PAX6 dosage effects on corneal development, growth, and wound healing

Abstract

The requirement for correct dosage of the transcription factor Pax6 during corneal growth and development was investigated using the Pax6-overexpressing (PAX77) transgenic mouse. Transgenics had a microcornea phenotype due to failure of postnatal growth, associated with reduction in the number of cells layers in the corneal epithelium. Cell cycle progression was monitored using bromodeoxyuridine, p63, cyclin E, and phosphohistone-3 labeling: proliferation rates were higher in PAX77+ than wild-type, without a concomitant increase in apoptosis. Hence, failure of proliferation did not underlie microcornea. PAX77+ corneal epithelia had reduced levels of cytokeratin-12, and exhibited severe wound healing delay that, in contrast to Pax6+/- mice, could not be modulated by exogenous growth factors. PAX77+ lenses showed partial failure of lens fiber differentiation. The data demonstrate that anterior eye development is very sensitive to Pax6 dosage. Although there are similarities between the eye phenotype of Pax6 heterozygotes and overexpressing mice, there are also striking differences. Developmental

Fig. 1

Histological analysis of the anterior segment of the eye of the PAX77⁺ transgenic mice. A,B: Comparison of the corneas of postnatal day (P) 10 wild-type (A) shows the corneal epithelium is stratified and is made up of two to three cell layers; the same pattern is seen in the PAX77⁺ (B). C,D: In the adult comparison of the corneal sections of both wild-type (C) and PAX77⁺ (D) revealed that the PAX77⁺ corneal epithelium was relatively normal. Quantitative analysis (see text) showed that, on average, it was composed of fewer cell layers than wild-type. E,F: Transmission electron microscope image of corneal stroma of wild-type (E) and PAX77⁺ (F) corneal stroma, showing that the interfibrillary gaps in (C) and (D) are processing artifacts and that collagen fiber lamination is normal in PAX77⁺. G: A normal ciliary body seen in the adult wild-type (arrowhead). H: Lack of a ciliary body (arrowhead) and iris hyperplasia in the PAX77⁺. l, lens; c, cornea. Scale bar = 50 μm in A-D, 2 μm in E,F.

Fig. 2

Corneal morphometry. A: Comparison of cornea diameter (mm) in PAX77⁺ and wild-types age postnatal day (P) 2, P10, and adult. B: Comparison of thickness of central (C) and peripheral (P) regions of the cornea in adult PAX77⁺ and wild-type mice. *P < 0.01, ***P < 0.0001 by unpaired t-test.

Fig. 3

Pax6 expression. A: Pax6 immunohistochemical staining in PAX77⁺ and wild-type eyes at postnatal day (P) 2, P10, and in adults. Pax6 expression in the cornea is increased in the PAX77⁺ at postnatal day 2 and postnatal day 10. B,C: Increased Pax6 expression in the cornea of the PAX77⁺ adults was confirmed by Western blot, and shown to be statistically significant when quantified and normalized against β-actin expression (C). Scale bar = 30 μm.

Fig. 4

Corneal epithelial proliferation. A,B: Frequency of bromodeoxyuridine (BrdU) -positive cells in the basal layer of the corneal epithelium in central and peripheral regions of the cornea in adult wild-type and PAX77⁺ mice. C: Distribution of p63-expressing cells in the peripheral and central cornea of the basal layer of the corneal epithelium of wild-type and PAX77⁺ eyes in adults and postnatal day (P) 10 and P2 neonates. D: Immunohistochemical staining to visualize p63-positive cells in the wild-type corneal epithelium. Immunohistochemical staining to identify proliferating cells in the wild-type corneal epithelium that have incorporated BrdU. Terminal deoxynucleotidyltransferase-mediated UTP end-labelling (TUNEL) labeling (green) of apoptotic cells in the superficial layers of the wild-type and PAX77⁺ corneal epithelium. 4′,6-diamidine-2-phenylidole-dihydrochloride (DAPI) nuclear counterstain is visible.

Fig. 5

Cytokeratin-12 expression. A,B: Bar charts grouping samples by K12 expression in both peripheral and central regions of the cornea in both wild-type and PAX77⁺ mice in adults (A) and postnatal day 10 (B). Bars are divided to represent the frequency of positive, patchy, and negatively stained samples. C: Western blot of K12 expression in wild-type and PAX77⁺ corneas.

Fig. 6

Wound healing. A: Histogram showing change in wound diameter (i.e., amount of healing) at 12 and 24 hr after wounding, in wild-type, Pax6^+/-, and PAX77⁺ eyes cultured in serum-free culture medium ± 10 ng/ml epidermal growth factor (EGF). B: Representative images of a single wild-type eye immediately after wounding and 6 hr and 12 hr after wounding, showing complete closure of the wound after 12 hr. The wounds were topically stained with fluorescein to visualize the wound area. C: Western blot analysis of EGF receptor (EGFR) expression in wild-type and PAX77 whole cornea, showing increased expression of EGFR in the PAX77⁺. Each lane was pooled from four separate corneas.

Fig. 7

Lens defects. A-F: Histology of postnatal day (P) 2 wild-type (A,C,E) and PAX77⁺ (B,D,F) lenses, with hematoxylin and eosin staining (A,B), anti-Prox1 immunohistochemistry (C, D), and anti-β-crystallin (E,F). G: Western blot analysis of Pax6 and β-crystallins in the PAX77⁺ lens. H: Lens epithelial dysgenesis in a PAX77⁺ lens. Top of image is anterior. Green arrowheads represent lens bows of displaced and fragmented epithelium. A composite high-resolution image is presented as Supplementary Figure 2. I,J: Ectopic nuclei in adult PAX77⁺ lens fiber region are Pax6-positive (I) and, with low frequency, BrdU-positive. Scale bars = 500 μm in A,B, 50 μm in C,D, 320 μm in H, 50 μm in I, 10 μm in J.