Histatin-5 Improves In Vivo Corneal Wound Healing and Immunohistochemical Neovascular Markers After Acute Alkali Injury

Abstract

Purpose

Histatin-5 (Hst5), an endogenous antimicrobial peptide, has recently been shown to promote corneal epithelial wound healing and reduce inflammation after injury. This study aimed to investigate whether topical Hst5 treatment can inhibit corneal neovascularization and promote tissue recovery in a murine model of alkali burn injury.

Methods

Mice were randomly assigned to two groups: balanced salt solution (BSS) and Hst5-treated (5 mL of 20 µM stock) after alkali injury. Alkali burns were induced on the cornea of each mouse and topically treated three times a day for up to two weeks. Wound closure was analyzed through epithelial wound healing, corneal opacity, and neovascularization. Histologic analysis was also conducted.

Results

Topical Hst5 treatment significantly accelerated corneal wound healing, reducing residual epithelial defect area by ∼60% at 18 hours and achieving complete closure by 24 hours, compared with BSS-treated controls. Corneal opacity scores were significantly lower in the Hst5 group from Day 7 with near-clear corneas by Day 14. Neovascularization scores were markedly reduced in Hst5-treated eyes at Day 14, and accompanied by decreased infiltration of CD45⁺ immune cells.

Conclusions

Hst5 reduces alkali burn-induced corneal epithelial wound healing time, opacification and neovascularization in murine models. Future research is required before clinical implementation; however, the results suggest a potential therapeutic role for Hst5 for alleviating the adverse effects associated with acute corneal alkali injuries.

Translational Relevance

Topical Histatin-5 promotes epithelial regeneration and suppresses inflammation-driven corneal neovascularization following alkali injury, suggesting Hst5 as a novel, biologically-derived treatment to accelerate corneal healing and prevent vision loss after chemical burns.

Introduction

Alkali eye injuries are potentially devastating events for patients around the world, with severe consequences including corneal melting, chronic wounds, corneal neo-vascularization, limbal stem cell deficiency and vision loss.¹ The widespread, rapid injury from alkali burns stems from the lipophilic nature of high pH solutions, allowing for the saponification of cell membrane fatty acids which allows for penetration into the cornea stroma.¹ Approaches to management of ocular alkali injuries include primary prevention, protective eye wear, supportive care with eye washes and urgent evaluation. In the United States, it is estimated that 15,865 cases of ocular chemical burn occur per year.^2,3 One study recorded 144,149 chemical diagnoses of ocular burns across the US over a 4 year period, with young children representing the highest risk age group for chemical injuries.^3,4 Sixty percent of ocular burns are caused by alkalis.^5,6 Acute chemical burns can cause irreversible damage to the ocular surface and limbus, resulting in chronic complications such as corneal neovascularization, scarring, and limbal stem cell deficiency. In these cases, limbal stem cell transplantation and corneal transplantation are considered.^3,5 Thus treatment in the acute phase of chemical injury is critical for preventing chronic complications. The promotion of corneal re-epithelialization is essential for recovery in the acute phase but effective treatments remain a challenge.³ Amniotic membrane transplantation has been studied, and while a very effective agent for ocular surface reconstruction and planned surgeries, randomized control trials have reported mixed results specifically for acute corneal burns.^7,8 Umbilical cord serum, autologous peripheral blood serum, and platelet-rich plasma have been studied for the their wound healing effects as well.³ However, their clinical use requires larger studies and the obtainment of these blood-derived serums remains a limiting factors for these treatments.³ Given the extent of injury that can cause sequelae of vision loss from alkali injuries, new treatments, particularly in the acute phase of injury, are urgently needed.

