Genetic background-dependent role of Egr1 for eyelid development

Abstract

EGR1 is an early growth response zinc finger transcription factor with broad actions, including in differentiation, mitogenesis, tumor suppression, and neuronal plasticity. Here we demonstrate that Egr1^-/- mice on the C57BL/6 background have normal eyelid development, but back-crossing to BALB/c background for four or five generations resulted in defective eyelid development by day E15.5, at which time EGR1 was expressed in eyelids of WT mice. Defective eyelid formation correlated with profound ocular anomalies evident by postnatal days 1-4, including severe cryptophthalmos, microphthalmia or anophthalmia, retinal dysplasia, keratitis, corneal neovascularization, cataracts, and calcification. The BALB/c albino phenotype-associated Tyr^c tyrosinase mutation appeared to contribute to the phenotype, because crossing the independent Tyr^c-2J allele to Egr1^-/- C57BL/6 mice also produced ocular abnormalities, albeit less severe than those in Egr1^-/- BALB/c mice. Thus EGR1, in a genetic background-dependent manner, plays a critical role in mammalian eyelid development and closure, with subsequent impact on ocular integrity.

Keywords: Egr1; eyelid development; genetic background-specific effects; ocular abnormalities; tyrosinase.

Fig. 1.

Defective eye development in adult Egr1^−/− BALB/c mice. (A and B) External view of normal eyes from adult Egr1^+/−, Egr1^−/− C57BL/6 mice (A) and adult Egr1^+/+, Egr1^+/−BALB/c mice (B). (C) External view of eyes from adult Egr1^−/− BALB/c mice showing abnormal periocular shape and eye size and corneal opacity. (D) Histology of adult Egr1^+/+ BALB/c mice. (E–G) Histology of adult Egr1^−/− BALB/c adult mice at F5 from the initial backcross, showing anophthalmia (E), severe microphthalmia (F), and microphthalmia (G). (H–J) Abnormalities in adult Egr1^−/− BALB/c mice. In G–J, numbered arrows indicate the following: 1, retinal dysplasia; 2, cataract; 3, calcification; 4, corneal neovascularization; 5, keratitis. (Scale bars: D, 800 µm; E and F, 400 µm; G and H, 200 µm; I, 100 µm; J, 50 µm.)

Fig. S1.

Histology of adult Egr1^−/− BALB/c mice at F4 from the second backcross. (A) Anophthalmia. (B) Severe microphthalmia. (C) Retinal rosettes and folds. (D) Keratitis and synechiae (adhesion of lens and cornea). Numbered arrows indicate (1) retinal dysplasia; (2) retinal folds; (3) cataract; (4) lens capsule rupture. (Scale bars: A and B, 400 µm; C and D, 200 µm.)

Fig. 2.

Abnormal ocular development in postnatal Egr1^−/− BALB/c mice. (A–G) Histology of eyes from newborn Egr1^+/− mice (A), Egr1^−/− mice at postnatal day 1 (B) and 2 (C), and from Egr1^−/− mice at postnatal days 3 and 4 (D–G). In B and D–G, numbered arrows refer to the following: 1, mild corneal neovascularization; 2, mild keratitis; 3, cataracts; 4, corneal perforation; 5, vitreous hemorrhage; 6, retinal detachment; 7, inflammation; 8, residual remnant retina. (H–J) Eyelids of newborn Egr1^+/+ (H) and Egr1^+/− (I) mice are formed, whereas those of Egr1^−/− mice are not formed, and eyes appear open (J). [Scale bars: A, B (Left), and D, 800 µm; C, 400 µm; E (Left), F, and G, 200 µm; B (Right) and E (Right), 100 µm.]

Fig. 3.

Eyelid formation in embryos. (A, C, E) Egr1^+/− BALB/c mice. (B, D, F) Egr1^−/− BALB/c mice. Two embryos of each indicated genotype are shown at days E16.5 (A and B), E17.5 (C and D), and E18.5 (E and F). The arrows in A and B indicate areas of eyelid merging or defects therein. (Scale bars: A–F, 800 μm.)

