), collagen type III (COL-3), aatrix metallopeptidase 1 (MMP1) and matrix metallopeptidase 3 (MMP3) were normalized against β-actin. Primers used in this study were synthesized by Sangon Biotech Co. (China) and are listed in Table 1.

Table 1.

Primer sequence for qPCR

Cell	Gene	Forward primer	Reverse primer
Fb	β-actin	ATCATGTTTGAGACCTTCAA	CATCTCTTGCTCGAAGTCCA
	COL-1	GGCGGCCAGGGCTCCGACCC	AATTCCTGGTCTGGGGCACC
	COL-3	TGGTGTTGGAGCCGCTGCCA	CTCAGCACTAGAATCTGTCC
	MMP1	GGAGCTGTAGATGTCCTTGGGGT	GCCACAACTGCCAAATGGGCTT
	MMP3	AGGACAAAGCAGGATCACAGTTG	CCTGGTACCCACGGAACCT
MGEC	GAPDH	CAGGAGGCATTGCTGATGAT	GAAGGCTGGGGCTCATTT
	MM9	GCACGACGTCTTCCAGTACC	GGTTCAACTCACTCCGGGAA
	IL6	AGACAGCCACTCACCTCTTCAG	TTCTGCCAGTGCCTCTTTGCTG
	TNF	CTCTTCTGCCTGCTGCACTTTG	ATGGGCTACAGGCTTGTCACTC
	CXCL2	CTCAAGAACATCCAAAGTGTG	ATTCTTGAGTGTGGCTATGAC
	CXCL3	CGAAAAGATACTGAACAAGGGGA	ATTTTCAGCTCTGGTAAGGGC
	CXCL8	ACTGAGAGTGATTGAGAGTGGAC	AACCCTCTGCACCCAGTTTTC

Open in a new tab

Ear implantation model of rabbit

All animal procedures were approved by the Ethics Committee of Shanghai Ninth People’s Hospital (No. SH9H-2023-A886-1) and conducted in accordance with the guidelines of the National Institutes of Health Guide for the Care and Use of Laboratory Animals. This study adhered to the ARRIVE 2.0 guidelines to ensure rigorous and comprehensive reporting of all in vivo experiments. Three New Zealand white rabbits were anesthetized by intramuscular injection of a Zoleti: xylazine hydrochloride mixture (8:1, 0.35 mL/kg), and one ear of each rabbit was used. A total of six 1 × 1 cm rectangular frames were marked along the posterior auricular vein on the dorsal surface of each rabbit ear, with 1 cm spacing between frames. ADM and bGA-ADM were placed on opposite sides of the vein. After skin disinfection, a diluted solution of articaine hydrochloride and epinephrine tartrate (1:1 dilution, Primacaine, France) was injected into the subcutaneous tissue of the marked regions for local infiltration anesthesia and tissue separation. The skin was incised, and subcutaneous tissue was bluntly dissected to form a pocket for material placement. The basement membrane side of the material faced the skin, and the dermal side faced the cartilage. The incision was closed using 6–0 sutures, cleaned and subsequently covered with erythromycin ointment for 3 consecutive days postoperatively. At 4 weeks postoperatively, ADM and bGA-ADM were harvested to evaluate cell infiltration and vascularization macroscopically and histologically. In addition, the inflammatory response within the scaffolds was assessed by a blinded pathologist using HE staining, based on the extent of immune cell recruitment.

Eyelid implantation model of rabbit

Three New Zealand white rabbits were anesthetized by intramuscular injection as described above, and the fur on their eyelids was trimmed. One upper eyelid of each rabbit was assigned to ADM, and the contralateral upper eyelid to bGA-ADM. A 1 × 1 cm surgical region was marked on the middle-outer portion of the upper eyelid. A corneal protective sheet was applied to protect the eyeball throughout the procedure. After disinfection, the surgical field was rinsed with saline, followed by local injection of a diluted solution articaine hydrochloride and epinephrine tartrate (1 :1 dilution). An incision approximately 5 mm above the upper edge of the tarsal plate, parallel to the eyelid margin, was made over a length of 1.5 cm. The conjunctival fold was fully dissected to form a cavity. The materials were placed on the upper edge of the tarsal plate and fixed to the outer layer using 7–0 PDS sutures. The dermal side faced the conjunctiva, whereas the basement membrane side faced the skin. The skin incision was closed, and a erythromycin ointment was applied. Penicillin (20 U/rabbit) was administered intramuscularly for 3 days postoperatively. Samples were harvested at 4 weeks postoperatively. For gross observation, the materials were fully exposed in situ. For histological analysis, hematoxylin and eosin (HE) staining and CD31 immunofluorescence staining (1:200, CST15585, Cell Signaling Technology, China; CY3-conjugated secondary antibody used for visualization) were performed to evaluate cell infiltration and vascularization of ADM and bGA-ADM in the eyelid environment. Cell infiltration depth and vessel number were further quantified.

Defect of posterior lamella of rabbit eyelid model and bGA-ADM repair of large-sized defect in situ

A rabbit eyelid model of posterior lamellar eyelid defect was established using a cutaneous approach as follows. After anesthesia, the fur on the upper eyelids of New Zealand white rabbits was trimmed. A defect region was marked on the middle-outer portion of the upper eyelid, and a local injection of diluted articaine hydrochloride and epinephrine tartrate solution (1: 1 dilution) was administered. A corneal protective sheet was used throughout the procedure. An incision parallel to the eyelid margin was made along the upper edge of the tarsal plate, approximately 4 mm from the eyelid margin, and the anterior and posterior layers of the eyelid were separated. Tissue dissection was extended to the conjunctival fold superiorly and to the eyelid margin inferiorly. Along the upper edge of the marked region (aligned with the upper margin of the tarsal plate), the conjunctiva was punctured from the skin side with a sharp knife and then cut using ophthalmic scissors. Two vertical incisions perpendicular to the eyelid margin, extending from the horizontal conjunctival incision to the eyelid margin, were made to completely resect the tarsal plate and corresponding conjunctiva.

Of the six New Zealand white rabbits included in this experiment, eight eyelids were allocated to the graft repairing, and four eyelids to the blank group without graft as controls. bGA-ADM, trimmed slightly larger than the defect, was used to repair the posterior lamellar defect. The basement membrane side was oriented toward the eyeball, and the dermal side faced the skin. First, the two ends of the eyelid margin were bridged with bGA-ADM using 7–0 PDS sutures, stitching at the anterior and posterior sides of the end, and the gray line. Next, the two ends of the upper margin of bGA-ADM were fixed to the anterior layer of the eyelid using 7–0 PDS sutures. Subsequently, the conjunctival tissue surrounding the defect was mobilized and sutured to the upper, left, and right edges of bGA-ADM with 7–0 PDS sutures. Finally, the eyelid skin was sutured to bGA-ADM at the level of the eyelid margin, and the skin incision was closed using 6–0 PDS sutures. The conjunctival sac was washed with saline, and the entire eye was covered with erythromycin ointment. The surgical region was dressed for 3 days postoperatively, and penicillin (20 U/rabbit) was administered intramuscularly.

At 1 month postoperatively, the repair outcome and eyelid mobility were documented using an SLR camera. Samples were harvested from the repair region, and processed sequentially by fixation, dehydration, paraffin embedding, and serial sectioning sequentially. HE staining and Masson staining were performed for histological observation. Periodic Acid-Schiff (PAS) staining (C0142S, Beyotime, China) and Pan-CK immunofluorescence staining (1:200, ab7753, Abcam, United States) were used to assess conjunctival regeneration. Goblet cells and conjunctival thickness were quantified in both the repair region and the native region. The fiber composition of the tarsal plate in the repair region and the native region was visualized using Sirius Red staining (G1472, Solarbio, China), and vessels were visualized by CD31 immunofluorescence staining (1:200, CST15585, Cell Signaling Technology, China). Goblet cell counts, conjunctival thickness, thickness of the posterior lamella, and areas of COLIII and COLI were further analyzed using Image J software. For vascular quantification, five high-power fields were randomly selected from each sample, and the red-stained tubular structures were counted. Three independent samples were analyzed per group.

RNA sequencing and qRT-PCR

Total RNA was extracted from MGECs (2 × 10⁵ cells) cultured on bGA-ADM and ADM, sized to fit the wells of a 24-well plate, for 10 days using TRIzol reagent. The RNA concentration and purity were assessed with a NanoDrop 2000 spectrophotometer (Thermo Scientific, United States), and RNA integrity was evaluated using an Agilent 2100 Bioanalyzer (Agilent Technologies, United States). Only high-quality RNA samples with an RNA integrity number (RIN) greater than 7 were selected for sequencing library construction. RNA libraries were amplified according to the Illumina protocol. Sequencing was performed on an Illumina NovaSeq 6000 platform, generating 150-bp paired-end reads. Differential expression analysis was conducted using DEseq2 software, with significantly differentially expressed genes (DEGs) defined by a q value < 0.05 and foldchange > 2 or foldchange < 0.5. These DEGs were then subjected to enrichment analysis of Gene Ontology (GO) functions and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. Transcriptome sequencing and primary analysis were performed by OE Biotech Co., Ltd. (Shanghai, China). Bioinformatic analyses were conducted using the OmicStudio tools (https://www.omicstudio.cn/tool) and the OECloud tools (https://cloud.oebiotech.com).

The crucial genes, including matrix metallopeptidase 9 (MMP9), interleukin 6 (IL6), tumor necrosis factor (TNF), and CXC chemokine ligands 2, 3, and 8 (CXCL2, CXCL3, and CXCL8) were further validated by qRT-PCR. Their expression levels were normalized to the housekeeping gene GAPDH. Primers were synthesized by Sangon Biotech Co. (China), and are listed in Table 1.

