Evaluation of the Algerbrush II rotating burr as a tool for inducing ocular surface failure in a mouse model

Ethics statement

All procedures were performed in accordance with the Stanford Laboratory Animal Care “APLAC” number 33,420, “Use of Animals for Scientific Purposes,” and the “ARVO Statement for the Use of Animals in Ophthalmic and Vision Research.” Male and female C57BL/6 mice aged 8–12 weeks were generously provided by the Irving L. Weissman laboratory. The animals were acclimatized to a 12-h light-dark cycle and provided water and feed ad libitum.

Surgical method

Based on the duration of exposure to the Algerbrush II, the animals were divided into four groups of 5 mice in group 1–3 with injury duration of 30–45 s, 60–75 s, 90–120 respectively. Group 4 contained 10 mice with two subgroups with injury duration of 3–4 min. The injury induced in one eye and the other eye was kept as control.

The animals were weighed before inducing intraperitoneal anesthesia which was a mixture of 100 mg/kg of ketamine hydrochloride (ANADA #200–055) and 5 mg/kg of xylazine (NADA #139–236) [16]. The inferior abdominal quadrants were selected as an injection site to avoid potential liver injury [17,18]. Lack of response to toe pinch was indicative of deep anesthesia.

All eyes were examined with a slit-lamp (Nidek SL-450, Nidek Co., Ltd, Aichi, Japan), before surgery to confirm the absence of any previous ocular injuries. The mice were placed in a lateral decubitus position on a surgical table prepared according to standard rodent surgery principles after anesthesia was induced [16]. A pillow measuring 8–10 mm (approximately 0.39 in) in height was placed under the head. Tetracaine hydrochloride (0.5%) eye drops (NDC 0065–0741–12) were administered for further anesthesia. An ophthalmic surgeon (AS) performed the surgery using a Zeiss S5 stereo microscope according to the following procedure. The ocular surface was dried with a surgical eye spear, and the eyelashes were trimmed. The eyelids were gently pressed against the orbital bones to induce proptosis, and the corneal and limbal epithelium was debrided using a sterile Algerbrush II with a 0.5-mm burr (catalog number: SVT14–5-2RB). To equalizing the pressure across the experimental groups the surgical instrument was held gently, and the angle between the burr and corneal surface was maintained at approximately 45° to avoid generating additional pressure. Thus, only the duration of exposure varied among experimental groups.

A timer was used to monitor the duration of Algerbrush II use. The burr was moved in a circular motion beginning at the central part of the cornea, then advanced subsequently to the superior and inferior quadrants without applying any additional pressure.

The duration of injury in group 1 was 30–45 s, and the entire corneal epithelium was debrided while maintaining an intact basement membrane (BM). In group 2, 60–75 s of injury duration was applied, and the corneal and limbal surfaces were shaved twice to remove the corneal and limbus epithelium. The injury duration to the corneal and limbus in group 3 was 90–120 s. To assess the impact of the size of the burr on the extent of injury, we divided group 4 into two groups of five mice. The duration of Algerbrush II use in both groups was 3–4 min; however, a 0.5-mm burr was used in the first subgroup, and a 1-mm burr was used in the second subgroup.

The tip was sanitized with 70% ethylene alcohol before and between surgeries, and the excess debris was cleared from the tip. Any detached epithelial sheets were cleared using a clean eye spear. Care was taken throughout the procedure to avoid damage to the limbal conjunctiva. Tail movement during the surgery was interpreted as a sign of pain and addressed by applying a drop of Tetracaine or injecting an additional bolus of 50% of the initial anesthesia volume.

Slit-lamp examination and fluorescent staining (FS) were performed postoperatively. FS was performed by administering 0.1% fluoresceine eye drops (0.1%; diluted from AK-fluor 10%, NDC 17,478-253–10 with buffered saline solution), drying the excess fluorescent liquid, and examining the eye using a cobalt blue filter (Figure 1 and Figure 2).

An ophthalmic ointment containing tetracycline and neomycin was applied postoperatively to prevent wound infection and corneal dryness, and 5 mg/kg of ibuprofen was injected subcutaneously to alleviate post-operative pain. The animals were examined on days 0, 2, 4, and 7.

Clinical evaluation

Slit lamp and Heidelberg Spectralis optical coherence tomography (OCT) examinations were performed to evaluate the cornea and anterior chamber on days 0, 2, 4, and 7 after the surgery. The animals were anesthetized before the examination, as described previously. Photographs were captured using the iPhone 13 camera in Cinematic mode. ImageJ was used on the fluorescent-stained images to calculate the ratio of epithelial defects (ED) to the entire corneal surface.

CO was scored on a scale of 1 to 4 based on the classification by Yoeruek et al. [19]: 0=normal, clear; 1=mild opacity; 2=severe opacity, but the iris and pupil are easily distinguishable; 3=iris and pupil are barely distinguishable; and 4=completely opaque with an invisible pupil.

The corneal NV was graded in accordance with the method by Bahar et al. with modifications [20]. Limbal NV was restricted to the limbus, paracentral NV extended from the limbus to the periphery of the cornea, and central NV extended to the pupillary area within 1 mm of the central cornea.

