Synergistic Effects of Riboflavin/UVA Cross-Linking, N-Acetylcysteine, and Polydatin on Collagenase Resistance in Human Corneal and Scleral Tissue

1. Introduction

Corneal collagen crosslinking (CXL) using riboflavin and ultraviolet-A (UVA) irradiation has become an established treatment for progressive keratoconus and corneal ectatic disorders.[1] By inducing the formation of reactive oxygen species, riboflavin/UVA irradiation promotes the development of additional covalent bonds between stromal collagen fibers, resulting in increased biomechanical rigidity and stabilization of the corneal extracellular matrix. [2] Clinical studies have consistently demonstrated the efficacy of CXL in slowing or halting ectatic progression and preserving visual function in keratoconus (KC) patients.[3] Beyond its biomechanical effects, increasing evidence suggests that CXL also exerts important biochemical effects on corneal tissue. [4] Spoerl et al. demonstrated that riboflavin/UVA crosslinking significantly increases resistance of corneal collagen to pepsin, trypsin, and collagenase digestion. [5]

Increased activity of matrix metalloproteinases, collagenases and extracellular matrix degradation are known to contribute to progressive stromal weakening in keratoconus and corneal melting disorders [5]. Growing attention has recently been directed toward adjunctive biochemical strategies capable of modulating extracellular matrix degradation and oxidative stress pathways. N-acetylcysteine (NAC), a thiol-containing antioxidant and precursor of glutathione, has demonstrated anti-inflammatory and antiproteolytic properties, including modulation of matrix metalloproteinase activity. [6,7,8]

Similarly, Polydatin, a natural glucoside precursor of resveratrol extracted from Polygonum cuspidatum (Pc) , exhibits antioxidant and extracellular matrix-protective effects in several experimental models.[9] Such compounds may therefore enhance tissue resistance to enzymatic degradation and potentially synergize with the stabilizing effect induced by collagen crosslinking.[10] Although the clinical role of corneal crosslinking is now well established in KC and other corneal ectasia, analogous extracellular matrix stabilization strategies may also have potential relevance in scleral remodeling associated with progressive myopia.[11,12] Progressive scleral remodeling, collagen disorganization, scleral thinning and weakening of scleral collagen-extra-cellular- matrix (ECM) complex play a central role in myopia progression.[13,14] Progressive myopia is increasingly recognized as a disease characterized by collagen disorganization, protease activation, and biomechanical weakening of the scleral ECM, ultimately contributing to equatorial stretching, ocular expansion and axial elongation.[15] Experimental studies[16,17] have demonstrated the feasibility of scleral crosslinking and suggest that increasing scleral rigidity may potentially counteract progressive deformation of the ocular wall.[18,19,20] Human cadaver eyes scleral collagen cross-linking by ultraviolet activation of riboflavin differentially increases scleral Young Modulus (YM) more in the equatorial than posterior sclera, and most in the lateral, equatorial sclera.[21] Cross-linking might be used to arrest progressive myopia or to prevent staphyloma formation.[22]

The purpose of the present study was therefore to evaluate the resistance of human corneal buttons and scleral strips to collagenase-mediated enzymatic degradation[5] following Riboflavin/UVA CXL, either alone or combined with adjunctive NAC and Pc treatment. By investigating both corneal and scleral tissue stabilization within the same experimental model, this study aims to provide further insight into combined biomechanical and biochemical ECM stabilization strategies with potential translational implications for corneal ectatic disorders and progressive myopia.

2. Materials and Methods

2.1. Setting and Donor Tissues

This experimental laboratory study was conceived at the Siena International Crosslinking Center, Monteriggioni, Siena, Italy and conducted at the Biotecnica Associates Laboratories, Castelfidardo, Ancona, Italy, in collaboration with the Veneto Eye Bank Foundation, Venice, Italy and University of Enna Kore Ophthalmology Department. The study was authorized by the Institutional Review Board of the Siena Crosslinking Center (Sclero-Corneal Enzymatic Dygestion Study 1, Code: SCED-S1) under the tenets of Helsinky declaration and respecting good laboratory practice (GLP) rules.

Twenty-four human donor corneas unsuitable for transplantation, together with their scleral rims, were obtained from the Veneto Eye Bank Foundation, Venice, Italy, for experimental research purposes. Samples were stored at 5 ± 3°C in the storage solution (Optisol™, Bausch and Lomb Inc, Rochester, NY, USA) until the beginning of the study.

Twelve corneal buttons were punched using a 6 mm trephine to obtain homogeneous samples. Included corneal buttons measured 6 mm in diameter with an average central corneal thickness of 503 ± 27 µm. Central thickness was measured with ultrasound portable US pachymetry (Pachette 4™, DGH Technology, Inc Exton PA, USA).

Twelve scleral strips measuring 6 mm in length × 2 mm in width were dissected from the corresponding scleral rims.

2.2. Experimental Groups

Corneal buttons and scleral strips were divided into six experimental groups:

• Group A: untreated controls
• Group B: riboflavin/UVA CXL alone
• Group C: NAC/Pc incubation alone
• Group D: NAC/Pc incubation before CXL
• Group E: NAC/Pc incubation after CXL
• Group F: NAC/Pc incubation before and after CXL

Corneal buttons were divided into 6 groups (2 specimens for each group).

2.3. Crosslinking Procedure

Crosslinking (CXL) treatment consisted of 15 minutes of Riboflavin soaking followed by continuous UVA irradiation for 10 minutes at 9mW/cm², corresponding to a total Fluence of 5.4 J/cm². Isotonic dextran-free 0.1% Riboflavin solution was applied during UV-A irradiation every two minutes.[23]

NAC/Pc Solution

The adjunctive bioactive solution was prepared by dissolving a 3.7 g mixture containing NAC 400 mg and Polygonum cuspidatum (Pc), Polidatin dry extract 200 mg in 250 mL of deionized water. Samples assigned to adjunctive treatment groups were incubated in this solution 20 minutes before CXL, 20 minutes after CXL, or both pre and post CXL according to Group allocation.

2.4. Enzymatic Degradation Assay

After treatment, samples were incubated at 25 ± 2°C and 60 ± 5% relative humidity in sterile multi-well plates containing 2 mL collagenase solution prepared in phosphate-buffered saline at pH 7.5.

Samples were evaluated daily until complete degradation, Figure 1. Corneal button diameter and scleral strip length were monitored to assess degradation kinetics. Protein levels released into the incubation medium from corneal samples were assessed using Bradford’s method.[24]

2.5. Statistical Analysis

Data were expressed as mean ± standard deviation and 95% confidence interval. Differences among groups were analyzed using one-way analysis of variance followed by Tukey post-hoc testing. Separate analyses were performed for corneal and scleral samples. A p value < 0.05 was considered statistically significant.

3. Results

3.1. Corneal Enzymatic Degradation

Untreated human corneal buttons underwent complete collagenase-mediated degradation after 7 ± 2 days. Riboflavin/UVA crosslinking alone significantly prolonged degradation time to approximately 14 ± 3 days, corresponding to a two-fold increase in resistance to enzymatic degradation compared with untreated controls. Incubation with NAC/Pc alone resulted in a modest prolongation of corneal preservation, increasing degradation time by approximately 3 ± 2 days compared with control corneas; however, this difference did not reach statistical significance. No differences were found in scleral strips incubation with NAC/Pc alone. Similarly, pre-incubation with NAC/Pc before CXL demonstrated only a limited not statistically significant additive effect of approximately 2 additional days compared with CXL alone. The greatest anti-degradative effect was observed in tissues receiving NAC/Pc both before and after CXL treatment, in which corneal digestion time was further prolonged by approximately 7 days compared with CXL alone, reaching approximately 21 days overall.

One-way ANOVA demonstrated significant differences among corneal experimental groups (F = 31.04, p = 3.01 × 10⁻⁸). Post-hoc analysis confirmed that the major statistically significant differences were observed between untreated controls and crosslinked groups, and between CXL alone and combined NAC/polydatin pre/post-CXL treatment. Corneal degradation kinetics are illustrated in Figure 2 and Figure 3, and displayed in Table 1.