Statistical analysis

Quantitative data were analyzed using GraphPad Prism version 10.0 software (GraphPad, United States), and are presented as the mean ± standard deviation (SD). Statistical comparisons were performed using Student’s t-test, one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test, or two-way ANOVA followed by Bonferroni’s method, as appropriate. Statistical significance was defined as *p < 0.05, and ns indicated no significant difference.

Results and discussion

The physicochemical properties of bGA-ADM

The preparation of the bGA-ADM is illustrated in Fig. 1. Briefly, porcine skin was decellularized to obtain sheet-like ADM, followed by sequential homogenization, freezing and freeze-drying with a thin basement membrane layer, heat-vacuum crosslinking, and glutaraldehyde crosslinking. Heat-vacuum crosslinking enhances the internal connections via the formation of amide bones between amino and carboxyl groups [16], whereas glutaraldehyde crosslinking further increases mechanical strength through condensation reactions between glutaraldehyde and amino or carboxyl groups [17].

Macroscopically, bGA-ADM appeared a porcelain-white and exhibited a clearly defined bilayer structure with distinguishable surfaces. The dermal layer was porous and spongy with a rough surface, while the basement membrane layer was relatively flat, with a distinct multi-directional texture. A clear layered structure of bGA-ADM was visible from the lateral side (Fig. 2a). SEM imaging revealed that the dermal layer contained variably sized pores, and the typical collagen fiber bundle structure was absent due to recombination of ADM-derived microparticles [18]. The basement membrane layer’s surface displayed local bulges corresponding to the papillary ridges. Cross-sectional observation showed that the basement membrane layer was firmly adhered to the dermal layer without any obvious gaps, confirming successful integration through freeze-drying and crosslinking methods. The average thickness of the dermal layer and the basement membrane layer was 758.09 μm and 146.73 μm, respectively (Fig. 2d), consistent with the overall thickness of the native posterior lamella of the eyelid (approximately 1 mm) [3].

Collagen fiber arrangement was further analyzed (Fig. 2e, Fig. S1). The dermal layer exhibited a larger deviation angle (29.6°) compared with the basement membrane layer (23.8°), indicating a looser internal architecture conducive to cell infiltration and matrix deposition, with collagen fiber orientation being closer to 45° [18]. SEM images (Fig. 2b) also showed micropores with an average diameter of 2.03 μm on the basement membrane surface and macropores with an average diameter of 30.59 μm on the dermal surface (Fig. 2f). This hierarchical pore structure suggests that cells preferentially adhere and form layers on the basement membrane surface, while they can infiltrate into the inner dermal layer through the macropores [19]. AFM analysis (Figs. 2c, 2g) further demonstrated that the basement membrane layer had significantly greater surface roughness than the dermal layer, a feature known to enhance cell adhesion [20]. Collectively, these findings indicate that the basement membrane layer promotes cell adhesion on the surface, whereas the dermal layer facilitates cell infiltration into the inner space.

The impacts of tissue source, decellularization protocol, and sterilization method on the host response post-implantation were also critically considered. Inadequate decellularization may leave residual antigens that trigger immune rejection and lead to graft failure. Generally, the residual DNA content below 50 ng/mg in decellularized matrices is considered acceptable [21]. The average residual DNA content in bGA-ADM was 25 ng/mg (Fig. 2h), well below this threshold, indicating thorough decellularization and suggesting favorable compatibility with host tissues and a low risk of immune rejection.

FT-IR analysis (Fig. 2 bGA-ADM presented characteristic absorption peaks at wavelengths of 1636.87 cm⁻¹, 1549.70 cm⁻¹ and 1338.15 cm⁻¹, corresponding to the amide I band (1600–1800 cm⁻¹), amide II band (1470–1570 cm⁻¹), and amide III band (1250–1350 cm⁻¹), respectively [22]. These classic peaks indicate the preservation of the collagen triple-helix secondary structure, which is essential for maintaining the functional integrity [23]. The presence of these characterized peaks of amino bonds following homogenization, thermal crosslinking and glutaraldehyde crosslinking confirms that the classic secondary structure of collagen was retained in bGA-ADM and that the original functions of matrix protein were largely preserved.

Dynamic changes in water contact angle on the surface of the basement membrane and dermal surfaces are shown in Fig. 2j, reflecting the liquid absorption behavior of the material. At 0 seconds, the mean water contact angles were 88.5° on the basement membrane side and 91.5° on the dermal side. The contact angles gradually decreased over time, and the water droplets disappeared by 28 seconds on the basement membrane side and 30 seconds on the dermal side. No statistically significant differences were observed between the two surfaces at 0, 20, and 30 seconds, with a significant difference only at 10 seconds. These data indicate comparable hydrophilicity of the two surfaces, which favors uniform tissue fluid distribution throughout the full thickness of bGA-ADM. Additionally, bGA-ADM displayed hydrophilicity similar to ADM (Fig. S3), which facilitates cell adhesion and infiltration [24].

The average porosities of ADM, the dermal layer of bGA-ADM, and bGA-ADM as a whole were 74.6%, 90.0%, and 86.0%, respectively (Fig. 2k). Both bGA-ADM and its spongy dermal layer exhibited significantly higher porosity than ADM, while no statistically significant difference was observed between them. Additionally, the average porosity of the basement membrane layer of bGA-ADM was 75.1%, which did not differ significantly from that of ADM (Fig. S4). These results indicate that the increased porosity of bGA-ADM is primarily attributable to structural modification of the dermal layer. Compared with the low-porosity materials, high-porosity scaffolds provide greater spatial space for nutrient transport, endothelial cell growth and migration, thereby promoting angiogenesis [25]. Given that material vascularization largely depends on porosity [26], it is reasonable to infer that bGA-ADM has faster vascularization than ADM.

Regarding mechanical support, which is critical for the eyelid, the mechanical properties of bGA-ADM were analyzed (Figs. 2l–2o). To evaluate the practicality and translational potential, bGA-ADM was compared with the native tarsal plate and a commercial product (ADM). The mean values of Young’s moduli for native tarsal plate, ADM, and bGA-ADM were 7.15 Mpa, 22.28 Mpa and 5.46 Mpa, respectively, while the average ultimate tensile strengths (UST) were 4.32 Mpa, 8.00 Mpa, and 1.10 Mpa. In addition, the strain at UST did not differ significantly among aforementioned three scaffolds. The Young’s modulus and UST of ADM were markedly higher than those of rabbit’s tarsal plate and bGA-ADM, whereas no significant difference was observed between the latter two. These findings indicate that bGA-ADM exhibits mechanical properties more closely resembling those of the native tarsal plate compared with ADM. The difference between ADM and bGA-ADM is likely attributable to the increased porosity of the latter [27]. With ongoing cell infiltration and matrix deposition, the mechanical strength of bGA-ADM is expected to improve. Moreover, bGA-ADM demonstrates sufficient mechanical support for clinical application, as evidenced by the reported tensile strength of human tarsal plate (1.73 ± 0.61 Mpa) by Sun MT et al. [28].

In summary, bGA-ADM is characterized by its high porosity, excellent hydrophilicity, biomimetic mechanical strength, low immunogenicity. Owing to its heterogeneous bilamellar structure, the basement membrane layer of bGA-ADM is well suited for cell adhesion, whereas the dermal layer facilitates cell infiltration. Thus, bGA-ADM holds great potential for the integrated reconstruction of tarso-conjunctival composite tissue.

The bGA-ADM possesses good biocompatibility with Fbs and HUVECs

The tarsal plate is composed of a fibrous matrix and meibomian glands, with the matrix primarily consisting of COL1, COL3 and Fbs. Previous studies have shown that Fbs are heterogeneous, with phenotypes and characteristics that vary according to tissue type and anatomical location [29]. Therefore, fibroblasts isolated from eyelids were used to evaluate the biocompatibility of bGA-ADM (Fig. 3a). Isolated primary Fbs were identified using vimentin, a specific marker expressed in all fibroblast’s subtypes, which stained positively in these cells (Fig. 3c). Figure 3b shows the viability of Fbs on both ADM and bGA-ADM by Live/Dead staining. Fbs on bGA-ADM exhibited superior adhesion and proliferation compared with those on ADM (Fig. 3e). This enhanced cell survival correlates with material porosity; high-porosity materials support improved cell adhesion and metabolism [30], indicating that bGA-ADM provides a more favorable microenvironment for Fb attachment and survival. Additionally, the cytotoxicity of bGA-ADM was indirectly assessed using material extracts (Fig. 3f). With complete DMEM medium as a control, Fbs were cultured in the conditioned media extracted from both ADM and bGA-ADM, and monitored for 7 days. Fb proliferation on days 1, 3, 5, and 7 showed no significant differences among groups, indicating that bGA-ADM exhibits good biocompatibility and no detectable cytotoxicity.

The adherent morphology of Fbs was further examined by cytoskeleton staining (Fig. 3d). Fbs adhering to both ADM and bGA-ADM exhibited a typical spindle shape with extended pseudopodia, demonstrating their capability to adhere and migrate on bGA-ADM. Moreover, expression of genes related to matrix formation was analyzed (Fig. 3g). Contrary to Fbs on ADM, those on bGA-ADM showed higher expression of COL1 and COL3 but reduced expression of MMP1 and MMP3. COL1 and COL3 are the major components of the tarsal plate matrix, whereas MMP1 and MMP3 are involved in matrix remodeling [31]. These results suggest that Fbs on bGA-ADM favor extracellular matrix synthesis over degradation, indicating early matrix deposition within bGA-ADM during the initial stage of tissue integration. Additionally, substrate stiffness has been reported to influence fibroblast behavior. Fbs cultured on stiffer scaffolds demonstrate enhanced proliferation and collagen production [32]. This suggests that Fbs within the crosslinked bGA-ADM are more functionally active than those in the non-crosslinked scaffolds, thereby promoting more effective scaffold remodeling.

