Application of Noggin-Coated Electrospun Scaffold in Corneal Wound Healing

Abstract

Purpose: The objective of this study is to develop and characterize electrospun corneal bandage infused with Noggin protein and evaluate its therapeutic potential in the treatment of superficial nonhealing corneal ulceration.

Methods: Electrospun nanofibrous scaffolds were created with different blend ratios of polycaprolactone and gelatin and coated with different concentrations of Noggin protein. Morphologic, mechanical, degradation, and surface chemistry of the developed scaffold was assessed. Biocompatibility of the developed scaffold with corneal epithelial cells was evaluated by looking at cell viability, proliferation, and immunostaining. In vitro wound healing in the presence of Noggin-coated scaffold was evaluated by measuring wound closure rate after scratch.

Results: Uniform nanofibrous scaffolds coated with Noggin were constructed through optimization of electrospinning parameters and demonstrated mechanical properties better than or similar to commercially available contact lenses used in corneal wound healing. In the presence of Noggin-coated scaffold, corneal epithelial cells showed higher proliferation and wound-healing rate.

Conclusions: This Noggin-coated electrospun scaffold represents a step toward, expanding treatment options for patients with indolent corneal ulcers.

Translational Relevance: In this study, the feasibility of Noggin-coated electrospun scaffold as a therapeutic for indolent corneal ulcer was evaluated. This study also provides a better perspective for understanding electrospun scaffolds as a tunable platform to infuse topical therapeutics and use as a corneal bandage.

Introduction

Superficial nonhealing (i.e., indolent) corneal ulceration (i.e., keratitis) is a painful condition, present in both humans and canines, which commonly requires prolonged and costly treatment. Recurrent corneal erosion syndrome (RCES) in humans and spontaneous chronic corneal epithelial defects (SCCEDs) in canines share a common set of symptoms, including redness, tearing, and cloudiness of the cornea.^1–5 Conservative treatment of either condition involves a progression of medical treatments such as artificial tears, antibiotics, and bandage contact lenses. Indolent ulcers, however, are often refractory to these treatments, necessitating surgical treatment. Debridement and diamond burr keratectomy are used to treat the separation between corneal epithelium and stroma associated with the disease by stimulating cell proliferation and adhesion. However, these surgical treatments are susceptible to microbial infection and pain associated with exposed nerve endings. Many cases are nonresponsive to topical treatments and have a recurrence of erosion.^6–11 A novel treatment method that would eliminate patient noncompliance, protect against microbial invasion, and encourage continued cellular proliferation and adhesion after surgery is needed to improve clinical outcomes. 

The morphologies of corneas experiencing SCCED and RCES tend to share similarities in the form of reduced or absent adhesion of the corneal epithelium to the underlying extracellular matrix and a deficient epithelial basement membrane.^1–5 It is hypothesized that these conditions may be a result of deficiencies in limbal epithelial progenitor cell (LEPC) proliferation and migration.^12–14 Therefore, effective treatments should seek to target molecular pathways involved in cellular proliferation and migration, such as the bone morphogenetic protein (BMP) or the Wnt pathway. This approach is supported by recent insight into SCCED development in Boxer dogs.¹⁵ Boxer dogs are known to be highly predisposed to the development of SCCEDs,^16–18 and a genetic study of affected Boxers showed a significant correlation in a defect of the NOG gene, resulting in a deficiency in functional Noggin protein, coincident with the development of SCCEDs.¹⁵ This finding is thought to be of particular importance as the Noggin protein is known to be a potent inhibitor of BMPs, and the BMP pathway, in combination with the Wnt pathway, is shown to affect the differentiation and growth of LEPC.¹⁹ 

With this information, to address the need for a treatment method that will improve indolent ulceration clinical outcomes, we aim to produce a biosynthetic Noggin-infused corneal bandage as a proof of concept. However, as a more solid understanding of the biochemical pathways in corneal wound healing and BMP's effect on such comes to light, the therapeutic to be infused in the bandage could be altered and customized. It is likely that there is a complex picture of crosstalk among pathways that control corneal wound healing, and treatments need to be tailored to the specific deficiencies of the patient responsible for their condition. 

