Fig 1.
Expression of ANGPTL4 in the peripheral ischemic inner retina and within adjacent retinal neovascular tissue of PSR eyes.
(A) H&E staining of proliferative vessels (red arrows) observed on the surface of the peripheral retina of a PSR eye. (B) Non-staining IgG (negative control). (C) CD34 staining (vascular endothelial cells) of proliferative vessels (red arrows) overlying non-perfused (i.e., no CD34 staining of retinal vessels) peripheral retina (blue arrows). (D) ANGPTL4 staining within non-perfused (i.e., no CD34 staining of retinal vessels) peripheral retina (orange arrows) and in the vascular endothelial cells and within the stroma (red arrows) of the retinal neovascular tissue overlying the peripheral non-perfused retina. Similar results were obtained in 5/5 PSR eyes tested.
Fig 2.
Expression of ANGPTL4 in the peripheral ischemic retina distant from retinal neovascular tissue of sickle cell retinopathy eyes.
(A) H&E staining (top), IgG non-staining (middle; negative control), and CD34 staining (bottom) in the central (posterior) vs. peripheral (anterior) retina of a PSR eye in a region in which no adjacent retinal neovascularization is detected under low power magnification. Sparse CD34-positive inner retinal vessels, but no overlying neovascular tissue, are observed in the peripheral ischemic retina. (B) ANGPTL4 staining in the posterior vs. peripheral inner retina. Staining of ANGPTL4 is noted in the peripheral ischemic retina (red arrows) but not in the perfused posterior retina. Similar results were obtained in 5/5 PSR eyes tested.
Fig 3.
VEGF and ANGPTL4 expression in the aqueous of patients with active PSR.
(A and B) Aqueous levels of VEGF (aVEGF) (A) and ANGPTL4 (aANGPTL4) (B) were measured in patients with active PSR and compared to control patients, as well as patients with active PDR. (C) Correlation between aVEGF and aANGPTL4 in control, PSR, and PDR patients. Control: non-diabetic, non-sickle cell patients; PDR: patients with active PDR. Aqueous levels of ANGPTL4 and VEGF are reported as concentration (ng/mL) or fold induction (compared to the average level for non-diabetic, non-sickle cell control patients, arbitrarily assigned as 1). NS = not significant; * p < 0.05; ** p < 0.01.
Table 1.
Patient aqueous samples.
Fig 4.
VEGF and ANGPTL4 expression in the vitreous of patients with active PSR.
(A and B) Vitreous levels of VEGF (vVEGF) (A) and ANGPTL4 (vANGPTL4) (B) were measured in patients with active PSR and compared to control patients, as well as patients with active PDR. (C) Correlation between vVEGF and vANGPTL4 in control, PSR, and PDR patients. Control: non-diabetic, non-sickle cell patients; PDR: patients with active PDR. Vitreous levels of ANGPTL4 and VEGF are reported as concentration (ng/mL) or fold induction (compared to the average level for non-diabetic, non-sickle cell control patients, arbitrarily assigned as 1). NS = not significant; * p < 0.05; **** p < 0.0001.
Table 2.
Patient vitreous samples.
Fig 5.
Schematic illustration depicting the progression from peripheral non-perfusion to the development of neovascular sea fans in PSR.
(A) FA image from a patient with PSR demonstrating extensive peripheral non-perfusion and developing neovascular sea fans. Expanded view of local area of peripheral non-perfusion following occlusion of vessels by sickled red blood cells. (B) Schematic demonstrating inner retinal non-perfusion leading to Müller cell hypoxia. (C) Nuclear accumulation of HIF-1 leads to increased expression of both VEGF and ANGPTL4, which together act on retinal vascular endothelial cells to promote pathological angiogenesis. (D) This, in turn, results in the formation of neovascular sea fans at the margin between perfused and non-perfused retina. Inhibition of HIF-1 or both VEGF and ANGPTL4 could be an effective approach for the treatment of neovascular sea fans in sickle cell patients.