Cooperation of Angiopoietin-2 and Angiopoietin-4 in Schlemm's Canal Maintenance

Figures

• 

  Figure 1.. Angpt4 expression in the mouse iridocorneal angle. (A) Angpt4-expressing cells visualized in cell fate–tracing Angpt4+/Cre;Rosa26mTmG/+ mice. Angpt4Cre-expressing cells are permanently marked with GFP (green), while Angpt4Cre-negative cells express mTomato (red). In corneal limbus whole mounts at P10, there are only very few GFP+ cells near the SC, while at P13, their number has increased. Areas framed with dashed lines are magnified in the boxes below. (B) The number of GFP+ cells further increases in adult corneal limbus area. SC is stained with CD31 (white). Mo, mouse age in months. (C) In sectioned tissue, GFP+ cells can be more precisely localized into the TM (arrows) and into the corneal stroma and sclera (arrowheads). Open arrowheads point to the GFP+ retinal (R) astrocytes. Area framed with dashed lines is magnified in the box on the left. Arrows in the insert point to projections of which cells of the narrow JCT region just beneath the SC make toward the SC endothelium (sparse dashed lines). CB, ciliary body; LVP, limbal vascular plexus. (D) Angpt4 expression visualized in the Angpt4LacZ reporter mice with X-gal staining (blue precipitate, arrowheads) at P19. (E–H) qPCR…

• 

  Figure 2.. Analyses of the SC area in 1-year-old Angpt4-deficient and Angpt1;Angpt4 double-deficient mice. (A) Representative closeup images of CD31 antibody staining (white) of whole-mount SC from WT and Angpt4−/− (A4−/−) mice. SC area (highlighted with solid white lines) was quantified in (B). n = 4 mice/genotype. (C) Ubiquitous Angpt1 deletion in Angpt1fl/fl;Angpt4Cre/Cre;Rosa26CreERT2 mice (A1del; A4−/− mice) was induced by weekly oral administration of tamoxifen at 4 to 8 weeks of age. Littermates received the same treatments (A4+/− mice are phenotypically normal). Representative closeup images of CD31 antibody staining (white) of whole-mount SC are shown. SC area (highlighted with solid white lines) was quantified in (D). n = 4/7/4/6 mice/genotype. **P < 0.01, ***P < 0.001 in one-way ANOVA followed by Tukey post hoc test in D.

• 

  Figure 3.. Analyses of SC morphology in 1-year-old Angpt2−/− and Angpt2−/−;Angpt4−/− mice. (A) Representative closeup images of CD31 antibody staining (white) of whole-mount SC. SC area (highlighted with solid white lines) was quantified in (B). n = 11/8/8/8 mice/genotype. (C) Representative images of whole-mount corneal limbus area. CD31 (magenta) strongly labels the SC and limbal blood vasculature and more faintly the LVs, and Lyve1 (green) distinctly marks LVs (open arrowheads) and also limbal macrophages. A, circular limbal artery; V, perilimbal vein. Arrowheads mark SC narrowing points quantified in (D), asterisks mark collecting vein branches quantified in (E), arrows mark corneal arcades of the limbal capillary plexus quantified in (F), and Lyve1+ corneolimbal LVs were quantified in (G). n = 7/8/4/5 mice/genotype in D, n = 13/8/10/9 mice/genotype in E, n = 6/8/4/5 mice/genotype in F, and n = 7/5/5/4 mice/genotype in G. *P < 0.05, **P < 0.01, ***P < 0.001 in one-way ANOVA followed by Tukey post hoc test in B and D–F or in Welch's ANOVA followed by Dunnett's T3 post hoc test in G.

• 

  Figure 4.. Analysis of glaucoma-like changes in 1-year-old Angpt2−/− and Angpt2−/−;Angpt4−/− mice. (A) IOP measurements. n = 9/7/11/7 mice/genotype. (B, C) Transmission electron microscopy images (B) were used to analyze GV number (arrowheads) in the SC inner wall endothelium in (C). n = 6/5/3/6 mice/genotype. (D) Hematoxylin and eosin–stained sections were used to analyze RGC (arrow) number (E) and RNFL (opposite arrowheads) thickness (F). Red bracket highlights RGC-deprived area in the Angpt2−/−;Angpt4−/− mice. n = 13/5/6/7 mice/genotype. *P < 0.05, **P < 0.01, ***P < 0.001 in one-way ANOVA followed by Tukey post hoc test.

• 

  Figure 5.. Analysis of Tie2 activation in the SC endothelium of Angpt2−/− and Angpt2−/−;Angpt4−/− mice. (A) Representative images of SC (highlighted with solid white lines) from 12-week-old mice stained with total Tie2 (green) and phosphorylated Tie2 (pTie2; red). Areas separated with dashed lines are magnified in the small boxes. (B) Quantification of pTie2/Tie2 immunofluorescence intensity ratio in the SC. n = 6–7 mice/genotype. *P < 0.05, ***P < 0.001 in one-way ANOVA followed by Tukey post hoc test.

• 

  Figure 6.. Analyses of SC morphology in 1-year-old Angpt2+/− and Angpt2+/−;Angpt4−/− mice. (A) Representative closeup images of CD31 antibody staining (white) of whole-mount SC. SC area (highlighted with solid white lines) was quantified in (B). n = 7/10 mice/genotype. (C) Representative images of whole-mount corneal limbus area. CD31 (magenta) strongly labels the SC and limbal blood vasculature and, more faintly, the LVs, and Lyve1 (green) distinctly marks the LVs (open arrowheads) and also limbal macrophages. A, circular limbal artery; V, perilimbal vein. Arrowheads mark SC narrowing points quantified in (D), asterisks mark collecting vein branches quantified in (E), arrows mark corneal arcades of the limbal capillary plexus quantified in (F), and Lyve1+ corneolimbal LVs were quantified in (G). n = 7 mice/genotype in D–G. *P < 0.05, **P < 0.01 in unpaired Student's two-tailed t-test.

• 

  Figure 7.. Analyses of glaucoma-like changes and Tie2 activation in 1-year-old Angpt2+/− and Angpt2+/−;Angpt4−/− mice. (A) IOP measurements. n = 7/9 mice/genotype. (B, C) GV number (arrowheads) in the SC inner wall endothelium was analyzed from transmission electron microscopy images (B) and quantified in (C). n = 4 mice/genotype. (D–F) RGC (arrow) number and RNFL (opposite arrowheads) thickness was quantified from hematoxylin and eosin–stained sections. n = 6/9 mice/genotype. (G) Representative images of SC (highlighted with solid white lines) stained with total Tie2 (green) and phosphorylated Tie2 (pTie2, red). (H) Quantification of pTie2/Tie2 immunofluorescence intensity ratio in the SC. n = 5/6 mice/genotype. **P < 0.01 in unpaired Student's two-tailed t-test.