Vascularization is a critical determinant of successful engraftment and integration of substitutes for eyelid defect repair. Therefore, the behavior of HUVECs, including proliferation, adhesion, spreading, and tube formation on bGA-ADM was evaluated in vitro and compared with ADM. HUVECs were seeded on the dermal layer of the scaffolds (Fig. 3h), and their proliferation on bGA-ADM was significantly higher than on ADM (Fig. 3i). To verify the ability of the scaffolds to capture ECs, the adhesion and spreading of HUVECs were examined. (Fig. 3j). The bGA-ADM group showed markedly higher numbers of adhered cells at 10 minutes and significantly greater cell spreading at 2 hours compared with the ADM group (Fig. 3m), indicating that bGA-ADM provides a more favorable interface for EC attachment and extension. Tube formation experiments were performed on ADM and bGA-ADM coated with matrix gel for 5 hours (Fig. 3k). Both the number of junctions and the total tube length were significantly greater in the bGA-ADM group than in the ADM group (Fig. 3l), suggesting that bGA-ADM has a stronger ability to support vessel formation. Scaffolds with larger pores are known to facilitate neovascular outgrowth [33], which may explain the superior EC behavior observed on bGA-ADM. Collectively, these findings demonstrate that bGA-ADM effectively promotes EC proliferation, adhesion, spreading, and tube formation, providing strong evidence for its potential in eyelid defect repair.

bGA-ADM supports the growth and function of HOKs and MGECs

Owing to the Collagen VIII and the native 3D structure, the basement membrane naturally facilitates the recruitment and migration of epithelial cells [34]. In clinical practice, oral mucosa is commonly used as a substitute for the conjunctiva, and oral mucosa epithelial cells exhibit migratory behavior comparable to conjunctival epithelial cells [35]. Repair of both oral mucosal and conjunctival defects relies on the proliferation and migration of surrounding normal epithelial tissues. Therefore, immortalized HOKs were used to investigate epithelial cell adhesion on the basement membrane of bGA-ADM (Fig. 4a).

As shown in Fig. 4b, HOKs exhibited a polygonal morphology, adherent growth, and a pavement-like arrangement. SEM observation further demonstrated that HOKs completely covered the basement membrane surface of both ADM and bGA-ADM, displaying a flattened morphology and arranged in either a single layer or in overlapping layers (Fig. 4c). The typical groove structure of the basement membrane was no longer visible. HOKs were labeled with Pan-CK and co-stained with phalloidin to visualize cell distribution (Fig. 4d). Pan-CK, a mixture of monoclonal anti-keratin antibodies, recognizes multiple keratins (cytoskeletal proteins), including types 1, 4, 5, 6, 8, 10, 13, 18, and 19, which are stably expressed in epithelial tissues. Phalloidin binds to F-actin, another cytoskeletal protein, resulting in HOKs staining in both green and red and appearing yellow in merged images. HOKs on both ADM and bGA-ADM presented a pebble-like morphology, consistent with SEM observations. The cell-covered area accounted for approximately 79.8% on ADM, and 75.5% on bGA-ADM (Fig. 4e), with no statistically significant difference. These findings indicate that the unique microstructure of the basement membrane surface was preserved during bGA-ADM preparation, maintaining its capacity to support epithelial cell adhesion, migration, and growth, and suggesting potential for self-conjunctivalization.

The meibomian gland, an integral component of the tarsal plate, consists of alveoli and ducts, with alveoli composed of MGECs that secret oil. Isolation and purification of human MGECs have long hampered progress in meibomian gland research. Substantial advances were not reported until 2010, when in vitro isolation, expansion, and immortalization of MGECs were first described [36]. However, immortalized MGECs exhibit several limitations, including activation of keratinization-related pathways, accumulation of lipid droplets in lysosomes, and a lack of key lipids such as (O-acyl)- ω-hydroxy fatty acids and wax esters [37]. To better simulate physiological interactions between MGECs and bGA-ADM, primary MGECs were therefore used to assess biocompatibility (Fig. 4f).

Under optical microscopy, isolated MGECs displayed oval or polygonal morphologies, assembled distribution, and a pebble-like arrangement (Fig. 4h). Neutral lipid droplets within MGECs were visualized using LipidTOX^TM staining to assess the intracellular lipid formation. Small lipid droplets were predominantly observed around the nuclei, consistent with previous reports that MGECs produce small amounts of lipids during rapid proliferation phases [38]. In addition, MGECs expressed Krt14, in line with the phenotype of glandular epithelial cells [39].

The adhesion morphology of MGECs on ADM and bGA-ADM was further examined by cytoskeleton staining (Fig. 4g) and SEM (Fig. 4i). MGECs adhered well to both ADM and bGA-ADM, exhibiting classic oval or polygonal shapes. Lipid production in MGECs after 10 days of co-culture with the scaffolds was then evaluated (Fig. 4j). Lipid droplets were detected in MGECs cultured on both ADM and bGA-ADM, and lipid content per cell was quantified (Fig. 4k). Although MGECs on bGA-ADM showed a trend toward higher lipid accumulation than those on ADM, the difference was not statistically significant. Notably, MGECs on bGA-ADM exhibited a more assembled distribution, in contrast to the relatively scattered pattern observed on ADM. Collectively, these results suggest that bGA-ADM provides a conducive environment for MGEC assembly and lipid production.

bGA-ADM exhibits fast vascularization and adaptive regeneration in vivo

Vascularization and cellularization are critical for the survival and integration of implanted constructs. Therefore, ADM and bGA-ADM were implanted in both the ear and the posterior lamella of the eyelid to evaluate blood vessel growth and host cell infiltration.

As illustrated in Fig. 5a, the scaffolds were placed in the subcutaneous layer of the rabbit ear, adjacent to the posterior auricular vein. Representative images obtained immediately after surgery and at 4 weeks postoperatively are shown in Fig. 5c. At 4 weeks, no necrosis was observed within the materials. HE and Masson staining were performed to assess cellularization of ADM and bGA-ADM, while CD31 staining was used to evaluate vascularization (Fig. 5b). Host cells primarily infiltrated through the dermal layer and adhered to the basement membrane side. In ADM, cells predominantly infiltrated the marginal regions adjacent to the host tissues, with sparse cellular distribution in the central regions. bGA-ADM exhibited nearly full-thickness cellular distribution, particularly in the peripheral areas. The mean cell infiltration depth was 503.33 μm in ADM, and 645.80 μm in bGA-ADM. Newly formed tissue filled the pores of the spongy dermal layer. Neither scaffold showed evident inflammatory cell infiltration, indicating their low immunogenicity. Both the relative cell infiltration level (ratio of infiltration depth to scaffold thickness) and vessel number were higher in bGA-ADM than in ADM (Fig. 5d).

A schematic diagram of the posterior lamellar implantation model in the rabbit eyelid is shown in Fig. 5e. ADM and bGA-ADM were inserted beneath the conjunctival layer. Macroscopic observation at 4 weeks post-implantation (Fig. 5g) revealed that both scaffolds maintained structural integrity and exhibited good biocompatibility, with the overlying conjunctiva appearing normal, although bGA-ADM showed a faster in vitro biodegradation rate (Fig. S2). A bulging contour toward the skin side was observed in both ADM and bGA-ADM, suggesting that the scaffolds possessed sufficient plasticity to conform to the curvature of the eyeball. HE and Masson staining (Fig. 5f) confirmed that ADM and bGA-ADM preserved their frameworks, providing guidance for cell alignment and structural support. These findings suggest that the degradation rate of bGA-ADM is well matched to the speed of tissue regeneration [40]. The pores within bGA-ADM were filled with newly formed collagenous tissues, and cells and vessels infiltrated nearly the full thickness of the scaffold. In contrast, cells and vessels in ADM were mainly distributed in the marginal regions. Quantification of cell infiltration depth and vessel number showed significant increases in bGA-ADM group (Fig. 5h).

Porosity is a key factor contributing to the biological differences between ADM and bGA-ADM. Previous studies have demonstrated that increased porosity enhances vascularization within scaffolds by improving the distribution of oxygen and other nutrients [41]. As shown in Fig. 2k, the average porosities of ADM and bGA-ADM were 74.6% and 86.0%, respectively, with a statistically significant difference. Recent research has also highlighted the pivotal role of macrophages in the revascularization of the decellularized tissues [42]. Macrophage polarization and their secretion of growth factors exert profound effects on angiogenesis. Larger pore sizes are associated with a shift from M1 to M2 macrophages phenotypes, with M2 cells secreting cytokines that promote vascular growth [43]. Matrix stiffness likewise affects macrophage behavior [44]; increased matrix stiffness can shift macrophages toward an M1 phenotype, leading to chronic inflammation and impaired tissue integration [34]. Moreover, the molecular pathways activated between bGA-ADM and ADM groups may also differ. Physicochemical surface properties, such as surface roughness, architecture, and stiffness, send regulatory signals to host cells, enabling precise control of biological responses [45]. This may explain why bGA-ADM exhibited superior vascularization and biocompatibility compared to ADM.