3.2. Scleral Enzymatic Degradation

Untreated scleral strips demonstrated greater intrinsic resistance to collagenase-mediated degradation compared with corneal tissue, with complete degradation occurring after approximately 16 ± 2 days. Riboflavin/UVA crosslinking alone significantly prolonged scleral preservation to approximately 24 ± 3 days. In contrast, incubation with NAC/polydatin alone did not produce significant differences in scleral degradation time compared with untreated controls. Additional incubation with NAC/Pc before CXL demonstrated only a mild additive effect, whereas combined NAC/Pc pre- and post-CXL treatment further prolonged scleral preservation by approximately 5 days compared with CXL alone, reaching approximately 29–30 days overall.

One-way ANOVA demonstrated significant differences among scleral experimental groups (F = 29.28, p = 4.77 × 10⁻⁸). Significant post-hoc differences were primarily observed between untreated controls and crosslinked tissues and between CXL alone and combined NAC/Pc pre/post-CXL treatment. Scleral degradation kinetics are shown in Figure 4 and Figure 5 and summarized in Table 1.

3.3. Protein Release Kinetics

Protein concentration within the incubation medium progressively increased during collagenase-mediated degradation and correlated with tissue dissolution and extracellular matrix breakdown. Untreated corneal samples demonstrated earlier and greater protein release into the incubation medium, consistent with accelerated collagen degradation. Crosslinked and adjunctively treated tissues showed delayed protein release kinetics, suggesting improved preservation of stromal architecture during collagenase incubation. The greatest difference in protein release was observed between untreated controls and the NAC/polydatin pre/post-CXL group, with an 84% reduction in protein concentration at day 5 (0.63 vs 0.10 µg/mL) and a 52% reduction at day 10 (1.23 vs 0.59 µg/mL). Protein release kinetics are illustrated in Figure 6.

Riboflavin/UVA CXL markedly prolonged tissue preservation in both corneal and scleral samples. NAC/polydatin incubation alone produced only a mild, non-significant prolongation of corneal degradation time and no significant effect in scleral strips compared with untreated controls. The greatest anti-degradative effect was observed after combined NAC/polydatin incubation before and after CXL. Values are expressed as mean ± standard deviation where applicable. NS = not significant; CXL = collagen crosslinking; NAC = N-acetylcysteine; Pc = Polygonum cuspidatum dry extract.

4. Discussion

The present study demonstrated that riboflavin/UVA collagen crosslinking (CXL) significantly enhances resistance of human corneal and scleral tissues to collagenase-mediated enzymatic degradation. Furthermore, adjunctive treatment with N-acetylcysteine (NAC) and Polydatin further potentiated this anti-degradative effect when combined with CXL, whereas NAC/Pc incubation alone produced only limited or non-significant effects. These findings support the concept that photochemical collagen stabilization represents the primary mechanism of corneal and scleral tissue preservation, while adjunctive biochemical modulation may enhance extracellular matrix stabilization when associated with crosslinking treatment.

The anti-enzymatic effect observed after corneal crosslinking is consistent with the experimental findings previously reported by Spoerl et al[5] and Aldahlawi et al[25] who demonstrated increased resistance of crosslinked corneal tissue to pepsin, trypsin, and collagenase digestion.[26] The present findings further expand this anti-degradative paradigm by demonstrating similar effects in human scleral tissue and by exploring adjunctive extracellular matrix-protective strategies based on NAC and polydatin treatment.

Untreated corneal buttons underwent complete degradation after approximately 7 days, whereas crosslinked corneas remained structurally preserved for approximately 14 days, confirming a two-fold increase in resistance following riboflavin/UVA treatment. The protective effect induced by crosslinking is likely multifactorial. Riboflavin/UVA irradiation generates reactive oxygen species leading to additional intermolecular and intramolecular collagen bonds within the extracellular matrix.[27,28] These structural modifications increase tissue stiffness and reduce accessibility of collagen cleavage sites to proteolytic enzymes.[29]

A major finding of the present study was the demonstration that adjunctive incubation with NAC[6] and Polydatin[10] further enhanced resistance to enzymatic degradation in both corneal and scleral samples. A protective effect was observed in samples receiving NAC/Pc treatment both before and after crosslinking, extending corneal degradation times up to approximately 21 days and scleral degradation times up to approximately 30 days.