Scoring system of the OCT images

Two different anterior OCT images were selected for each eye. The corneal thickness was measured in pixels at three separate locations in each image, the values were converted to micrometers, and the average and standard deviation for each group were calculated and compared with those of other groups.

Histologic evaluation

Enucleation was performed after euthanizing the animals. Histological evaluation was performed on postoperative days 4 and 7. The enucleated eyes were fixed in 10% formalin overnight at room temperature, dehydrated using a series of graded alcohols and xylene, and subsequently encased in paraffin. The paraffin blocks were cut into 6-µm-thick sections and mounted on glass microscope slides. Hematoxylin and eosin (H&E) and periodic acid-Schiff (PAS) staining were performed. All layers of the cornea and anterior chamber were evaluated, and the severity of inflammation was assessed for each group. The remaining sections were subsequently used for immunostaining. The slides were deparaffinized using xylene and graded alcohol. Epitope retrieval (ER) was performed before immunostaining by immersing the deparaffinized slides in Citrate Buffer with pH 6.0 for 20 min at 85 °C. The slides were immersed for another 20 min at room temperature to cool down, then reheated as described previously. After rinsing in staining buffer (10 mM Tris buffered saline with 0.025% Triton X-100), antigen blocking was performed for one hour using BSA (BSA) 5%. The slides were incubated in a buffer containing primary keratin 13 antibodies (K13; a 1:100 dilution of mouse monoclonal, Cat. No. ab92551, Abcam, Cambridge, UK) overnight at 4 °C. Donkey anti-Mouse-IgG (A10037) at a dilution of 1:500 was used to detect the binding of the primary antibodies. Negative controls were created by circumventing the primary antibody incubation step. Positive controls consisted of non-wounded tissue sections stained with keratin 13 antibodies. The slides were imaged using a Leica Thunder Imager microscope.

Statistical analysis

Statistical analysis comparing multiple groups with parametric data was performed by one way ANOVA (GraphPad Prism) analysis. All data were presented as mean ± standard deviation. The Tukey’s multiple comparison test was used for additional subgroup analysis. A value of p<0.05 was considered statistically significant.

Slit lamp examination findings

Table 1 summarizes the main findings of the study. Corneal edema and exudation release into the anterior chamber, immediately after injury induction, were minimal in group 1. However, exudate release was observed after around 60 s of injury and was significantly increased on advancing the exposure time to 2 min and more. Group four has the most severe exudate release compared to other groups (Figure 1).

During the first week post injury in group 1 and 2, re-epithelialization occurred, and edema decreased continuously from the peripheral to the central cornea. In group 3, 75% of eyes were completely re-epithelialized with negative FS, which was a clinical sign of basement membrane disease [21]. In group 4, none of eyes were completely re-epithelialized after 7 days of injury and the mean epithelial defect was 50.38%.

On day 7, the mean CO in group 1 was 0.5+ and 25% developed corneal NV. The mean CO in group 2 was 0.8+ and 33% of eyes developed corneal NV. The mean score of CO in group 3 and 4 were 2+ and 2.2+ respectively which was diffusely distributed to the cornea. 100% of the eyes showed corneal NV in both groups. A statistically significant difference in CO was observed between group 1, group 3, and 4 (Figure 2).

Regarding the tip size of Algerbrush, the mean CO induced by the 1-mm and 0.5-mm burrs were 2.33 and 1.9, respectively, which was statistically significant (p value=0.029).

OCT findings

The corneal thickness was measured in all groups after injury induction. Statistical analysis revealed significant differences between the normal and 60–75 s and 90–120 s injury duration (Figure 1B). However, there was no significant difference between normal corneal and the 3–4 min injured cornea due to development of corneal edema after prolonged injury (Figure 3).

Histology and immunohistology findings

The H&E-stained corneal section of the C57BL/6 mice were observed under the light microscope with 40X magnification is shown in Figure 4A,B. The epithelium, which comprised 4–5 cell layers and BM, covered the corneal stroma.

Immediately after the injury, the corneal epithelium was eliminated and the cornea exhibited an irregular stromal surface. The iris vessels were dilated and the anterior chamber was filled with exudate (Figure 4C,D). On day 4 after the injury, the conjunctival epithelial cell layers proliferated, with prominent nuclei migrating through the limbus to the central corneal area (Figure 5A).

On day 7 after the injury, the corneal surface was covered with a thin and atrophic layer of epithelial cells in all groups except the eyes in group 4, which still exhibited stroma with central denudation. The endothelium was reactive and covered with pigments and exudate.

Goblet cells were detected on the corneal surface in the PAS-stained slides in all groups which was a marker of conjunctivalization (Figure 5B,C). The immunohistochemical staining, revealed the expression of K13 on the corneal surface which were expressed from the epithelial cells with conjunctival origin that covered the injured cornea and confirmed conjunctivalization (Figure 6).

3.4 Complications of Algerbrush II injury

The vibrations from the Algerbrush II burr transferred to the eyeball and likely resulted in pigment release from the ciliary body and iris into the anterior chamber, especially in groups that were exposed for more than 2 min (Figures 3). Another complication included bleeding from the conjunctival or limbal vessels, which was avoided by applying gentle pressure on the limbus using the Algerbrush II. Furthermore, cataracts and iris neovascularization were detected in the eyes with prolonged exposure to the Algerbrush II (Figure 3C).