Histatin peptides are a 13-member family of endogenous antimicrobial peptides that have a wide variety of biological effects in mucosal biological systems.⁹ They were originally discovered in the saliva but have since been found in the tears and ocular surface. Different members of the histatin peptide family have different functional specifications, with histatins 1, 2, and 5 reported to have wound healing properties in multiple tissue types. These peptides are also reported to have immunomodulatory functions that have been shown to improve recovery following injury. Recent research has found that the effects of histatin peptides allows for improved corneal wound healing.^10–12 An in vivo study induced 2 mm epithelial wounds on rabbits and treated the wounds with varying concentrations of histatin-1. They found accelerated wound recovery, faster hourly healing rates, greater percentage recovered area, and no adverse effects compared to the control group.¹³ These findings suggest that these endogenous peptides have the potential to improve care by hastening recovery, preventing infection and reducing the chances of long-term negative consequences of alkali injury. In this study, we tested and quantified the ability of topically applied histatin-5 (Hst5) to improve corneal re-epithelialization, reduce inflammatory cell infiltration and reduce corneal neovascularization in a mouse model of corneal alkali injury.

Material and Methods

Corneal Alkali Burn Injury in Animals

All animal procedures were conducted approved by the Institutional Animal Care and Use Committee at the University of Michigan and adhered to the ARVO statement for the use of animals in Ophthalmic and Vision Science research. Six-to-eight-week-old female C57BL/6 mice (The Jackson Laboratory, Bar Harbor, ME, USA) were acclimatized in standard environmental conditions for seven days and inspected for signs of corneal abnormality prior to the beginning of the study. Mice were randomly assigned to two experimental groups (n = 5 per group): balanced salt solution (BSS)–treated and Hst5-treated. Mice were anesthetized by intraperitoneal injection of ketamine (100 mg/kg) and xylazine (5 mg/kg). A 1.5 mm diameter circle was cut from Whatman no. 3 filter paper (GE Healthcare Life Sciences, Piscataway, NJ, USA) using a trephine and soaked in 1 N sodium hydroxide solution. After applying two drops of topical 0.5% proparacaine, the filter paper was placed at the center of the cornea of the left eye for 30 seconds, then carefully removed. The ocular surface was immediately irrigated with 10 mL of saline solution and treated with erythromycin ointment (Bausch & Lomb, Rochester, NY, USA). Hst5 (5 mL of 20 µM stock) or BSS was subsequently applied topically to the cornea three times per day up to two-week periods. The concentration of Hst5 used was determined from prior studies on Hst5 wound-healing effects where a relatively low concentration of 20 µM was observed to be highly effective in promoting epithelial wound healing.¹⁴ The two-week period was used to evaluate the sustained effects of Histatin-5 on stromal repair and overall corneal recovery. Determination of treatment length was based on stromal layer healing. Histatin-5 was administered topically three times daily at five-hour intervals. This schedule was maintained throughout the study to ensure uniform exposure of the ocular surface to the peptide.

Analysis of Wound Closure

Corneal wound closures were quantified with fluorescein-based imaging to quantify the assessment of wound area closure. The corneas were stained with fluorescein (one drop of FUL-GLO fluorescein sodium ophthalmic strips) and imaged under a cobalt filter with standard slit-lamp biomicroscopy. For the next two to three days (variation in timing based on whether wounds closed before end of three days), the wounds were measured every six hours and quantified using Image J software. The wound area remaining was calculated until wound closure and evaluated daily after closure to ensure no wound recurrence.¹⁵

Assessment of Corneal Opacity and Neovascularity

Corneal opacity was evaluated on days 1, 2, 4, 7, 9, 11, and 14 after injury to assess treatment response. The anterior segment of each mouse eye was imaged using a slit-lamp biomicroscope (Nikon D200 camera; Nikon, Melville, NY, USA), and corneal clarity was graded by a masked observer using a standardized Fantes scoring: 0 = no opacity, completely clear cornea; 1 = slightly hazy, iris and lens visible; 2 = moderately opaque, iris and lens still detectable; 3 = severely opaque, iris and lens hardly visible; and 4 = completely opaque, with no view of iris and lens.^16,17