The bGA-ADM repairs the large posterior lamellar defects of the eyelid in situ and returns satisfactory results in the early stage

Dogs, rabbits, and mice are commonly used as models for researching meibomian gland-related diseases. Owing to their similarities to humans in eyeball size, eyelid structure, and eyelid activity, New Zealand white rabbits were selected for establishing an eyelid defect model [46]. Anatomical and histological analyses revealed that rabbit eyelids are divided into an anterior layer, consisting of skin and muscle, and a posterior layer, composed of the tarsal plate (where meibomian glands are distributed) and conjunctiva (Fig. S5). Unlike humans, rabbits predominantly rely on lower eyelid movement. The average length and height of the tarsal plate in the upper eyelid were 20 mm and 3 mm, respectively, whereas in the lower eyelid they were 23 mm and 2 mm (Fig. 6a). Considering the differences in eyelid activity and tarsal plate height, the upper eyelid, which exhibits lower blinking activity but has a larger tarsal plate, was chosen for posterior lamellar defect modeling.

To date, there have been few reports on rabbit eyelid models of posterior lamellar defect. Existing models typically involve only small-to-medium-sized tarsal plate defects and exclude defects at the eyelid margin. These models are usually created via a conjunctival approach, which is technically demanding, difficult for tissue dissection, and places knots on the ocular side, thereby risking corneal irritation. In this study, a parallel skin incision was made 4 mm from the eyelid margin. Dissection was performed extensively from the incision edge to the superior conjunctival dome and inferior eyelid margin. A short horizontal incision was then made 1 mm above the upper edge of the tarsal plate, from the conjunctival side to the skin side, for precise positioning. Tarso-conjunctival tissues of varying sizes could thus be removed as required. The entire thickness of the tarsal plate, including eyelid margin, was resected (Fig. 6b). To our knowledge, this is the first reported rabbit model of a large-sized posterior lamellar defect that includes eyelid margin resection.

As illustrated in Fig. 6c, bGA-ADM was used to repair the large-sized defect in the posterior lamella of the eyelid in situ. The basement membrane side faced the eyeball, whereas the dermal side faced the skin. The efficacy of bGA-ADM in reconstructing the posterior lamella of eyelid was evaluated by continuous postoperative observation. At 1 month postoperatively, bGA-ADM successfully restored the normal eyelid appearance and continuity of the eyelid margin (Fig. 6d, Fig. S6). The basement membrane surface was resurfaced by conjunctiva, and the eyelid demonstrated normal function (Supplementary Video). In contrast, an obvious defect persisted in the blank group (without bGA-ADM repair), where the eyelid margin remained discontinuous. HE staining and Masson staining corroborated these macroscopic findings (Fig. 6e). In the graft group, bGA-ADM was contiguous with native tarsal plate, and its surface was covered by conjunctiva that seamlessly integrated with the surrounding tissues. Cells and vessels infiltrated nearly the full-thickness of the scaffold, and abundant connective tissue was deposited within the pores. By comparison, a gap was evident in the blank group due to the absence of structural support, indicating that bGA-ADM facilitated conjunctival migration into the defect, supported the matrix deposition, and restored eyelid margin continuity. Furthermore, the thickness of the posterior lamella in the graft region did not differ significantly from that of native tissue (Fig. 6k), indicating that bGA-ADM provides sufficient structural support.

Further analyses were performed to compare the fibrous matrix and conjunctiva between the repair region and native tissue. Conjunctival goblet cells, containing mucoproteins, were stained purple on PAS staining (Fig. 6f), which play a crucial role in ocular surface homeostasis by producing and secreting soluble mucins to stabilize the tear film and prevent microbial infection [47]. Additionally, cytokeratins CK4 and CK13, specifically expressed in the conjunctival epithelial cells, were detected by Pan-CK immunostaining and appeared green (Fig. 6g, Fig. S7). Goblet cells were observed in both the repair and native regions, and conjunctival epithelial cells were arranged in multiple layers. The epithelial layers in native conjunctiva were much more than in the repair region, as confirmed by quantitative analysis of conjunctival thickness (Fig. 6j). Previous studies have shown that regenerative conjunctiva after injury often exhibits thinner epithelial layers and reduced goblet cell density [48]. Although not as robust as native conjunctiva at 1 month postoperatively, both epithelial stratification and goblet cell numbers in the bGA-ADM group were substantially higher than in the blank group. Moreover, the conjunctiva beneath the bGA-ADM was continuous with surrounding conjunctival tissue, indicating that bGA-ADM guided conjunctival epithelial cell migration to achieve rapid “self-conjunctivalization”.

Using native matrix as a reference, regeneration of matrix components in the bGA-ADM was evaluated by Sirius Red staining (Fig. 6h) and CD31 staining (Fig. 6i). Sirius Red staining showed that COL1, which appears bright red, was the predominant component of the extracellular matrix in both native tarsal plate and the bGA-ADM, while the COL3, which appears green, was relatively sparsely. Although the COL3/COL1 ratio in native tissue was significantly higher than in the graft region (Fig. 6k), ongoing matrix remodeling is expected to gradually adjust the collagen composition of bGA-ADM toward the physiological ratio [49]. Vessels were observed in both native and graft regions, with no significant difference in vessel density. Furthermore, bGA-ADM exhibited good histocompatibility and did not induce marked inflammation, as confirmed by IL-6 immunostaining (Fig. S8). Taken together, these findings support that bGA-ADM can adaptively regenerate conjunctiva and tarsal matrix within the eyelid environment. Nevertheless, given the substantial differences between the human and rabbit immune systems, the outcomes observed in this model may not be fully generalizable to humans.

Repairing large posterior lamellar defects of the eyelid with bGA-ADM is straightforward and time-efficient, without the need for additional conjunctive transplantation. Because of its sufficient supply, bGA-ADM may help avoid severe postoperative lagophthalmos associated with limited autologous tissue. Its low immunogenicity and lack of toxicity make it suitable for both pediatric and adult patients. Although the scaffold has demonstrated encouraging early results in terms of aesthetics and function, transition from beach to beside will require more comprehensive clinical trials to confirm safety, reproducibility and efficacy [50]. To sum up, bGA-ADM showed promising repair outcomes in the early stage, particularly in terms of structural support, conjunctival regeneration, and matrix deposition. Nevertheless, its long-term durability and functionality require further investigation through extended observation.

The potential for bGA-ADM to construct meibomian gland

The tarsal plate hosts numerous vertically arranged meibomian glands, whose secretions are essential for tear film formation and stabilization. Dysfunction of these glands can result in dry eye, ocular irritation and inflammation [51]. Due to their limited regenerative capacity, meibomian glands are often deficient in cases involving large posterior lamellar defects. Current treatments mainly aim to restore an acceptable appearance, while reconstruction of functional meibomian gland remains a major challenge. In this study, we further explored the superiority of bGA-ADM to support the regeneration of functional meibomian gland analogues by comparing differences in MGECs cultured in ADM and bGA-ADM using RNA-seq analysis.

Comparative transcriptomic analysis between the bGA-ADM and ADM groups identified 185 significantly upregulated and 125 significantly downregulated genes in MGECs cultured within bGA-ADM (Fig. 7a). The expression of lipogenic genes associated with meibomian glands was then analyzed [52, 53], with expression levels visualized on a heatmap where red indicates high expression and blue indicates low expression (Fig. 7b). Among the lipogenic genes, only CD36, which regulates fatty acid transport, was highly expressed, indicating no significant difference in lipogenesis between MGECs cultured in bGA-ADM and ADM. This finding is consistent with in vitro lipid quantification results (Fig. 4k). DEGs were further subjected to GO enrichment analysis, visualized as a chord diagram (Fig. 7c). Highly and weakly expressed genes in descending order were displayed on the left part, while their corresponding GO terms were arranged on the right part. These DEGs, especially for downregulated genes, were mainly enriched in processes related to inflammatory regulation [54], including “cytokine-mediated signaling pathway,” “inflammatory response,” and “chemokine-mediated signaling pathway” (Fig. 7d). Thus, the downregulated genes were further analyzed by KEGG pathway enrichment (Fig. 7e). Several pathways related to the suppression of pro-inflammatory responses were found, including the TNF signaling pathway, NF-κB signaling pathway, IL-17 signaling pathway, and NOD-like receptor signaling pathway. These results indicate that the expression of inflammation-related genes is suppressed in MGECs cultured in bGA-ADM compared with ADM.

Fig. 7 — RNA-seq reveals the differences of genes expression of MGECs cultured within bGA-ADM and ADM. (a) Volcano plot of DEGs between bGA-ADM and ADM groups. (b) Expression patterns of lipogenic genes between bGA-ADM and ADM groups. (c) Chord diagram of GO enrichment analysis. (d) Bar chart of top30 enriched GO terms of downregulated DEGs in bGA-ADM groups compared to that in ADM groups. (e) Scatter plot of top 20 enriched KEGG pathways of downregulated DEGs in bGA-ADM groups compared to that in ADM groups. (f) Expression patterns of the IL-17 signaling pathway between bGA-ADM and ADM groups. (g) Expression levels of IL17 pathway-related genes (MMP9, IL6, TNF, CXCL2/3/8) measured by qRT-PCR. (n = 4) (h) Schematic diagram of MGECs cultured within bGA-ADM. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001

Previous studies have demonstrated that inflammation can lead to abnormal lipid accumulation in MGECs and induce ductal hyperkeratinization, contributing to meibomian gland dysfunction [55, 56]. Therefore, IL-17 pathway, which had the highest enrichment degree was further explored (Fig. 7f). The expression of downstream genes of IL-17 pathway [57], including chemokines (CXCL1/2/3/5/8), cytokines (IL6, TNF, PTGS2, CSF2/3), and tissue remodeling regulators (MMP1/3/9), was downregulated in MGECs cultured within bGA-ADM. The expression levels of key genes (MMP9, IL6, TNF, CXCL2, CXCL3, and CXCL8) in IL-17 pathway were further verified by qRT-PCR, which demonstrated a statistically significant decreasing trend in line with the transcriptomic findings (Fig. 7g). Overall, the suppression of inflammation-related gene expression in MGECs cultured in bGA-ADM compared with ADM suggests that bGA-ADM can create an environment conducive to constructing meibomian gland analogues through inflammatory regulation (Fig. 7h).