There are a couple of important distinctions about indolent corneal ulcers and where our proposed solution may fit. Keratitis is used to describe a condition where there is an open sore on the cornea and there is inflammation present, but the epithelium tissue is typically intact.²⁰ On the other hand, indolent corneal ulcers are used to specifically describe circumstances in which the healing of an ulcer is slow, there may be recurrent injury, and it is believed that the epithelium tissue has an inability to adhere to the stroma layer.²¹ There can be confusion about indolent corneal ulcers when talking about humans versus canines as fungal-based ulcers in humans may be called “indolent,” but the context described in this article is fitting of the presentation of RCES.²² Most cases for keratitis tend to involve infectious causes, while an indolent ulcer can be spontaneous and is not generally believed to be caused by infection but rather has infections as a secondary complication.^20,21 With this study, we focus more on a possible solution for indolent corneal ulcers in an epithelial defect sense but are hopeful there is valuable insight to be gained for other cases of keratitis. 

An electrospun nanofibrous scaffold can be used to develop biodegradable corneal bandages. Polycaprolactone (PCL) is a synthetic biodegradable polymer with excellent mechanical properties and is blended with gelatin to enhance cell adherence and proliferation.^23–25 Electrospinning also allows easy modification of the corneal bandage due to solution properties and electrospinning parameters.^26–31 Using different polymers and blend ratios, the morphology and the mechanical properties of the scaffolds can be altered.³¹ 

Materials and Methods

Preparation of PCL-Gelatin Fibrous Scaffolds

PCL (Mn = 42,000 g/mol, 181609; Sigma-Aldrich, St. Louis, MO, USA) and Gelatin Type-A from Porcine Skin (gel strength ∼300 g Bloom, G2500; Sigma-Aldrich) were dissolved overnight in 1,1,2,3,3,3-hexafluoro-1-propene (HFP) (H0424; TCI America, Portland, OR, USA) by magnetic stirring to obtain an electrospinning solution at a 10% w/v concentration. The fibrous scaffolds were electrospun using electrospinning solutions at different PCL-gelatin ratios (2:1, 1:1, 1:2, and 0:1). The solution was loaded into a syringe with a blunt 20-gauge needle and extruded at 3 mL/h with a 24-kV applied voltage during the electrospinning process. The PCL-gelatin fibers were collected on copper shim at a distance of 16.5 cm. The resultant layer of fibers was placed in a desiccator to allow the evaporation of any remaining HFP. 

Noggin Coating

Before coating, 1-cm² × 1-cm² samples were cut. The scaffolds were placed in cell culture-treated 24-well plates (142475; ThermoFisher, Waltham, MA, USA) and immersed in 70% ethanol for 30 minutes for sterilization. After sterilization, the scaffolds were washed with phosphate-buffered saline (PBS) three times for 5 minutes each time. After washing, the scaffolds were incubated overnight at 4°C in recombinant human Noggin (N17001; Sigma-Aldrich) solutions at different concentrations (0 ng/mL, 100 ng/mL, 300 ng/mL, 500 ng/mL, and 1000 ng/mL) diluted with PBS. 

Characterization of Scaffold

Morphologic Analysis

For each blend ratio, samples were mounted on a 12.7-mm scanning electron microscope (SEM) stub with carbon adhesive followed by sputter-coating (SC7620 Mini Sputter Coater; Quorum Technologies, East Sussex, UK) to achieve a layer of gold and palladium with a thickness of 10 nm. The morphology of the electrospun scaffold was examined using the Hitachi TM4000 SEM (Tokyo, Japan) at 10 kV. The diameter of fibers and pore size were measured using DiameterJ,³² a plugin of ImageJ/Fiji.^33,34 Based on the morphologic analysis results, as discussed in the Results section, only PCL-gelatin scaffolds at a 1:1 ratio were used in the subsequent experiments. 

Mechanical Characterization

Mechanical analysis was performed using a modified ASTM standard D5035, “Textile Strip Method,” as described previously.³⁵ Briefly, 4-cm² × 1-cm² samples (n = 5) were prepared and mounted on a tensile tester (Q-Test 5; MTS, Eden Prairie, MN, USA). The tensile test was run using a 5 lbf (pound-force) load cell (22.24 N), 2.54-cm grip separation, 50% break sensitivity, and 0.01-mm/s strain rate. 

