Notch signaling promotes the corneal epithelium wound healing

Animal model

To create cornea epithelium-specific NICD transgenic mice, we crossed K14-cre^+/+ transgenic homozygous mice (stock number 004782; derived from B6xCBA F1; Jackson Laboratory, Bar Harbor, ME) with R26^fN1-ICD transgenic mice (Stock number 008159, Jackson Lab) to generate two types of mice, K14-cre^+/−/NICD⁺, and K14-cre^+/− mice (ratio of 1:1) [8,9]. All studies are conformed to ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and the institutional IACUC protocol.

Cornea epithelium debridement, fluorescein staining, and Richardson’s staining

Adult K14-cre^+/−/NICD⁺, K14-cre^+/− and wild type (WT) mice at the ages of 6–10 weeks were selected and anesthetized with an intraperitoneal injection of ketamine (50 mg/kg) and xylazine (5 mg/kg). After a few drops of 0.5% topical proparacaine were applied, the central epithelium, excluding the limbal area, was demarcated with a 1.5-mm-diameter biopsy punch and removed by gently scraping with a blade [10]. This type of wound only removes the epithelium and leaves the basement membrane intact. The epithelial wound was monitored right after injury with the fluorescein staining. Corneal surface integrity was examined and photographed with a slit lamp biomicroscope equipped with a cobalt blue light, after a 2-min surface staining with 2 μl of 0.5% fluorescein sodium salt and intense washout of the extra fluorescein with 1 ml PBS. The wound-healing process after 6, 12, 18, and 24 h was further monitored and recorded using the fluorescein staining method described earlier, after which the animals were euthanized, and the eyes were collected for either immunohistochemistry or isolation of RNA and protein from the corneas. The wound areas were measured in Portable Document Format (PDF) files using the measure tool and expressed in square millimeters (mm²). To avoid bias, genotyping of WT and transgenic mice were masked during the wounding experiments, image taking, and wound-area measurements.

In addition, we monitored the wounding area by Richardson’s staining to reveal exposed basement membrane. At 0, 18, and 24 h after epithelium debridement, eyes with wounded cornea in K14-cre^+/−/NICD⁺ and K14-cre^+/− mice were fixed in 4% paraformaldehyde (PFA) for 2 h. Four eyes were used for each time point. Whole eyes were stained with Richardson's solution (equal mixture of 1% azure II and 1% methylene blue/1% borax dissolved in 100 ml autoclaved water) to reveal exposed cornea basement membrane. Then corneas were removed and photographed with a digital camera connected to a dissecting microscope. The results are presented as average±SD and a value of p<0.05 was considered to be statistically significant.

Tissue collection

Corneas for RNA or protein preparation were dissected from WT, K14-cre^+/−, and the K14-cre^+/−/NICD⁺ mice under a dissecting microscope. For cell proliferation assay and immunohistochemistry, whole eyes were dissected 2 h after 5-bromo-2'-deoxyuridine (BrdU; 150 mg/kg bodyweight) injection and fixed in 4% PFA at 4 °C for 24 h and followed by routine paraffin-embedding and sectioning procedure. The sections were cut serially at 5 μm.

Immunohistochemical analysis

Sagittal eye sections (5 μm) were treated with antigen retrieval procedure by boiling in 10 mM Tris-ethylendiaminetetraacetic acid (EDTA) buffer (for 1 liter: 1.2 g Tris and 0.37 g EDTA, the solution was about pH 9) for 30 min, and then slowly cooled down in the same heating buffer at room temperature. After blocking with the blocking buffer (1× PBS, 0.2% Triton X-100, 5% normal serum), the slides were incubated with primary antibody in the blocking buffer for 1 h at room temperature; followed by three washes with PBS, 10 min each; and then by a 30 min incubation with appropriate secondary antibodies. The primary antibodies are rabbit anti-K14 (1:500, Cat# PRB-145P; Covance Research Products, Denver, PA), mouse anti-p63 (1:200, Cat# sc-8431; Santa Cruz Biotechnology, Santa Cruz, CA), goat anti-K12 (1:200, Cat# sc-17101; Santa Cruz Biotechnology), rat anti-BrdU antibody (1:800, Cat# MAS 250c; Harlan-Sera Lab, Belton Loughborough, Leicestershire, England), and rabbit anti-zonula occludens-1 (ZO-1; 1:600, Cat # 61–7300; Zymed Laboratories Inc., San Francisco, CA). The secondary antibodies were conjugated with either Cy3 or FITC and obtained from Jackson ImmunoResearch Laboratories, Inc. (West Grove, PA). Nuclei were stained by diamidinophenylindole (DAPI).

