Inhibition of retinal angiogenesis by α-defensins. After 5 days of hyperoxia, mice were brought back to room air on p12 and were injected intraperitoneally or subconjunctivally with different treatments at different time points. (A) Typical photographs from lectin staining of retina whole-mount from a mouse that was treated with buffer alone or a mouse treated with synthetic α-defensin (HNP-1) intraperitoneally from p14 to p16 are shown. Arrowheads indicate the neovascular tufts. Neovascular regions are far more extensive in retinas from buffer-treated mice than in retinas from α-defensin–treated mice. (B) Retinal neovascularization was quantified on p17, as described in “Materials and methods.” Mice were treated from p14 until p16 once daily intraperitoneally with buffer, synthetic α-defensin (30 μg/mouse), blocking antibody to α5-integrin (30 μg/mouse), combination α-defensin and antibody to α5-integrin (each 30 μg/mouse), or β-defensin (30 μg/mouse). Retinal neovascularization is presented as percentage of control, defined as neovascularization in the presence of buffer only. Data are mean ± SD (n = 5 mice). (C) Retinal neovascularization was quantified on p17. Mice were treated on p14 with a subconjuctival injection into one eye of buffer or synthetic α-defensin (5 μg/eye). Retinal neovascularization is presented as percentage of control, defined as neovascularization in the presence of buffer only. Data are mean ± SD (n = 5 mice). * P < .05 compared with buffer; ns indicates not significant. (D) Induction of apoptosis by α-defensins in the ROP. Apoptosis in the retinas of 4 mice treated with buffer or 4 mice treated with α-defensin (30 μg/mouse) intraperitoneally from p14 until p17 was quantified. Apoptosis was determined by TUNEL by counting 15 sections per eye and is shown as number of apoptotic nuclei per section. *P < .05 compared with buffer.