Compared with clinically-used materials such as ear cartilage, allogeneic sclera, and hard palate, bGA-ADM supports the adhesion, proliferation, and migration of epithelial cells, thereby promoting “self-conjunctivalization.” This reduces the need for additional conjunctival grafts, and decreases the risks of epithelial keratinization, simplifying surgical procedures and improving postoperative outcomes. Importantly, bGA-ADM is readily available due to its abundant source. Additionally, compared with synthetic scaffolds primarily composed of polymers such as poly(3-hydroxybutyrate-co-3-hydroxyhexanoate) [58], bGA-ADM offers superior biocompatibility and lower immunogenicity. However, its long-term repairing effects and therapeutic safety remain to be fully elucidated. Although a production system for bGA-ADM has been established, substantial work is still required before clinical translation. Optimization of the preparation process is necessary to improve productivity, and the underlying mechanisms must be more comprehensively investigated to better understand the regenerative process. Moreover, because the rabbit model cannot fully recapitulate human eyelid anatomy and function, additional animal models are warranted to further validate the practicality of bGA-ADM. Overall, bGA-ADM, a novel variant of traditional sheet-like ADM, demonstrates unparalleled advantages in repairing large posterior lamellar defects of the eyelids, and holds considerable promise for future clinical application.

Conclusion

bGA-ADM, an ADM-derived bilamellar bionic scaffold, is fabricated through a series of processes including homogenization, freezing, freeze-drying, heat-vacuum crosslinking, and glutaraldehyde crosslinking. It features a paper-like basement membrane layer and a spongy dermal layer, and is characterized by good hydrophilicity, high porosity, biomimetic mechanical strength, low immunogenicity, and negligible cytotoxicity. Owning to its heterogenous structure, bGA-ADM supports adhesion, proliferation, and functional activity of fibroblasts, meibomian gland and endothelial cells within the dermal layer, as well as epithelial cells on the basement membrane surface. Whether positioned ectopically in the ear or placed in situ in the posterior lamella of the eyelid, bGA-ADM demonstrates extensive cell infiltration, rapid vascularization, and effective integration with surrounding tissues. In this study, a novel rabbit model of a large posterior lamellar defect involving the eyelid margin was successfully established using an external skin incision. This model represents the first of its kind for large-size posterior lamellar defects and provides a valuable platform for further research. When used to repair of the posterior lamellar defects, bGA-ADM effectively restores the continuity of the eyelid margin, provides adequate structural support, and guides conjunctival regeneration and matrix deposition, thereby achieving both structural and functional reconstruction. Furthermore, RNA-seq and qRT-PCR analyses revealed that bGA-ADM reduces the expression of inflammatory genes in MGECs compared with traditional ADM, providing mechanism evidence that bGA-ADM can create a favorable microenvironment for the construction of meibomian gland analogues. Collectively, bGA-ADM presents a promising biomaterial with broad clinical application prospects for the repair of the posterior lamellar defects of the eyelid.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1^{(9.4MB, docx)}

Supplementary Material 2^{(12MB, zip)}

Author contributions

Xing Huang: Methodology, Data curation, Investigation, Animal Experiment, Writing-original draft, Writing-review & editing. Lin Lu: Data curation, Animal Experiment, Investigation. Yi Ding: Date curation, Animal Experiment, Investigation. Mengling Chang: Date curation, Investigation. Xiao Liang: Date curation, Investigation. Zhaoqi Yuan: Date curation, Investigation. Feixue Ding: Date curation, Investigation. Peiyi Li: Date curation, Investigation. Rui Jin: Animal Experiment, Conceptualization, Writing-review & editing. Xusong Luo: Animal Experiment, Conceptualization, Funding acquisition, Writing-review & editing. All authors reviewed and approved the paper.

Funding

The study is sponsored by National Natural Science Foundation of China (No. 82372536), Shanghai Municipal Key Clinical Specialty, China (No. shslczdzk00901), the project of Biobank from Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine (YBK202502), and Wuxi Taihu Lake Talent Plan, Supports for Leading Talents in Medical and Health Profession.

Data availability

Date will be made available on request.

Declarations

Ethics approval and consent to participate

Tissues were harvested from patients who provided written informed consent with ethics approval from the Ethics Committee (No.SH9H-2020-T341-2). All animal protocols were permitted by the Ethics Committee of Shanghai Ninth People’s Hospital (No.SH9H-2023-A886-1) and performed in accordance with the guidelines of the National Institutes of Health Guide for the Care and Use of laboratory animals.

Conflict of interest

The authors have no financial interest to declare in relation to the content of this article.

Footnotes

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Xing Huang, Lin Lu and Yi Ding co-first authors and contributed equally to this Work.

Contributor Information

Rui Jin, Email: dr.jinrui@hotmail.com.

Xusong Luo, Email: luoxs71@126.com.