Degradation Rate

The degradation rate was calculated as a weight loss percentage of the dry scaffold. Both enzymatic and PBS degradation was observed. The enzymatic degradation was prepared by using a 1:1 solution of 0.25 mg/mL collagenase (type I, >125 U/mg, 17018029; ThermoFisher) in 0.1 M Tris-HCl and 0.005 M CaCl₂ and 0.0025 mg/mL Pseudomonas lipase (type XIII, ≥15 U/mg solid, L9518; Sigma-Aldrich) in PBS as described previously.^35,36 Each scaffold (n = 5) was weighed at the beginning of the test and then submerged in degradation solution at 37°C for different durations (7, 14, and 21 days). Samples were dried after the submersion period and weighed. SEM images of the samples were also taken after using similar time points mentioned above. The degradation rate was calculated using Equation (1):  

\begin{eqnarray} Weight{\rm{\ }}loss{\rm{\% }} = \left( {\frac{{{W_0} - {W_t}}}{{{W_0}}}} \right){\rm{*}}100{\rm{\% }}\quad \end{eqnarray}

(1)

 

Here, W₀ is the initial weight of the sample and W_t is the weight of the sample after a certain time point of degradation. 

Fourier Transform Infrared Spectroscopy

The functional groups of polymers and prepared scaffolds were assessed using Fourier transform infrared spectroscopy (FTIR) spectra (Nicolet iS10 spectrometer; Thermo-Fisher) in the range of 4000 to 700 cm⁻¹. 

Contact Angle

The water contact angle of scaffolds without and with Noggin coating (1000 ng/mL) was observed. For each condition, 2.54-cm² × 2.54-cm² samples (n = 5) were prepared to facilitate multiple measurements. The contact angle was measured using a goniometer (Dataphysics OCA System; FDS Corp, OCA 15 System, Dataphysics Instruments, Charlotte, NC, US) with 3 µL water droplets. 

Immunostaining for Noggin on Scaffold Surface

Immunostaining was done to confirm the presence of Noggin on the scaffold surface. The samples were washed with PBS once and blocked with 2% bovine serum albumin + 2% goat serum for 1 hour. Then the samples were incubated with Polyclonal Rabbit IgG Noggin Antibody (1:50, PA5-109231; ThermoFisher) overnight at 4°C. Later, the samples were washed with PBS and incubated with goat anti-rabbit IgG (H + L) Alexa Fluor 488 (1:250, A11008; ThermoFisher) at room temperature for 1 hour. Imaging was done using a fluorescent microscope (EVOS FL Auto 2; ThermoFisher). 

In Vitro Experiments

Cell Culture and Seeding

Human corneal epithelial cells (CEpiC) (36045-16; Celprogen, Torrance, CA, USA) were maintained in CEpiC Culture Complete Media with Serum (M36045-16S; Celprogen) at 37°C and 5% CO₂. The culture media were changed every 2 or 3 days. The cells were passaged regularly using Trypsin-EDTA (0.25%) (25200072; Thermo-Fisher) upon reaching approximately 80% confluence. 

To observe interactions of CEpiCs with scaffolds, 1-cm² × 1-cm² samples were placed into 24-well plates (one sample per well), which were prepared as described earlier. CEpiCs were seeded atop the scaffold at a density of 1 × 10⁵ cells/cm² per scaffold. CEpiCs seeded at the same density into wells of a 24-well plate coated with FNC coating mix (0407; Athena Enzyme Service, Baltimore, MD, USA) were used as control. 

Cell Viability

Cell viability assays were conducted on days 1, 3, 5, and 7 postseeding of CEpiC cells. LIVE/DEAD Cell Imaging Kits (488/570) (R37601; ThermoFisher) were used to observe cell viability following the manufacturer's recommendations using a 1:1 dilution ratio with PBS. Following a 30-minute incubation period at 37°C, the samples were imaged using a fluorescent microscope (EVOS FL Auto 2; ThermoFisher). Cell proliferation assays were performed on days 1, 3, 5, and 7 postseeding. The cell proliferation rate was quantified using alamarBlue Cell Viability Reagent (DAL1100; ThermoFisher). Triplicate samples were washed with PBS and then incubated with alamarBlue reagent diluted 1:10 v/v in cell culture medium for 2 hours at 37°C in the dark. Supernatants were collected from each well and pipetted in triplicate on a 96-well plate (100 µL each well, 10861-666; VWR, Radnor, PA, USA), and the resulting fluorescence value was read using a microplate reader (Synergy HT; BioTek, Winooski, Vermont, US) at 540-nm excitation and 590-nm emission wavelengths. 