Cornea neovascularization was observed and scored under slit-lamp microscope two, four, seven, nine, 11, and 14 days after alkali burn in five mice from each group. Corneal neovascularization was scored by a masked observer by vessel extent using a scale from 0 to 4, where 0 = no vessels on the corneal limbus; 1 = vessels advance over the corneal limbus, covering 0%–25% of the burned area; 2 = vessels that reach 25%–50% of the burned area; 3 = vessels that reach 50%–75% of the burned area; and 4 = vessels that extend to the entire burned area.¹⁸

Histologic Staining and Assessment

For histologic staining, whole mouse eyes were snap-frozen in optimal cutting temperature (OCT) compound (Fisher Healthcare, Waltham, MA, USA). The frozen tissues were cut into 10-µm cryosections (NX50 cryomicrotome; Thermo Fisher Scientific, Waltham, MA, USA), then sections were mounted on Superfrost Plus slides (Thermo Fisher Scientific). For the hematoxylin-eosin staining, slides were fixed using 4% paraformaldehyde (Thermo Fisher Scientific) for 30 minutes, followed by several phosphate-buffered saline solution (PBS) washes and then stained with Harris hematoxylin solution, modified (Sigma-Aldrich, St. Lois, MO. USA) and Eosin-Y with Phloxine (Epredia, Kalamazoo, MI, USA). Slides were examined using a revolve microscope (ECHO, San Diego, CA. USA) using a ×10 objective. For counting of CD45⁺ or F4/80⁺ immune cells, slides were incubated overnight at 4 °C with mouse anti-mouse CD45 or mouse anti-mouse F4/80 from Biolegend (San Diego, CA, USA) as primary antibodies, washed three times for five minutes each time in PBS before incubation with Alexa fluor 568 goat anti-mouse IgG (Invitrogen, Walthman, MA, USA) or FITC goat anti-mouse Ig (BD Phamingen, Franklin Lake, NJ, USA) as secondary antibodies for 30 minutes. A solution of 1 µg of 4′,6-diamidino-2-phenylindole (DAPI) in 1 mL of PBS were counterstained for three minutes and observed under a confocal microscope (Zeiss LSM 710 Confocal Microscope, Oberkochen, Germany) using a 20× objective.

Statistical Analysis

The results are expressed as the mean ± standard deviation (SD). Normality of the data was tested using the Shapiro-Wilk test. Kruskal-Wallis test followed by post hoc Dunn's test was used for evaluation of corneal surface staining data and assessment of statistical significance among multiple comparison groups. One-way analysis of variance (ANOVA) followed by post hoc Tukey's test were utilized for evaluation of immunohistochemistry and TUNEL staining data and assessment of statistical significance among multiple comparison groups. All analyses were performed with Prism software (GraphPad; California, US). *P < 0.05 and ***P < 0.001 were considered statistically significant.

Results

Effects of Histatin-5 on Corneal Wound Healing and Opacification

Mice treated with Hst5 demonstrated substantially faster rates of re-epithelialization compared with BSS treated control mice at both 18 and 24 hours after alkali injury (Fig. 1). Corneal opacity progressed after alkali injury from Day 1 to 14. Continuous observation from Day 1 to Day 14 revealed a progressive increase in corneal opacity after alkali burn injury. In contrast, treatment with Hst5 significantly reduced corneal opacity from Day 7 through the end of the study (Day 14) compared with injured mice eyes (Fig. 2). As shown in Figure 2A, the healing effect of Hst5 became more evident on Day 14, when treated mice displayed nearly clear corneas, whereas injured mice treated with BSS continued to exhibit marked corneal opacity.

Figure 1.