References

1.Walker MK, Schornack MM, Vincent SJ. Anatomical and physiological considerations in scleral lens wear: eyelids and tear film. Cont Lens Anterior Eye. 2021;44(5):101407. [DOI] [PubMed] [Google Scholar]
2.Di Girolamo N, Park M. Cell identity changes in ocular surface Epithelia. Prog Retin Eye Res. 2023;95:101148. [DOI] [PubMed] [Google Scholar]
3.Yan Y, Fu R, Ji Q, Liu C, Yang J, Yin X, et al. Surgical strategies for eyelid defect reconstruction: a review on principles and techniques. Ophthalmol Ther. 2022;11(4):1383–408. [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Huang X, Ding Y, Lu L, Jin R, Di S, Yang J, et al. Biomaterials for tarsal plate reconstruction and our innovative work. Chin J Plast Reconstructive Surg. 2021;3(3):150–54. [Google Scholar]
5.Jirsova K, Jones GLA. Amniotic membrane in ophthalmology: properties, preparation, storage and indications for grafting-a review. Cell Tissue Bank. 2017;18(2):193–204. [DOI] [PubMed] [Google Scholar]
6.Sabater-Cruz N, Figueras-Roca M, González Ventosa A, Padró-Pitarch L, Tort J, Casaroli-Marano RP. Current clinical application of sclera and amniotic membrane for ocular tissue bio-replacement. Cell Tissue Bank. 2020;21(4):597–603. [DOI] [PubMed] [Google Scholar]
7.Gao Q, Hu B, Ning Q, Ye C, Xie J, Ye J, et al. A primary study of poly(propylene fumarate)-2-hydroxyethyl methacrylate copolymer scaffolds for tarsal plate repair and reconstruction in rabbit eyelids. J Mater Chem B. 2015;3(19):4052–62. [DOI] [PubMed] [Google Scholar]
8.Xu P, Chen P, Gao Q, Sun Y, Cao J, Wu H, et al. Azithromycin-carrying and microtubule-orientated biomimetic poly (lactic-co-glycolic acid) scaffolds for eyelid reconstruction. Front Med (Lausanne). 2023;10:1129606. [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Chen L, Yan D, Wu N, Zhang W, Yan C, Yao Q, et al. 3D-Printed poly-caprolactone scaffolds modified with biomimetic extracellular matrices for tarsal plate tissue engineering. Front Bioeng Biotechnol. 2020;8:219. [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Lee H, Park YH, Kang HJ, Lee H. Three-dimensional bioprinting of tarsal plates with adipose-derived mesenchymal stem cells: evaluation of meibomian gland reconstruction in a rat model. Biomedicines. 2024;12(11). [DOI] [PMC free article] [PubMed]
11.Xu P, Gao Q, Feng X, Lou L, Zhu T, Gao C, et al. A biomimetic tarso-conjunctival biphasic scaffold for eyelid reconstruction in vivo. Biomater Sci. 2019;7(8):3373–85. [DOI] [PubMed] [Google Scholar]
12.Cottler PS, Olenczak JB, Ning B, Seaman SA, Thuman JM, Sun N, et al. Fenestration improves acellular dermal matrix biointegration: an investigation of revascularization with photoacoustic microscopy. Plast Reconstr Surg. 2019;143(4):971–81. [DOI] [PubMed] [Google Scholar]
13.Rehman U, Shemie M, Sarwar MS, Adebayo O, Brennan PA. Use of biomaterials in the reconstruction of posterior lamellar eyelid defects: a systematic review and meta-analysis. Br J Oral Maxillofac Surg. 2023;61(7):464–74. [DOI] [PubMed] [Google Scholar]
14.Huang X, Ding Y, Pan W, Lu L, Jin R, Liang X, et al. A Comparative study on two types of porcine acellular dermal matrix sponges prepared by thermal crosslinking and thermal-glutaraldehyde crosslinking matrix microparticles. Front Bioeng Biotechnol. 2022;10:938798. [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Binder H, Hoffman L, Zak L, Tiefenboeck T, Aldrian S, Albrecht C. Clinical evaluation after matrix-associated autologous chondrocyte transplantation: a comparison of four different graft types. Bone Joint Res. 2021;10(7):370–79. [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Dunn RM. Cross-linking in biomaterials: a primer for clinicians. Plast Reconstr Surg. 2012;130(5 Suppl 2):18s–26s. [DOI] [PubMed] [Google Scholar]
17.Gu L, Shan T, Ma YX, Tay FR, Niu L. Novel biomedical applications of crosslinked collagen. Trends Biotechnol. 2019;37(5):464–91. [DOI] [PubMed] [Google Scholar]
18.Huang X, Zhu Z, Lu L, Jin R, Sun D, Luo X. Frozen bean curd-inspired xenogeneic acellular dermal matrix with triple pretreatment approach of freeze-thaw, laser drilling and ADSCs pre-culture for promoting early vascularization and integration. Regen Biomater. 2022;9:rbac053. [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Zhang Y, Zhang M, Cheng D, Xu S, Du C, Xie L, et al. Applications of electrospun scaffolds with enlarged pores in tissue engineering. Biomater Sci. 2022;10(6):1423–47. [DOI] [PubMed] [Google Scholar]
20.Deng W, Yang X, Yu J, Omari-Siaw E, Xu X. Recent advances of physiochemical cues on surfaces for directing cell fates. Colloids Surf B Biointerfaces. 2025;250:114550. [DOI] [PubMed] [Google Scholar]
21.Liu K, He Y, Lu F. Research progress on the immunogenicity and regeneration of acellular adipose matrix: a mini review. Front Bioeng Biotechnol. 2022;10:881523. [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Ji Y, Yang X, Ji Z, Zhu L, Ma N, Chen D, et al. DFT-Calculated IR spectrum amide I, II, and III band contributions of N-methylacetamide fine components. ACS Omega. 2020;5(15):8572–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Zheng X, Chen Y, Dan N, Dan W, Li Z. Highly stable collagen scaffolds crosslinked with an epoxidized natural polysaccharide for wound healing. Int J Biol Macromol. 2021;182:1994–2002. [DOI] [PubMed] [Google Scholar]
24.Meng C, Liu X, Li R, Malekmohammadi S, Feng Y, Song J, et al. 3D Poly (L-lactic acid) fibrous sponge with interconnected porous structure for bone tissue scaffold. Int J Biol Macromol. 2024;268(Pt 1):131688. [DOI] [PubMed] [Google Scholar]
25.Rnjak-Kovacina J, Wise SG, Li Z, Maitz PK, Young CJ, Wang Y, et al. Tailoring the porosity and pore size of electrospun synthetic human elastin scaffolds for dermal tissue engineering. Biomaterials. 2011;32(28):6729–36. [DOI] [PubMed] [Google Scholar]
26.Seymour AJ, Shin S, Heilshorn SC. 3D printing of microgel scaffolds with tunable void fraction to promote cell infiltration. Adv Healthc Mater. 2021;10(18):e2100644. [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Pais AI, Belinha J, Alves JL. Advances in computational techniques for bio-inspired cellular materials in the field of biomechanics: current trends and prospects. Materials (Basel). 2023;16(11). [DOI] [PMC free article] [PubMed]
28.Sun MT, Pham DT, O’Connor AJ, Wood J, Casson R, Selva D, et al. The Biomechanics of eyelid tarsus tissue. J Biomech. 2015;48(12):3455–59. [DOI] [PubMed] [Google Scholar]
29.Melchionna R, Trono P, Di Carlo A, Di Modugno F, Nisticò P. Transcription factors in fibroblast plasticity and caf heterogeneity. J Exp Clin Cancer Res. 2023;42(1):347. [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Pattnaik A, Sanket AS, Pradhan S, Sahoo R, Das S, Pany S, et al. Designing of gradient scaffolds and their applications in tissue regeneration. Biomaterials. 2023;296:122078. [DOI] [PubMed] [Google Scholar]
31.Zhou BY, Wang WB, Wu XL, Zhang WJ, Zhou GD, Gao Z, et al. Nintedanib inhibits keloid fibroblast functions by blocking the phosphorylation of multiple kinases and enhancing receptor internalization. Acta Pharmacol Sin. 2020;41(9):1234–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Cheng J, Chen Z, Lin D, Yang Y, Bai Y, Wang L, et al. A high clinically translatable strategy to anti-aging using hyaluronic acid and silk fibroin co-crosslinked hydrogels as dermal regenerative fillers. Acta Pharm Sin B. 2025;15(7):3767–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Alkazemi H, Mitchell GM, Lokmic-Tomkins Z, Heath DE, O’Connor AJ. Hierarchically vascularized and suturable tissue constructs created through angiogenesis from tissue-engineered vascular grafts. Acta Biomater. 2024;189:168–78. [DOI] [PubMed] [Google Scholar]
34.Li R, Liu K, Huang X, Li D, Ding J, Liu B, et al. Bioactive materials promote wound healing through modulation of cell behaviors. Adv Sci (Weinh). 2022;9(10):e2105152. [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Borrelli M, Geerling G, Spaniol K, Witt J. Eye socket regeneration and reconstruction. Curr Eye Res. 2020;45(3):253–64. [DOI] [PubMed] [Google Scholar]
36.Liu S, Hatton MP, Khandelwal P, Sullivan DA. Culture, immortalization, and characterization of human meibomian gland epithelial cells. Invest Ophthalmol Vis Sci. 2010;51(8):3993–4005. [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Brown SH, Kunnen CM, Duchoslav E, Dolla NK, Kelso MJ, Papas EB, et al. A comparison of patient matched meibum and tear lipidomes. Invest Ophthalmol Vis Sci. 2013;54(12):7417–24. [DOI] [PubMed] [Google Scholar]
38.Han X, Liu Y, Kam WR, Sullivan DA. Effect of brimonidine, an α2 adrenergic agonist, on human meibomian gland epithelial cells. Exp Eye Res. 2018;170:20–28. [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Hampel U, Schröder A, Mitchell T, Brown S, Snikeris P, Garreis F, et al. Serum-induced keratinization processes in an immortalized human meibomian gland epithelial cell line. PLoS One. 2015;10(6):e0128096. [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Qiu J, Li J, Wang G, Zheng L, Ren N, Liu H, et al. In vitro investigation on the biodegradability and biocompatibility of genipin cross-linked porcine acellular dermal matrix with intrinsic fluorescence. ACS Appl Mater Interface. 2013;5(2):344–50. [DOI] [PubMed] [Google Scholar]
41.Wen N, Qian E, Kang Y. Effects of macro-/micro-channels on vascularization and immune response of tissue engineering scaffolds. Cells. 2021;10(6). [DOI] [PMC free article] [PubMed]
42.Ostadi Y, Khanali J, Tehrani FA, Yazdanpanah G, Bahrami S, Niazi F, et al. Decellularized extracellular matrix scaffolds for soft tissue augmentation: from host-scaffold interactions to bottlenecks in clinical translation. Biomater Res. 2024;28:0071. [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Peng Y, Zhuang Y, Liu Y, Le H, Li D, Zhang M, et al. Bioinspired gradient scaffolds for osteochondral tissue engineering. Exploration (Beijing). 2023;3(4):20210043. [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Cao W, Zhang Y, Li L, Liu B, Ding J, Chen X. Physical cues of scaffolds promote peripheral nerve regeneration. Appl Phys Rev. 2024;11(2).
45.Rahmati M, Silva EA, Reseland JE, Ah C, Haugen HJ. Biological responses to physicochemical properties of biomaterial surface. Chem Soc Rev. 2020;49(15):5178–224. [DOI] [PubMed] [Google Scholar]
46.Huang W, Huang L, Li W, Saglam MS, Tourmouzis K, Goldstein SM, et al. Once-daily topical phosphosulindac is efficacious in the treatment of dry eye disease: studies in rabbit models of its main clinical subtypes. J Ocul Pharmacol Ther. 2022;38(1):102–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
47.Shi R, Jin M, Jin J, Xu L, Liu Z, Zheng L, et al. Retinoic acid promotes conjunctival epithelium differentiation and goblet cell regeneration: evidence from novel 3D conjunctival organoids and whole-mount PAS staining. Ocul Surf. 2025;37:301–13. [DOI] [PubMed] [Google Scholar]
48.Xu L, Wang G, Shi R, Zeng B, Zhang Y, Liu Z, et al. A cocktail of small molecules maintains the stemness and differentiation potential of conjunctival epithelial cells. Ocul Surf. 2023;30:107–18. [DOI] [PubMed] [Google Scholar]
49.Brown M, Li J, Moraes C, Tabrizian M, Li-Jessen NYK. Decellularized extracellular matrix: new promising and challenging biomaterials for regenerative medicine. Biomaterials. 2022;289:121786. [DOI] [PubMed] [Google Scholar]
50.Zhang Y, Xu Y, Kong H, Zhang J, Chan HF, Wang J, et al. Microneedle system for tissue engineering and regenerative medicine. Exploration (Beijing). 2023;3(1):20210170. [DOI] [PMC free article] [PubMed] [Google Scholar]
51.Bu J, Guo Y, Wu Y, Zhang R, Zhuang J, Zhao J, et al. Models for meibomian gland dysfunction: in vivo and in vitro. Ocul Surf. 2024;32:154–65. [DOI] [PubMed] [Google Scholar]
52.Butovich IA, Wilkerson A. Dynamic changes in the Gene expression patterns and lipid profiles in the developing and maturing meibomian glands. Int J Mol Sci. 2022;23(14). [DOI] [PMC free article] [PubMed]
53.Phan MAT, Madigan MC, Stapleton F, Willcox M, Golebiowski B. Human meibomian gland epithelial cell culture models: current progress, challenges, and future directions. Ocul Surf. 2022;23:96–113. [DOI] [PubMed] [Google Scholar]
54.Liu R, Tu M, Xue J, Xiao B, Li J, Liang L. Oleic acid induces lipogenesis and NLRP3 inflammasome activation in organotypic mouse meibomian gland and human meibomian gland epithelial cells. Exp Eye Res. 2024;241:109851. [DOI] [PubMed] [Google Scholar]
55.Liu R, Li J, Xu Y, Chen Z, Ye H, Tang J, et al. Melatonin attenuates LPS-Induced proinflammatory cytokine response and lipogenesis in human meibomian gland epithelial cells via MAPK/NF-κB pathway. Invest Ophthalmol Vis Sci. 2022;63(5):6. [DOI] [PMC free article] [PubMed] [Google Scholar]
56.Yamaguchi T. Inflammatory response in dry eye. Invest Ophthalmol Vis Sci. 2018;59(14):Des192–Des9. [DOI] [PubMed] [Google Scholar]
57.Kang MH, Kim MK, Lee HJ, Lee HI, Wee WR, Lee JH. Interleukin-17 in various ocular surface inflammatory diseases. J Korean Med Sci. 2011;26(7):938–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Zhou J, Peng SW, Wang YY, Zheng SB, Wang Y, Chen GQ. The use of poly(3-hydroxybutyrate-co-3-hydroxyhexanoate) scaffolds for tarsal repair in eyelid reconstruction in the rat. Biomaterials. 2010;31(29):7512–18. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Material 1^{(9.4MB, docx)}

Supplementary Material 2^{(12MB, zip)}

Data Availability Statement

Date will be made available on request.