Immunofluorescence

Cytokeratin 12 (CK12) staining was performed 7 days postseeding of CEpiCs. CK12 is a unique marker for CEpiCs and is also responsible for maintaining the structural integrity of CEpiCs.³⁷ Immunocytochemistry of CK12 will indicate if seeding on Noggin-coated scaffolds impacted the normal functionality or phenotype of CEpiCs. Samples were first washed with PBS and then fixed with 4% paraformaldehyde solution at room temperature for 20 minutes. After fixing, samples were washed with PBS three times for 5 minutes each and then permeabilized using 0.25% Triton X-100 for 30 minutes. Serum blocking was incubated for 1 hour using 2% bovine serum albumin + 2% goat serum to block nonspecific binding. Anti–cytokeratin 12 antibody (1:100; sc-515882, Santa Cruz Biotechnology, Dallas, TX, USA) was diluted in a blocking buffer and incubated overnight at 4°C. After three washes of PBS 5 minutes each, the secondary antibody goat anti-mouse Alexa Fluor 488 (1:250; A-21141; ThermoFisher) diluted in 10% blocking buffer was added and incubated overnight at 4°C. Nuclei were counterstained with Hoechst 33342 and imaged with a fluorescent microscope (EVOS FL Auto 2; ThermoFisher). 

In Vitro Wound Healing

For this experiment, 1 × 10⁵ cells/cm² of GFP–Human Corneal Epithelial Cells (cAP-0211GFP; Angio-Proteomie, Boston, MA, USA) maintained in Epithelial Growth Medium (cAP-45; Angio-Proteomie) were seeded in cell culture–treated 24-well plates and grown until reaching confluence. After that, a linear, cell-free defect was made in the cell monolayer by scratching with a sterile 1000-µL pipette. Cells were washed with PBS to remove cellular debris prior to adding plain culture media, culture media with 500 ng/mL Noggin, or a Noggin-coated scaffold (1000 ng/mL) in culture media. All plates were incubated at the conditions described earlier. Images of the scratches were captured at 0, 6, and 12 hours postscratching using a fluorescent microscope (EVOS FL Auto 2; ThermoFisher). To ensure the same area was imaged at each time point, we identified the coordinates at which images of scratches were taken at hour 0 and developed an automated scan protocol to take images at hour 6 and hour 12 using built functionality within the fluorescent microscope. An ImageJ plugin³⁸ was used to measure cell-free area at each time point, and wound closure percentage was determined using Equation (2):  

\begin{eqnarray} Wound{\rm{\ }}closure{\rm{\% }} = \left( {\frac{{{A_0} - {A_t}}}{{{A_0}}}} \right){\rm{*}}100{\rm{\% }}\quad \end{eqnarray}

(2)

 

Here, A₀ is the initial wound area, and A_t is the wound area after a certain time point of the initial scratch. 

Statistical Analysis

At least three replicates were carried out for each measurement. All data are presented as mean ± standard error of the mean unless indicated otherwise. Statistical analysis was carried out using GraphPad Prism version 9.4.1 for Windows (GraphPad Software, San Diego, CA, USA) by two-way analysis of variance followed by Tukey's multiple comparison post hoc test, and P < 0.05 was considered statistically significant. 

Results

Morphology of Electrospun Scaffolds

Observations from SEM images show that the polymer/protein blend ratio played a critical role in the shape of nanofibers obtained by electrospinning (Fig. 1). The presence of gelatin caused the fibers to have a more ribbon-like morphology with a flattened shape, whereas the increase in PCL content caused the fibers to take on a rounded appearance. Additionally, reduced fiber diameter and more uniform distribution of fiber diameter were observed with an increase in PCL content in the blend ratio. The mean fiber diameter, fiber diameter distribution, and pore size are presented (Fig. 1). For subsequent experiments, the blend ratio of 1:1 (PCL-gelatin) was used as it created nanofibrous membranes with optimal handleability and a porous structure with uniform fiber diameter. 

Figure 1.

 

Representative SEM images and fiber diameter distribution of PCL-gelatin electrospun scaffolds with blend ratios of (A) 2:1; (B) 1:1; (C) 1:2, and (D) 0:1. Scale bar: 20 µm.