N-acetylcysteine[7] is a well-known antioxidant and glutathione precursor with documented anti-inflammatory and antiproteolytic properties[8]. Similarly, Polydatin has demonstrated antioxidant and extracellular matrix-protective effects in several experimental models.[9] Although the precise molecular mechanisms were not specifically investigated in the present study, the observed prolongation of tissue preservation suggests a synergistic interaction between photochemical/photodynamic collagen stabilization and biochemical modulation of extracellular matrix degradation pathways.

Although the clinical role of corneal crosslinking is now well established in Keratoconus[2,4] and iatrogenic post laser vision correction corneal ectasia[30], supplemental extracellular matrix stabilization strategies may also have potential relevance in scleral remodeling associated with progressive myopia.[11] Progressive myopia is increasingly recognized as a disease characterized by collagen disorganization, protease activation, and biomechanical weakening of the scleral extracellular matrix, ultimately contributing to eye expansion and axial elongation.[12,13]

In the present study, untreated scleral strips demonstrated greater intrinsic resistance to collagenase-mediated degradation compared with corneal tissue, likely reflecting the denser collagen architecture and lower hydration of scleral connective tissue.[14] Nevertheless, Riboflavin/UVA crosslinking significantly prolonged scleral preservation agains enzymatic degradation, while adjunctive NAC/Polydatin exposure further enhanced resistance to extracellular matrix breakdown. These findings provide preliminary experimental evidence supporting the concept that combined biomechanical and biochemical stabilization strategies may reduce scleral ECM remodeling thus enhancing its preservation.

The protein release kinetics further supported the morphologic findings. Untreated corneal samples demonstrated earlier and greater protein release into the incubation medium, whereas crosslinked and adjunctively treated tissues showed delayed protein release kinetics, suggesting slower collagen breakdown and improved preservation of stromal ultrastructure.

Several limitations of the present study should be acknowledged. First, the study was performed in vitro and therefore cannot fully reproduce the complexity of in vivo ocular remodeling, inflammatory responses, and pharmacokinetics. Second, the collagenase digestion model represents an accelerated experimental system that may not directly reflect physiologic extracellular matrix turnover. Third, the precise contribution of direct collagenase inhibition versus tissue-level collagen stabilization induced by NAC and Polydatin was not specifically investigated.

Despite these limitations, the present study provides novel evidence that combined Riboflavin/UVA crosslinking increase scleral dygestion time and adjunctive NAC/Polydatin treatment significantly enhance resistance of human corneal and scleral tissues to collagenase-mediated degradation. These findings support further translational investigation of combined extracellular matrix stabilization strategies in corneal ectatic disorders and progressive myopia.

Interestingly, NAC/polydatin incubation alone produced only limited or non-significant effects, whereas its use in association with CXL, particularly in the pre/post-CXL protocol, further prolonged tissue preservation. This pattern suggests that the bioactive compounds may not independently induce substantial collagen stabilization, but may potentiate the anti-degradative effect of photochemical crosslinking in an exposure-dependent manner. Future studies testing different concentrations and incubation times are required to confirm a true dose–response relationship.

5. Conclusions

Riboflavin/UVA collagen crosslinking significantly enhances the resistance of human corneal and scleral tissues to collagenase-mediated enzymatic degradation. Adjunctive NAC and Polydatin treatment further potentiate this anti-degradative effect, suggesting a possible synergistic strategy for extracellular matrix stabilization. These findings support future translational investigation of combined biomechanical and biochemical approaches in corneal ectasia and progressive myopia, where scleral crosslinking may potentially emerge as a clinically relevant strategy for controlling disease progression. Future studies evaluating different NAC/Polydatin concentrations and incubation times are warranted to determine whether the observed protocol-dependent potentiation reflects a true dose–response relationship.