[CR1] 1.Walker MK, Schornack MM, Vincent SJ. Anatomical and physiological considerations in scleral lens wear: eyelids and tear film. Cont Lens Anterior Eye. 2021;44(5):101407. [DOI] [PubMed] [Google Scholar]

[CR2] 2.Di Girolamo N, Park M. Cell identity changes in ocular surface Epithelia. Prog Retin Eye Res. 2023;95:101148. [DOI] [PubMed] [Google Scholar]

[CR3] 3.Yan Y, Fu R, Ji Q, Liu C, Yang J, Yin X, et al. Surgical strategies for eyelid defect reconstruction: a review on principles and techniques. Ophthalmol Ther. 2022;11(4):1383–408. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR4] 4.Huang X, Ding Y, Lu L, Jin R, Di S, Yang J, et al. Biomaterials for tarsal plate reconstruction and our innovative work. Chin J Plast Reconstructive Surg. 2021;3(3):150–54. [Google Scholar]

[CR5] 5.Jirsova K, Jones GLA. Amniotic membrane in ophthalmology: properties, preparation, storage and indications for grafting-a review. Cell Tissue Bank. 2017;18(2):193–204. [DOI] [PubMed] [Google Scholar]

[CR6] 6.Sabater-Cruz N, Figueras-Roca M, González Ventosa A, Padró-Pitarch L, Tort J, Casaroli-Marano RP. Current clinical application of sclera and amniotic membrane for ocular tissue bio-replacement. Cell Tissue Bank. 2020;21(4):597–603. [DOI] [PubMed] [Google Scholar]

[CR7] 7.Gao Q, Hu B, Ning Q, Ye C, Xie J, Ye J, et al. A primary study of poly(propylene fumarate)-2-hydroxyethyl methacrylate copolymer scaffolds for tarsal plate repair and reconstruction in rabbit eyelids. J Mater Chem B. 2015;3(19):4052–62. [DOI] [PubMed] [Google Scholar]

[CR8] 8.Xu P, Chen P, Gao Q, Sun Y, Cao J, Wu H, et al. Azithromycin-carrying and microtubule-orientated biomimetic poly (lactic-co-glycolic acid) scaffolds for eyelid reconstruction. Front Med (Lausanne). 2023;10:1129606. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR9] 9.Chen L, Yan D, Wu N, Zhang W, Yan C, Yao Q, et al. 3D-Printed poly-caprolactone scaffolds modified with biomimetic extracellular matrices for tarsal plate tissue engineering. Front Bioeng Biotechnol. 2020;8:219. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR10] 10.Lee H, Park YH, Kang HJ, Lee H. Three-dimensional bioprinting of tarsal plates with adipose-derived mesenchymal stem cells: evaluation of meibomian gland reconstruction in a rat model. Biomedicines. 2024;12(11). [DOI] [PMC free article] [PubMed]

[CR11] 11.Xu P, Gao Q, Feng X, Lou L, Zhu T, Gao C, et al. A biomimetic tarso-conjunctival biphasic scaffold for eyelid reconstruction in vivo. Biomater Sci. 2019;7(8):3373–85. [DOI] [PubMed] [Google Scholar]

[CR12] 12.Cottler PS, Olenczak JB, Ning B, Seaman SA, Thuman JM, Sun N, et al. Fenestration improves acellular dermal matrix biointegration: an investigation of revascularization with photoacoustic microscopy. Plast Reconstr Surg. 2019;143(4):971–81. [DOI] [PubMed] [Google Scholar]

[CR13] 13.Rehman U, Shemie M, Sarwar MS, Adebayo O, Brennan PA. Use of biomaterials in the reconstruction of posterior lamellar eyelid defects: a systematic review and meta-analysis. Br J Oral Maxillofac Surg. 2023;61(7):464–74. [DOI] [PubMed] [Google Scholar]

[CR14] 14.Huang X, Ding Y, Pan W, Lu L, Jin R, Liang X, et al. A Comparative study on two types of porcine acellular dermal matrix sponges prepared by thermal crosslinking and thermal-glutaraldehyde crosslinking matrix microparticles. Front Bioeng Biotechnol. 2022;10:938798. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR15] 15.Binder H, Hoffman L, Zak L, Tiefenboeck T, Aldrian S, Albrecht C. Clinical evaluation after matrix-associated autologous chondrocyte transplantation: a comparison of four different graft types. Bone Joint Res. 2021;10(7):370–79. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR16] 16.Dunn RM. Cross-linking in biomaterials: a primer for clinicians. Plast Reconstr Surg. 2012;130(5 Suppl 2):18s–26s. [DOI] [PubMed] [Google Scholar]

[CR17] 17.Gu L, Shan T, Ma YX, Tay FR, Niu L. Novel biomedical applications of crosslinked collagen. Trends Biotechnol. 2019;37(5):464–91. [DOI] [PubMed] [Google Scholar]

[CR18] 18.Huang X, Zhu Z, Lu L, Jin R, Sun D, Luo X. Frozen bean curd-inspired xenogeneic acellular dermal matrix with triple pretreatment approach of freeze-thaw, laser drilling and ADSCs pre-culture for promoting early vascularization and integration. Regen Biomater. 2022;9:rbac053. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR19] 19.Zhang Y, Zhang M, Cheng D, Xu S, Du C, Xie L, et al. Applications of electrospun scaffolds with enlarged pores in tissue engineering. Biomater Sci. 2022;10(6):1423–47. [DOI] [PubMed] [Google Scholar]

[CR20] 20.Deng W, Yang X, Yu J, Omari-Siaw E, Xu X. Recent advances of physiochemical cues on surfaces for directing cell fates. Colloids Surf B Biointerfaces. 2025;250:114550. [DOI] [PubMed] [Google Scholar]

[CR21] 21.Liu K, He Y, Lu F. Research progress on the immunogenicity and regeneration of acellular adipose matrix: a mini review. Front Bioeng Biotechnol. 2022;10:881523. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR22] 22.Ji Y, Yang X, Ji Z, Zhu L, Ma N, Chen D, et al. DFT-Calculated IR spectrum amide I, II, and III band contributions of N-methylacetamide fine components. ACS Omega. 2020;5(15):8572–78. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR23] 23.Zheng X, Chen Y, Dan N, Dan W, Li Z. Highly stable collagen scaffolds crosslinked with an epoxidized natural polysaccharide for wound healing. Int J Biol Macromol. 2021;182:1994–2002. [DOI] [PubMed] [Google Scholar]

[CR24] 24.Meng C, Liu X, Li R, Malekmohammadi S, Feng Y, Song J, et al. 3D Poly (L-lactic acid) fibrous sponge with interconnected porous structure for bone tissue scaffold. Int J Biol Macromol. 2024;268(Pt 1):131688. [DOI] [PubMed] [Google Scholar]

[CR25] 25.Rnjak-Kovacina J, Wise SG, Li Z, Maitz PK, Young CJ, Wang Y, et al. Tailoring the porosity and pore size of electrospun synthetic human elastin scaffolds for dermal tissue engineering. Biomaterials. 2011;32(28):6729–36. [DOI] [PubMed] [Google Scholar]

[CR26] 26.Seymour AJ, Shin S, Heilshorn SC. 3D printing of microgel scaffolds with tunable void fraction to promote cell infiltration. Adv Healthc Mater. 2021;10(18):e2100644. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR27] 27.Pais AI, Belinha J, Alves JL. Advances in computational techniques for bio-inspired cellular materials in the field of biomechanics: current trends and prospects. Materials (Basel). 2023;16(11). [DOI] [PMC free article] [PubMed]

[CR28] 28.Sun MT, Pham DT, O’Connor AJ, Wood J, Casson R, Selva D, et al. The Biomechanics of eyelid tarsus tissue. J Biomech. 2015;48(12):3455–59. [DOI] [PubMed] [Google Scholar]

[CR29] 29.Melchionna R, Trono P, Di Carlo A, Di Modugno F, Nisticò P. Transcription factors in fibroblast plasticity and caf heterogeneity. J Exp Clin Cancer Res. 2023;42(1):347. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR30] 30.Pattnaik A, Sanket AS, Pradhan S, Sahoo R, Das S, Pany S, et al. Designing of gradient scaffolds and their applications in tissue regeneration. Biomaterials. 2023;296:122078. [DOI] [PubMed] [Google Scholar]