Hst5 accelerated corneal wound closure rates in a murine corneal alkali burn injury model. (A) Slit-lamp images of murine corneas using a cobalt filter and fluorescein dye staining of the wounded areas were shown among the experimental groups (n = 5 for each group) (Hst5 [20 µM]). (B) Wound areas at multiple time points were measured using ImageJ software. Measurement of percentage remaining corneal wound area at 18 and 24 hours compared to baseline showed statistically significant improvement in Hst5 treated group compared to BSS. Un (Uninjured, Baseline). Statistical significance was determined by a two-way ANOVA with multiple comparison. *P < 0.05. % Wound area = (wound area at time x/wound area at time 0) × 100.

Figure 2.

Hst5 decreases corneal opacification in corneal alkali injury. (A) White light images were taken using slit-lamp biomicroscopy and images showed the area of corneal epithelial injury of nontreated and the Hst5 (20 µM). Corneal opacity was significantly lower in 20(S)-OHC-treated mice than in the BSS group. (B) The histogram represents the scoring of corneal opacity at the indicated time after alkali injury with BSS or Hst5 treatment. The values are expressed as floating bars (min to max) with mean (—) and are representative of five animals per group. Un = Uninjured, Baseline. Corneal opacification was scored using a system of scaling from 0 to 4 described in materials and methods. All values are expressed as mean ± SEM (n = 5 per group). Statistical significance was determined by two-way ANOVA with multiple comparisons. **P < 0.01, ***P < 0.001, ****P < 0.0001.

Effects of Histatin-5 on Corneal Structural Recovery After Alkali Injury (Histological Evaluation of Corneal Structural Recovery After Histatin-5 Treatment)

In addition to direct clinical observation, we carried out the histological examination of alkali burned corneas from both BSS- and Hst5-treated groups (Fig. 3). As shown in Figure 3A, alkali burn injured cornea showed severe stromal edema. However, corneas treated with Hst5 recovered towards baseline with mild edema in the stroma area. Statistical analysis demonstrated a significant recovery in cornea stromal edema between BSS and Hst5 treated groups (**P < 0.01) after alkali burn (Fig. 3D). Furthermore, alkali burn to the cornea resulted in significant thinning of the epithelial layer (4.3 folds) compared to uninjured healthy mice, which was restored after treatment of Hst5 (2 folds) compared to injured mice (Fig. 3E).

Figure 3.

Histologic and immunofluorescent staining of corneas at day 14 after alkali burn injury. (A) Histology of alkali burned corneal stained with hematoxylin-eosin (H&E). H&E staining shows numerous mononuclear cells present and loss of the epithelium the control (BSS) group, which was distinct from the uninjured group (Un). In contrast, the Hst5-treated groups showed attenuated mononuclear cell infiltration and recovery of the corneal epithelium, similar to those of normal corneas. (B) Immunofluorescent staining of CD45 in cross-sectioned murine corneas demonstrated that CD45⁺cells infiltrated the corneal stroma on day 14 after the alkali burn induction. However, Hst5 suppressed the infiltration of corneal CD45⁺ immune cell. (C) Bar graph representing the mean ± SD of CD45⁺ immune cell number counted from three eyes per treatment group. (D) Stromal edema was graded into four groups: 0–3 (0 = absence; 1 = mild; 2 = moderate; 3 = severe). Hst5 treatment recovered the stromal edema induced by alkali injury. (E) Epithelial thickness was measured in each group using Image J. The thickness of epithelial layer was reduced by injury and was normalized in mice treated with Hst5. Statistical significance was determined by Student's t-test. **P < 0.01, ***P < 0.001, ****P < 0.0001. Scale bar: 100 µm.

To investigate the effect of Hst5 peptide on inflammation associated with the alkali burn injury model, the CD45⁺ immune cell infiltrates were measured and compared between BSS and Hst5 treated group. The CD45⁺ immune cell numbers counted were 30.5 ± 9.15 in the BSS group and 5.5 ± 3.72 in the Hst5 group (Figs. 3B, 3C). Statistical analysis showed a significant elevation in CD45⁺ immune cell number between BSS and Hst5 groups (***P < 0.001) after injury. Similarly, F4/80+ macrophages numbers were compared in the stroma of injured animals with substantial reductions in numbers of these inflammatory cells in the mice treated with Hst5 (Fig. 4). The F4/80⁺ macrophages numbers counted were 7.5 ± 6.2 in the BSS group and 1 ± 1.4 in the Hst5 group (Fig. 4B) after injury.

Figure 4.

Hst5 inhibits the infiltration of F4/80 positive macrophages induced by alkali burn injured murine corneas. (A) The uninjured cornea (Un). Immunofluorescent staining of F4/80 antibody in cross-sectioned murine cornea demonstrated that F4/80-positive macrophages infiltrated the corneal stroma on day 14 after the alkali burn induction. Fluorescence images showing infiltration of F4/80 positive cells (white arrows) in the alkali burn injured group at day 14 (B). Hst5 inhibits F4/80 positive inflammatory cells recruitment (C). (D) A bar graph representing the mean ± SD of F4/80⁺ cells number counted from three eyes per treatment group. Corneal sections were stained with immunofluorescent antibodies for positive cells, as shown (Scale bar: 100 µm).

Histologic sections of murine corneas after alkali injury were compared between treatment and control groups for several different features associated with injury, including CD45+ immune cell infiltration, stromal edema, and epithelial thickness (Fig. 3). Notably, there was statistically significantly fewer CD45+ cells, reduced stromal thickness, and increased epithelial thickness in corneas treated with Hst5 versus controls. Similarly, CD3+ T-cells and F4/80+ macrophages numbers were compared in the stroma of injured animals with substantial reductions in numbers of these inflammatory cells in the mice treated with Hst5.

Effects of Histatin-5 on Corneal Neovascularization After Alkali Injury

Post-alkali injury measurements of neovascular vessel growth into the cornea were measured over time and scored by a masked observer by vessel lengths and contents using a scale from 0 to 4 described in materials and methods. Corneal neovascularization increased in both groups post-injury until day 7, but regression occurred substantially more in Hst5 treated mice with near total regression by days 11 and 14 (Fig. 5). Statistical analysis showed a significant regression in neovascularization between BSS and Hst5 groups at day 11 (*P < 0.05) and 14 (***P < 0.001) after injury (Fig. 5B).

Figure 5.

Alkali burn injury-induced corneal neovascularization in corneas with or without Hst5 treatment. (A) Representative white-light images show corneal neovascularization after alkali injury in eyes treated three times daily with either BSS or Hst5 (20 µM). A notable reduction in neovascularization is observed in corneas treated with Hst5. (B) Neovascularization was graded on a scale from 0 to 4, as described in the Material and Methods section. Results are presented as floating bars (min to max) with the mean indicated (—), based on five animals per group. Un = Uninjured, Baseline. Statistical significance was determined by two-way ANOVA with multiple comparisons. *P < 0.05, ***P < 0.001.

Discussion

Management of devasting alkali chemical injuries of the ocular surface is complex, patient centric, and differs over the course of time from injury.¹⁹ From an acute timeline, copious irrigation, topical steroids, prophylactic antibiotics, ocular lubrication, and intraocular pressure therapeutics are often used depending on the extent of injury. Compared to a corneal abrasion, which is often a mechanical injury to superficial epithelium with the basement membrane often intact, alkali burns may cause deeper tissue destruction due to saponification of the cornea cell membrane fatty acids. The healing process for deeper injuries may be prolonged compared to superficial injuries. Such deep injuries that include the corneal stroma may lead to long-term vision loss such as scarring and chronic inflammation if not aggressively and immediately neutralized.²⁰ Severe alkali injuries may require surgical intervention or devices containing amniotic membrane tissue or other biological bandages. Chronic management depends on the sequela of the injury (e.g., corneal opacification, limbal stem cell deficiency) and can include lamellar corneal transplantation, ocular surface reconstruction, limbal stem cell transplantation amongst other interventions.²¹ Development of new, accessible therapeutics that can be employed at the time of injury to hasten recovery and prevent sequelae of these injuries is desperately needed. Hst5 is an endogenous, multifunctional peptide with domains that promote wound healing and abrogate infections and thus a potential therapy for acute alkali injuries.²²

In this study, the results observed that topical application of Hst5 after an acute corneal alkali injury in a murine model reduces time to re-epithelialization, reduces acute stromal edema, inflammatory cell infiltration, and neovascularization. The Hst5-treated eyes showed effective recovery of both the epithelial and stromal layers, with reduced stromal opacification and neovascularization compared to the vehicle-treated group, supporting a prolonged treatment regimen. These results are supported by a series of studies that indicate histatin-5 can improve rates of wound healing in multiple biological systems.^14,23–25 These findings are supportive of previous studies demonstrating acceleration of corneal epithelial wound healing by Hst5,^10,14 and significantly advance our understanding of translational applications of this salutary peptide by testing proof of concept of enhancing corneal injury response in a challenging model of alkali injury. Moreover, these results indicate that the critical drivers of vision loss after chemical injury, notably corneal inflammation, neovascularization, and opacification, are substantially improved with Hst5 treatment.

Although these results are exciting and support deeper investigation of the potential applications of ophthalmic applications of histatin peptides, several questions remain. Histatin-1 has been reported to exert biological effects through interaction with TMEM97, an ER protein that is associated with cell migration, lipid processing, and pathogen processing through its interaction with NPC1.²⁶ Further study is required to determine whether the biological effects of histatin-5 are mediated through this or other pathways. The potential mechanisms of action of Histatin-5, including its roles in epithelial wound healing and modulation of inflammatory responses, have been explored in our previous reports.^14,27,28 Recent studies indicate that MAPKinase signaling is important to the wound healing effects of histatin-5 and that comports well with studies that show the importance of these pathways in corneal epithelial wound healing in other models. Studies have also shown that histatin-5 can promote cellular migration which is necessary for wound healing. This proposed mechanism and molecular characteristics responsible for the wound healing properties of histatin-5 have been explored in cellular studies. In vitro studies with epithelial scratch assays demonstrated the SHRGY residue of the C-terminal in histatin-5 was found to be critical for migration.²¹ Truncated histatin-5 peptides without the SHRGY sequence did not achieve rates of increased wound closure compared to the controls.²¹ By increasing cellular migration, this possibly allows for expedited migration of corneal epithelial cells to cover the areas of insult, which may allow for increased wound closure. However, cellular interactions between histatin-5 and the surrounding corneal epithelial environment remains an area that requires further investigation. Additionally, future research can be done to understand the effect of histatin-5 on burns that are not alkali derived (e.g., acidic or thermal burns). Based on the results of this study and the aforementioned in vitro scratch assay experiment, it is likely that histatin-5 will provide benefit in the wound closure of non-alkali chemical burns.

Although there are merits to this research, there are also limitations. One limitation is that the mechanisms behind histatin-5 are not fully elucidated. The results of these experiments still require additional experiments to fully understand the mechanistic properties of histatin-5 wound healing. Another limitation is the sample size of the treatment groups. Because this study was primarily designed as a pilot to assess feasibility and preliminary efficacy, future research to increase the sample size will be helpful when investigating additional parameters such as dose or treatment duration. Another limitation is that there are inherent differences between murine and human corneas, as well as the limbal stem cells that are critical for corneal wound healing.²⁹ Henriksson et al.³⁰ noted the morphologic differences between human and murine corneas, however, it is not well understood if these differences play a significant role in severe alkali injuries. Although murine models have been a well-established and advantageous model organism for corneal injury research, future research with large animal models and ultimately human clinical trials will be necessary to advance Hst5 as a viable treatment option. Ultimately, the results of this study are encouraging, and future research is required to understand the full effects of Hst5 on alkali injuries.