Degradation Rate and Mechanical Properties

The degradation behavior of the electrospun scaffolds was observed in both enzymatic and PBS solutions. Faster degradation, severe fiber breakage, and deformation were observed with the enzymatic solution compared to PBS only (Fig. 2A). After 21 days, samples lost almost 40% of their initial weight in enzymatic conditions and almost 17% in PBS only (Fig. 2B). However, there were no statistically significant differences in degradation between the two conditions. The presence of collagenase and lipase in enzymatic solution acted as catalysts, which promoted faster degradation compared to passive hydrolysis in PBS.³⁹ Also, the presence of these enzymes switches the mechanisms of degradation from bulk to surface erosion, leading to more controlled and uniform degradation.^39,40 

Figure 2.

 

Bulk property characterization of PCL-gelatin (1:1). (A) SEM images of degradation over 21 days. Arrows indicate fiber breakage. Scale bar: 20 µm. (B) Weight loss percentage due to enzyme and PBS degradation over 21 days. (C) Representative stress–strain curve and (D) mechanical properties.

The mechanical properties of the scaffolds were analyzed and summarized (Figs. 2C, 2D). The stress at break was 1.198 ± 0.240 MPa with Young's modulus of 52.305 ± 6.130 MPa. These results indicate that fabricated scaffolds possess better or similar mechanical properties than amniotic membranes or commercially available contact lenses used in corneal wound healing.^41–43 

Surface Characterization

FTIR spectra were used to analyze the surface composition of the scaffolds (Fig. 3A). Bands characteristic to both PCL and gelatin were identified in the electrospun scaffold. Carbonyl stretching at 1726 cm⁻¹ and symmetric C–O–C stretching at 1171 cm⁻¹ were observed in the scaffold corresponding to the characteristic bands of PCL. Also, amide I at 1651 cm⁻¹, amide II at 1541 cm⁻¹, and amide A at 3305 cm⁻¹ characteristic bands of gelatin/protein were observed. A slight shift in bands was observed in the electrospun scaffolds compared to PCL pellets and gelatin powders. This can be attributed to the interaction between the ester group of PCL and the amine group of gelatin.⁴⁴ 

Figure 3.

 

Surface characterization of prepared scaffold. (A) FTIR spectra. (B) Water contact angle. (C) Noggin antibody staining (green) at different concentrations on scaffold. Scale bar: 100 µm.

It is difficult to distinguish between Noggin and gelatin using FTIR spectra as they share similar functional groups, causing overlapping of bands, and the quantity of Noggin was quite low compared to gelatin. Therefore, Noggin antibody staining was performed to detect the presence of Noggin protein on the surface of the scaffold (Fig. 3C). Using fluorescent microscopy, the presence of Noggin was detected. As Noggin concentration was increased, there was also an increase in fluorescent signal intensity and coverage corresponding to the presence and distribution of Noggin on the scaffold surface. Also, a slight decrease in water contact angle was observed when noncoated (0 ng/mL) and coated (1000 ng/mL) scaffolds were compared (Fig. 3B). This could be attributed to the surface of the scaffold becoming more hydrophilic with the presence of Noggin on it. 

Biocompatibility

For the scaffold to perform optimally as a corneal bandage, corneal cells need to be able to survive and proliferate on the scaffold. Fluorescent microscopy utilizing a Live/Dead assay demonstrated high cytocompatibility between the PCL-gelatin scaffold and CEpiCs, with minimal cytotoxicity (Fig. 4). An increase in the number of CEpiCs was observed across all scaffolds other than 300 ng/mL from day 1 to day 5, with high viability on day 7. This indicates that the fabricated scaffolds support cell proliferation. Upon closer visual examination, migration and attachment of the cells to different layers of the scaffold were observed, regardless of Noggin coating concentration. Decreased visibility of cells observed on day 7 can be attributed to the migration of cells within the inner structure of the scaffold. Also, in some cases, cells on the scaffold reached confluency at day 5, which caused contact inhibition. This indicates that the scaffolds are suitable for biointegration and emulate a physiologically favorable microenvironment for CEpiCs. 

Figure 4.

 

Cell viability on scaffolds with varying concentrations of Noggin at different time points. Live cells are shown in green and dead cells are in red. Representative images shown here. Scale bar: 275 µm.

The metabolic activity of the CEpiCs maintained on the scaffolds with different concentrations of Noggin protein was quantified using an alamarBlue assay. CEpiCs seeded onto scaffolds demonstrated an increase in metabolic activity over the course of the 7 days, confirming there were no cytotoxic issues and good cytocompatibility (Fig. 5). It can be noted that there was a decrease in metabolic activity for the 100-, 500-, and 1000-ng/mL Noggin-coated scaffolds from days 5 to 7. This decrease can be attributed to contact inhibition commonly observed in the CEpiCs and other cell phenotypes, which are dependent on the cell type, seeded cell density, and confluency.^35,45–49 Our prior studies repeatedly observed this phenomenon with electrospun scaffolds seeded with murine fibroblasts³⁵ and corneal endothelial cells (data not shown). With a seeding density of 1 × 10⁵ cells/cm² per scaffold, it follows a similar reasoning that the CEpiC had reached a level of confluency on the scaffolds on day 5 that caused them to reduce metabolic activity. Additionally, in a three-dimensional environment, the location of cells can vary tremendously, which can cause variation in the timeframe in which contact inhibition occurs between different samples. Overall, during the analysis before contact inhibition, scaffolds coated with Noggin showed higher or similar metabolic activity compared to the cell-only control and the noninfused scaffold. Specifically, a notable and significant difference (****P < 0.0001) in the metabolic activity on days 3 and 5 was observed for the 1000-ng/mL scaffold compared to all the other samples. Similar trends were observed in the qualitative cell viability assay (Fig. 4), which suggests that the effect of Noggin becomes prominent at the concentration of 1000 ng/mL. As such, 1000-ng/mL scaffolds were selected for the wound-healing assay. Also, strong expression of CK12 was observed in all conditions regardless of the concentration of Noggin on the scaffold (Fig. 6). This indicates that the presence of Noggin does not affect the phenotype and functionality of CEpiC. 

Figure 5.

 

Metabolic activity on scaffolds with varying concentrations of Noggin at different time points. The error bars represent standard deviation and **** denotes statistical significance at P < 0.0001.

Figure 6.

 

Protein expression after 7 days postseeding with cell nuclei stained blue and CK12 proteins stained green. Scale bar: 100 µm. Cells were seeded on PCL-gelatin scaffolds coated with (A) no Noggin or (B) 100 ng/mL, (C) 300 ng/mL, (D) 500 ng/mL, and (E) 1000 ng/mL Noggin.

Wound-Healing Assay

An in vitro wound-healing assay was performed to assess the closure of an artificially induced wound on the cellular monolayer of CEpiCs in the presence of a Noggin-coated scaffold (1000 ng/mL). The wound closure rate was 40.78% ± 2.45% and 61.36% ± 4.48% in the control (cells only) well after 6 and 12 hours of incubation, respectively (Fig. 7). Noggin-added culture media had a similar wound closure rate as the control well. The Noggin-coated scaffold had a slightly higher wound-closing rate than the control well. However, no significant difference was observed between the control and others. The difference in wound closure rates between Noggin directly added to the culture and the Noggin-coated scaffold may be due to the difference in release rate. Noggin-coated scaffolds release Noggin protein at a gradual rate instead of a single initial bolus. This result indicates that the presence of Noggin does not impair wound healing and may promote wound healing. 

Figure 7.

 

In vitro wound-healing (scratch assay) study of Noggin-coated scaffold on CEpiCs. Scale bar: 100 µm.

Discussion

Electrospun biomimetic constructs using blends of synthetic and natural polymers have long since been appreciated for their supportive capacity in corneal tissue regeneration. With this study, we have sought to contribute to the expansion of the versatility and applicability of these constructs as an adjunctive ophthalmic treatment by demonstrating a novel therapeutic protein-infused device. To achieve this goal, we have investigated the feasibility of a nanofibrous polymeric membrane coated with a therapeutic protein in the application of supporting corneal wound healing in conditions such as RCES. 

Through optimization of electrospinning parameters, we were able to obtain continuous and defect-free nanofibers with different blend ratios of PCL-gelatin. We observed that an increase in the PCL blend ratio facilitates a uniform fiber diameter distribution and circular morphology, consistent with prior literature.^44,50 We also demonstrated that the quantity of a therapeutic protein fixed on a scaffold surface could be controlled by varying concentrations of protein in solution based on the principles of protein adsorption on polymeric surfaces.^51,52 This was done alongside producing a scaffold with desirable mechanical strength properties to improve handleability and suture retention. In addition to surgically relevant mechanical properties, the scaffolds must have appropriate mechanical integrity to withstand postimplantation physiologic forces such as intraocular pressure, eyelid motion, and tear film motion. This scaffold (52.305 ± 6.130 MPa) has a higher stiffness than commercially available soft contact lenses (0.3–0.6 MPa) and comparable stiffness to crosslinked amniotic membranes (46.5 ± 9.8 MPa).^41–43 

Our hypothesis of using a Noggin-coated scaffold was based on recent findings in canine models in tandem with skin wound models.¹⁵ The canine models found that there was a significant –2.8-fold change in NOG expression for SCCED-affected Boxer dogs (a predisposed breed) versus non-Boxer canines affected with the same condition. This finding pairs with our in vitro observation that Noggin protein does not have any adverse effect on CEpiC even at higher concentrations and might even promote proliferation. These findings are additionally consistent with the observations seen in other areas, such as a neurorestorative study in mice that found Noggin improved ischemic brain tissue repair and reduced glial scar thickness.⁵³ It is likely that Noggin plays a very similar role in corneal wound healing as it does in wound-healing processes across the body. However, more investigation is needed as no direct correlation has yet to be shown in the cornea. 

Regarding therapeutic delivery, it is interesting to note that a higher wound-healing activity was observed when a Noggin-coated scaffold was used to introduce Noggin instead of directly adding Noggin to the culture. This might indicate that in the case of CEpiC, a gradual release or a continuous dosage of exogenous Noggin is more beneficial than a single-burst dosage. Our future work will focus on investigating a controlled release rate with the scaffold while maintaining the simplicity of the initial design. Also, there is a possibility that wound healing promoted by Noggin in CEpiC might use a different pathway than inhibiting the BMP-4 pathway exhibited in keratinocytes.^19,54–58 Thus, our future work will also focus on elucidating mechanisms/pathways for indolent corneal ulcers using in vitro models and translating these findings into clinical applications. 

Our proposed solution operates on an assumption of a therapeutic that targets the wound-healing issue of indolent corneal ulcers. This assumption is limited in its scope of clinical application outside of the specific condition. For example, many of the infection-based ulcers will involve a pathogen releasing a toxin damaging corneal cells, cytopathic effects, or inducing an inflammatory response that will involve enzymes such as matrix metalloproteinases that break down structural proteins and cause adhesion issues.²⁰ There is no evidence that Noggin or members of its relevant pathway can necessarily stop this aspect. Our study only suggests that the relevant pathways may have some sort of role in the cell-signaling aspect of wound healing. However, a molecular-based therapeutic such as Noggin can be used alongside treatments that take care of the “active threat” posed by pathogens to possibly speed up healing and ensure more positive outcomes. This is the appeal of the proposed bandage that can incorporate those pathogen treatments all in one treatment option. 

As new options for ophthalmic disease treatments are being investigated, nanofiber membranes offer a unique middle ground between tissue engineering and drug delivery solutions for ophthalmic challenges. This is mainly due to the ability of nanofibrous structures to emulate the endogenous extracellular matrix while having customizable technical properties such as protein adsorption. Upon refining the scaffold characteristics to reach clinical applicability, we are excited to investigate a diverse pool of topical therapeutics for infusion into the scaffold. Typically, a postoperation bandage will be used alongside topical treatments meant to improve healing times. Our intention is to incorporate those additional treatments into the scaffold, effectively reducing provider and client burden. 

Conclusions

In summary, we developed a Noggin-coated electrospun scaffold that can be used as a corneal bandage for applications in corneal indolent ulcers. We found that the fabricated scaffold showed nanofibrous porous morphology with desired mechanical properties, biocompatibility, and supported wound healing. These findings provide an opportunity of developing an alternative treatment for corneal indolent ulcers using tissue engineering principles. 

Acknowledgments

The authors thank Judy Elson, Teresa White, and Birgit Andersen (Textile Engineering, Chemistry, and Science, North Carolina State University) for their help with the experimental work and sample testing. 

Supported by the Wilson College of Textiles (JMG), the Textile Engineering, Chemistry and Science Department (JMG), North Carolina Textile Foundation–Wilson College Fellowship (NM), and CMI Young Scholar Award (NM, ES, DR). 

Disclosure: N. Mahmood, None; E. Sefat, None; D. Roberts, None; B.C. Gilger, None; and J.M. Gluck, None