BrdU- positive cell counts and statistical analysis

On each corneal section, the BrdU-positive cells in corneal and limbal epithelia were enumerated on a digital micrograph taken at 100× magnification. The numbers of BrdU positive cells were divided by the length of corneal or limbal epithelia, respectively, to produce a number of BrdU positive cells per 1 mm. The results were expressed as the average counts from three sections per eye with 4 to10 eyes in each experimental group. All data are shown as the average±standard deviation (SD). The differences were considered statistically significant if the p values were less than 0.05.

RNA isolation and qPCR

Total RNA was prepared from four eyes in each genotype using TRIzol reagent (Invitrogen, San Diego, CA). The A260/A280 ratio of all RNA samples was >2.0 as measured by NanoDrop 1000 Spectrophotometer Nanodrop (Thermo Fisher, Wilmington, DE). cDNAs were reverse-transcribed using random primers and SuperScript VILO cDNA synthesis kit (Invitrogen). qPCRs were performed in a SYBR green-based PCR reaction mixture on a MX3005p system (Agilent Technologies, Inc., Santa Clara, CA), with a 10-min initial hot-start activation of Taq polymerase at 95 °C and followed by 40 cycles of amplification (95 °C for 25 s, 56 °C for 30 s, and 72 °C for 30 s). The comparative Ct (threshold cycle) method normalized to β-actin (Actb) was used to analyze relative changes in gene expression. The qPCR primer sequences for NICD, Hes1, and Hey1 were listed in Table 1 and described previously [11].

NICD transgene expression in corneal epithelium does not affect the corneal homeostasis

The Notch signaling pathway is critically involved in cell fate decisions during the development of many tissues and organs. To analyze the effect of NICD overexpression on corneal epithelium homeostasis and wound healing, we generated the transgenic mice with cornea epithelium-specific expression of NICD cDNA by breeding the previously described K14-cre^+/+ and R26^fN1-ICD mice for the generation of K14-cre^+/−/NICD⁺ mice (Figure 1A) [8,9]. RT-qPCR consistently showed a higher expression of Notch1 in adult K14-cre^+/−/NICD⁺ corneas than in WT corneas (Figure 1B). Among the immediate Notch target genes, Hes1 and Hey1 were upregulated in NICD transgenic corneas (Figure 1B). The newborn K14-cre^+/−/NICD⁺ pups showed severe skin blistering in the limb area, however, the pups appeared healthy after the first week. The skin blistering in NICD1 transgenic mice was caused by the loss of integrin expression and consequently, the loss of hemidesmosomes due to NICD induction, as described previously [12]. In addition, the NICD1 transgenic mice showed various proliferation and differentiation defects in the epidermis [12]. Surprisingly, a normal corneal morphogenesis in the transgenic mice at 2 and 4 months old was observed by the hematoxylin and eosin (H&E) staining of paraffin-embedded sections (Figure 2A); further examination of the progenitor markers p63, cytokeratin 14 (K14), K12, and tight junction maker ZO-1 by immunohistochemistry showed no abnormality in corneal epithelium differentiation (Figure 2A). In both WT and K14-cre^+/−/NICD⁺ transgenic mice, the basal layer maker K14 and p63 were similarly distributed and mostly confined to the corneal basal epithelial layer. K12 was strongly shown in the corneal suprabasal layer cells that normally form tight junctions at the cornea surface as judged by ZO-1 immunostaining (Figure 2A).

To further examine cell proliferation in the cornea epithelium, we injected animals with BrdU, a thymidine analog that incorporates into DNA during the S-phase of the cell cycle. The similar numbers of BrdU labeling cells were observed between K14-cre^+/−/NICD⁺ transgenic and WT corneas, suggesting that ectopic expression of NICD did not change cell cycling in cornea epithelium (Figure 2B).