[CR31] 31.Zhou BY, Wang WB, Wu XL, Zhang WJ, Zhou GD, Gao Z, et al. Nintedanib inhibits keloid fibroblast functions by blocking the phosphorylation of multiple kinases and enhancing receptor internalization. Acta Pharmacol Sin. 2020;41(9):1234–45. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR32] 32.Cheng J, Chen Z, Lin D, Yang Y, Bai Y, Wang L, et al. A high clinically translatable strategy to anti-aging using hyaluronic acid and silk fibroin co-crosslinked hydrogels as dermal regenerative fillers. Acta Pharm Sin B. 2025;15(7):3767–87. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR33] 33.Alkazemi H, Mitchell GM, Lokmic-Tomkins Z, Heath DE, O’Connor AJ. Hierarchically vascularized and suturable tissue constructs created through angiogenesis from tissue-engineered vascular grafts. Acta Biomater. 2024;189:168–78. [DOI] [PubMed] [Google Scholar]

[CR34] 34.Li R, Liu K, Huang X, Li D, Ding J, Liu B, et al. Bioactive materials promote wound healing through modulation of cell behaviors. Adv Sci (Weinh). 2022;9(10):e2105152. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR35] 35.Borrelli M, Geerling G, Spaniol K, Witt J. Eye socket regeneration and reconstruction. Curr Eye Res. 2020;45(3):253–64. [DOI] [PubMed] [Google Scholar]

[CR36] 36.Liu S, Hatton MP, Khandelwal P, Sullivan DA. Culture, immortalization, and characterization of human meibomian gland epithelial cells. Invest Ophthalmol Vis Sci. 2010;51(8):3993–4005. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR37] 37.Brown SH, Kunnen CM, Duchoslav E, Dolla NK, Kelso MJ, Papas EB, et al. A comparison of patient matched meibum and tear lipidomes. Invest Ophthalmol Vis Sci. 2013;54(12):7417–24. [DOI] [PubMed] [Google Scholar]

[CR38] 38.Han X, Liu Y, Kam WR, Sullivan DA. Effect of brimonidine, an α2 adrenergic agonist, on human meibomian gland epithelial cells. Exp Eye Res. 2018;170:20–28. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR39] 39.Hampel U, Schröder A, Mitchell T, Brown S, Snikeris P, Garreis F, et al. Serum-induced keratinization processes in an immortalized human meibomian gland epithelial cell line. PLoS One. 2015;10(6):e0128096. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR40] 40.Qiu J, Li J, Wang G, Zheng L, Ren N, Liu H, et al. In vitro investigation on the biodegradability and biocompatibility of genipin cross-linked porcine acellular dermal matrix with intrinsic fluorescence. ACS Appl Mater Interface. 2013;5(2):344–50. [DOI] [PubMed] [Google Scholar]

[CR41] 41.Wen N, Qian E, Kang Y. Effects of macro-/micro-channels on vascularization and immune response of tissue engineering scaffolds. Cells. 2021;10(6). [DOI] [PMC free article] [PubMed]

[CR42] 42.Ostadi Y, Khanali J, Tehrani FA, Yazdanpanah G, Bahrami S, Niazi F, et al. Decellularized extracellular matrix scaffolds for soft tissue augmentation: from host-scaffold interactions to bottlenecks in clinical translation. Biomater Res. 2024;28:0071. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR43] 43.Peng Y, Zhuang Y, Liu Y, Le H, Li D, Zhang M, et al. Bioinspired gradient scaffolds for osteochondral tissue engineering. Exploration (Beijing). 2023;3(4):20210043. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR44] 44.Cao W, Zhang Y, Li L, Liu B, Ding J, Chen X. Physical cues of scaffolds promote peripheral nerve regeneration. Appl Phys Rev. 2024;11(2).

[CR45] 45.Rahmati M, Silva EA, Reseland JE, Ah C, Haugen HJ. Biological responses to physicochemical properties of biomaterial surface. Chem Soc Rev. 2020;49(15):5178–224. [DOI] [PubMed] [Google Scholar]

[CR46] 46.Huang W, Huang L, Li W, Saglam MS, Tourmouzis K, Goldstein SM, et al. Once-daily topical phosphosulindac is efficacious in the treatment of dry eye disease: studies in rabbit models of its main clinical subtypes. J Ocul Pharmacol Ther. 2022;38(1):102–13. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR47] 47.Shi R, Jin M, Jin J, Xu L, Liu Z, Zheng L, et al. Retinoic acid promotes conjunctival epithelium differentiation and goblet cell regeneration: evidence from novel 3D conjunctival organoids and whole-mount PAS staining. Ocul Surf. 2025;37:301–13. [DOI] [PubMed] [Google Scholar]

[CR48] 48.Xu L, Wang G, Shi R, Zeng B, Zhang Y, Liu Z, et al. A cocktail of small molecules maintains the stemness and differentiation potential of conjunctival epithelial cells. Ocul Surf. 2023;30:107–18. [DOI] [PubMed] [Google Scholar]

[CR49] 49.Brown M, Li J, Moraes C, Tabrizian M, Li-Jessen NYK. Decellularized extracellular matrix: new promising and challenging biomaterials for regenerative medicine. Biomaterials. 2022;289:121786. [DOI] [PubMed] [Google Scholar]

[CR50] 50.Zhang Y, Xu Y, Kong H, Zhang J, Chan HF, Wang J, et al. Microneedle system for tissue engineering and regenerative medicine. Exploration (Beijing). 2023;3(1):20210170. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR51] 51.Bu J, Guo Y, Wu Y, Zhang R, Zhuang J, Zhao J, et al. Models for meibomian gland dysfunction: in vivo and in vitro. Ocul Surf. 2024;32:154–65. [DOI] [PubMed] [Google Scholar]

[CR52] 52.Butovich IA, Wilkerson A. Dynamic changes in the Gene expression patterns and lipid profiles in the developing and maturing meibomian glands. Int J Mol Sci. 2022;23(14). [DOI] [PMC free article] [PubMed]

[CR53] 53.Phan MAT, Madigan MC, Stapleton F, Willcox M, Golebiowski B. Human meibomian gland epithelial cell culture models: current progress, challenges, and future directions. Ocul Surf. 2022;23:96–113. [DOI] [PubMed] [Google Scholar]

[CR54] 54.Liu R, Tu M, Xue J, Xiao B, Li J, Liang L. Oleic acid induces lipogenesis and NLRP3 inflammasome activation in organotypic mouse meibomian gland and human meibomian gland epithelial cells. Exp Eye Res. 2024;241:109851. [DOI] [PubMed] [Google Scholar]

[CR55] 55.Liu R, Li J, Xu Y, Chen Z, Ye H, Tang J, et al. Melatonin attenuates LPS-Induced proinflammatory cytokine response and lipogenesis in human meibomian gland epithelial cells via MAPK/NF-κB pathway. Invest Ophthalmol Vis Sci. 2022;63(5):6. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR56] 56.Yamaguchi T. Inflammatory response in dry eye. Invest Ophthalmol Vis Sci. 2018;59(14):Des192–Des9. [DOI] [PubMed] [Google Scholar]

[CR57] 57.Kang MH, Kim MK, Lee HJ, Lee HI, Wee WR, Lee JH. Interleukin-17 in various ocular surface inflammatory diseases. J Korean Med Sci. 2011;26(7):938–44. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR58] 58.Zhou J, Peng SW, Wang YY, Zheng SB, Wang Y, Chen GQ. The use of poly(3-hydroxybutyrate-co-3-hydroxyhexanoate) scaffolds for tarsal repair in eyelid reconstruction in the rat. Biomaterials. 2010;31(29):7512–18. [DOI] [PubMed] [Google Scholar]

PERMALINK

Heterogeneous-structural bilamellar bionic scaffold for integrated regeneration of tarso-conjunctival composite tissue

Xing Huang

Lin Lu

Yi Ding

Mengling Chang

Xiao Liang

Zhaoqi Yuan

Feixue Ding

Peiyi Li

Rui Jin

Xusong Luo

Abstract

Background

Methods

Results

Conclusion

Graphical Abstract

Supplementary information

Introduction

Fig. 1.

Materials and methods

Material preparation

Characterization of physicochemical properties

Isolation, culture, and identification of fibroblasts

Culture and expansion of oral keratinocytes

Culture and expansion of human umbilical vein endothelial cells

Isolation, culture, and identification of meibomian gland epithelial cells

Toxicity and cytocompatibility of materials

Adhesion of varied cells on materials

Function of varied cells on materials

Quantitative real-time polymerase chain reaction (qRT-PCR)

Table 1.

Ear implantation model of rabbit

Eyelid implantation model of rabbit

Defect of posterior lamella of rabbit eyelid model and bGA-ADM repair of large-sized defect in situ

RNA sequencing and qRT-PCR

Statistical analysis

Results and discussion

The physicochemical properties of bGA-ADM

Fig. 2.

The bGA-ADM possesses good biocompatibility with Fbs and HUVECs

Fig. 3.

bGA-ADM supports the growth and function of HOKs and MGECs

Fig. 4.

bGA-ADM exhibits fast vascularization and adaptive regeneration in vivo

Fig. 5.

The bGA-ADM repairs the large posterior lamellar defects of the eyelid in situ and returns satisfactory results in the early stage

Fig. 6.

The potential for bGA-ADM to construct meibomian gland

Fig. 7.

Conclusion

Electronic supplementary material

Author contributions

Funding

Data availability

Declarations

Ethics approval and consent to participate

Conflict of interest

Footnotes

Contributor Information

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases