Retinal hypoxia and angiogenesis with methamphetamine

Abstract

Central retinal artery occlusion, retinopathy, and retinal neovascularization have been reported in methamphetamine (METH) abusers. In the current study, we investigated whether METH induces retinal neovascularization in a mouse model, and if so, whether the neovascularization is associated with increased hypoxia, hypoxia-inducible factor 1α (HIF-1α), and vascular endothelial growth factor (VEGF). Mice were administrated METH by intraperitoneal injection over a 26-day period, or injected with saline as a vehicle control. The number of retinal arterioles and venules were counted using in vivo live imaging following infusion with fluorescein isothiocyanate-dextran. Excised retinas were stained with griffonia simplicifolia lectin I and flat mounted for a measurement of vascularity (length of vessels per tissue area) with AngioTool. Retinal hypoxia was examined by formation of pimonidazole adducts with an anti-pimonidazole antibody, and HIF-1α and VEGFa protein levels in the retina were detected by immunoblot. METH administration increased vascularity (including the number of arterioles) measured on Day 26. Retinal VEGFa protein level was not changed in METH-treated mice on Day 5, but was increased on Day 12 and Day 26. Hypoxia (pimonidazole adduct formation) was increased in retinas of METH-treated mice on Day 12 and Day 26, as were HIF-1α protein expression levels. These results indicate that METH administration induces hypoxia, HIF-1α, VEGFa, and angiogenesis in the retina.

Keywords: Angiogenesis; HIF-1α; Hypoxia; Methamphetamine; VEGF.

Figure 1.

Effect of METH on retinal vascularity. Fixed eyes were stained with GSL-1 and flat mounted for microscopic photographs. Graphs provide total retinal blood vessel length per unit area as measured by AngioTool, with N=3–5 per group. Panels A, mid-peripheral retina; B, peripheral retina. Scale bar = 100 μm. *p<0.05 and ***p<0.001 vs saline-injected controls.

Figure 1.

Effect of METH on retinal vascularity. Fixed eyes were stained with GSL-1 and flat mounted for microscopic photographs. Graphs provide total retinal blood vessel length per unit area as measured by AngioTool, with N=3–5 per group. Panels A, mid-peripheral retina; B, peripheral retina. Scale bar = 100 μm. *p<0.05 and ***p<0.001 vs saline-injected controls.

Figure 2.

A. Number of retinal arterioles per venule in control (saline-) and METH-injected mice at Day 26. N=7 per group. **p<0.01. B. Example of arteriolar and venular filling of fluorescent dye to determine the number of primary arterioles (filling first; numbered A1-A7) and venules (filling subsequently).

Figure 3.

Effect of METH on protein expression of VEGFa in the retina. VEGFa protein expression in the retina at Day 5 (A, N=3 per group), Day 12 (B, N=5 per group), and Day 26 (C, N=6–8 per group) was examined by immunoblot. The relative protein expression level of VEGFa was normalized to β-actin expression using ImageJ. *p<0.05 vs saline-injected controls.

Figure 4.

Detection of retinal hypoxia. As a positive control, retinal hypoxia induced by CCAO in mice was detected with formation of pimonidazole (Hypoxyprobe) adducts in retinal lysates on nitrocellulose membranes (A). The fixed eyes were stained with Hypoxyprobe primary antibody (red) and GSL-1 (green) and dissected and flat mounted for microscopic photographs (B).

Figure 5.

Effect of METH on hypoxia in the retina at Day 12. Pimonidazole (Hypoxyprobe) adducts from retina whole lysates were detected on nitrocellulose membrane (A). The relative pimonidazole adduct level was normalized to total protein using Image J (B). *p<0.05 vs saline-injected controls; N=3. The fixed eyes were stained with Hypoxyprobe primary antibody (red) and GSL-1 (green) and dissected and flat mounted for microscopic photographs (C). Scale bar = 1 mm.

Figure 6.

Effect of METH on hypoxia in the retina at Day 26. Pimonidazole (Hypoxyprobe) adducts from retina whole lysates were detected on nitrocellulose membrane (A). The relative pimonidazole adduct level was normalized to total protein using Image J (B). *p<0.05 vs saline-injected controls; N=4. The fixed eyes were stained with Hypoxyprobe primary antibody (red) and GSL-1 (green) and dissected and flat mounted for microscopic photographs (C). Scale bar = 1 mm.

Figure 7.

Effect of METH on protein expression of HIF-1α in the retina, examined by immunoblot. The relative protein expression level of HIF-1α was normalized to β-actin expression using ImageJ. (A) Day 12 (N=5), (B) Day 26 (N=8). ***P < 0.001 vs saline-injected controls.