Fig. 4.

Immunohistochemistry of Egr1 during embryogenesis. Two embryos of C57BL/6 WT mice at days E13.5 (A and B), E15.5 (C and D), and E17.5 (E and F) and of BALB/c WT mice at day E15.5 (G and H). A, C, E, and G show controls stained with the Egr1 Ab-blocking peptide mix, and B, D, F, and H show staining with the Egr1 antibody. In B, D, F, and H, Left are the same magnification as in panels A, C, E, and F, respectively, whereas B, D, F, and H, Right show a higher-magnification view of the relevant region of the left panel. [Scale bars: A, B (Left), C, D (Left), E, F (Left), G, and H (Left), 200 µm; B (Right), D (Right), F (Right), and H (Right), 100 µm.]

Fig. 5.

Gene-expression profiles in eyes from Egr1^+/− and Egr1^−/− BALB/c mice. (A) Il1b mRNA expression is significantly increased in adult Egr1^−/− BALB/c but not in Egr1^−/−C57BL/6 mice. (B) Il1b mRNA is increased in Egr1^−/− BALB/c newborn mice. RNA was normalized relative to the expression of Rpl7. Comparison between samples was done by the Student’s t test. (C–G) Gene-expression profiles in eyes from Egr1^+/− and Egr1^−/− BALB/c mice at postnatal day 3. (C) MDS analysis for individual samples from eyes from Egr1^+/− and Egr1^−/− mice with formed eyelids (indicated by a “C” for closed) vs. Egr1^−/− mice without eyelid formation (indicated by an “O” for open). Differences in dimension 1 (Dim1, x axis) indicates greater differences in samples than do differences in dimension 2 (Dim 2, y axis). (D) 2D scatter plot shows expressed genes. Genes highlighted in blue, green, and orange indicate differentially expressed genes with |log2FC|>3, (|log2FC|>2), and |log2FC|>1, respectively. “log2FC” refers to the log (base 2) of the fold change of mRNA expression in eyes with unformed eyelids vs. eyes with formed eyelids. log2CPM, log base 2 of counts per million. (E) Heat map of the top 50 differentially expressed genes. The expression matrix on rows is normalized by z-scores, and the colors of the heat map are mapped linearly to the z-scores. Samples ending in “O” were from open eyes of a given embryo; those ending in “C” were from closed eyes. (F) Volcano plot showing the top 50 differentially expressed genes. The x axis represents the log2FC between eyes from Egr1^−/− mice with unformed eyelids (open) vs. Egr1^−/− or Egr1^+/− mice with fused eyelids (closed), and the y axis shows the −log10 of the P value. (G) Bar plots show the expression of Il1b, Krt16, Sprr1a, and Egr1. OD, right eye; OS, left eye. E24, E28, E29, and E32 are the originally designations for these embryos based on when they were collected. RPKM, reads per kilobase of transcript per million mapped reads.

Fig. 6.

Histology of Egr1^−/−, Tyr^c-2J/c-2J C57BL/6 mice. (A) Normal structure of eyes from Egr1^−/−, Tyr^+/c-2J C57BL/6 adult mice. (B–D) Abnormalities in Egr1^−/−, Tyr ^c-2J/c-2J C57BL/6 adult mice. (B) Mild corneal neovascularization (1). (C) Poor corneal closure with epithelial ingrowth (2), retinal dysplasia (3), anterior angle dysplasia and ciliary body hypoplasia (4), and PHPV (5). The left and upper panels are higher- magnification views of the indicated areas. (D) Staphyloma, posterior coloboma, and PHPV. [Scale bars: A, C (Bottom), and D, 800 µm; B, 50 µm; C (Left and Top), 100 µm.]

Fig. S2.

Histology of eyes in Egr1^+/− BALB/c (A) and Egr1^−/− BALB/c (B) embryos at day E14.5. (Scale bars: A and B, 100 µm.)