NICD expression promotes corneal epithelial wound healing

The process of cell proliferation, differentiation, and migration during corneal wound healing response is highly coordinated and precisely controlled, and Notch plays a role in those events. We therefore examined the effect of ectopic expression of NICD on the corneal epithelial debridement wound healing. Mice at 2 months of age were wounded by a corneal debridement procedure. A 1.5-mm diameter circle at the central epithelium, excluding the limbal region, was removed. The corneal surface wound was then observed and monitored by fluorescein dye. Based on the measurement of fluorescein stained area at 18 h and 24 h after injury, we found that overexpression of NICD in corneal epithelium promoted wound healing process (Figure 3A,B). Eighteen hours post injury, the wound area was significantly reduced in the K14-Cre^+/−/NICD⁺ transgenic mice compared to the age matched WT mice, as assessed by fluorescein staining (0.355 mm² [n=10] versus 1.121 mm² [n=8]; p<0.05; Figure 3A,B).

We further confirmed the results with a better staining technique, Richardson's staining, to quantify the healing rate at 18 h and 24 h after injury. After the removal of central cornea epithelium demarcated with a 1.5 mm biopsy punch, the initial wound areas in wild type control mice (1.71 mm²) and NICD transgenic mice (1.70 mm²) were similarly produced as determined by the Richardson’s stained cornea (Figure 3C,D). However, the remaining wound area at 18 h after wounding in NICD transgenic corneas was significantly less than in controls (0.036 mm² [n=4] versus 0.612 mm² [n=4]; p<0.005; Figure 3C,D). At 24 h after debridement, the wounds were completely healed in NICD transgenic mice, while the wound areas were still present at the wounding site in WT cornea (0 mm² [n=4] versus 0.263 mm² [n=4]; p<0.005; Figure 3C,D). The results indicate that NICD overexpression in vivo enhances epithelial wound closure.

NICD overexpression alters the kinetics of corneal epithelial proliferation in response to wounding

To further dissect the cellular mechanisms underlying the promotion of corneal epithelial debridement wound healing by active nuclear NICD, we first examined the dynamics of cell proliferation in response to epithelial debridement wounding, using BrdU labeling and immunostaining quantification. The incorporated BrdU was primarily localized in the basal corneal cell layer. After wounding, the BrdU positive cells were gradually increased in a time window of 12 to 18 h post surgery in the WT corneal epithelium (Figure 4A). However, the kinetics of cell proliferation in the transgenic corneal epithelium was dramatically shifted, with a rapid increase in the number of BrdU labeled cells in the period of 6 to 12 h after wounding, as compared to the WT control (Figure 4A). The average number of BrdU-positive cells in a fixed length of transgenic cornea epithelium on 5 μm sagittal corneal sections was doubled to that of the WT controls during the same period following injury (Figure 4). However, by 18 h after wounding, fewer BrdU-labeled cells were observed in the NICD transgenic mice than in the WT animals. Such proliferation kinetic patterns were also observed in the limbal areas. Those data suggest that active Notch promotes rapid cell proliferative response.

Notch1 expression and its activity in WT mice are upregulated after wounding, correlating with proliferation induction

The results shown above suggest that increased Notch activity can promote cell proliferation in the wound healing response. To further correlate corneal epithelial cell proliferation initiated by wounding in normal mice with expression and activity of Notch, we examined mRNA levels for Notch1, Hes1, and Hey1 during the process of wound healing. Corneas from WT mice at 0, 3, 6, 12, 18, or 24 h after debridement were used for RNA isolation and qPCR quantification. No significant change was observed in the mRNA levels of Notch1, Hes1, and Hey1 at the time points of 0, 3, 6 or 12 h after debridement. However, by 18 to 24 h after injury, Notch1, Hes1, and Hey1 mRNAs were significantly increased as shown by real time qPCR assays (Figure 5). The induced Notch activity (Figure 5) was correlated to the BrdU incorporation pattern with the peak cell proliferation response starting at 18 h after injury (Figure 4). These observations support the hypothesis that an increased Notch activity during wound healing process contributes to corneal epithelial proliferation response.