Retinal Angiogenesis in Methamphetamine Self-Administration Rats

Abstract

Purpose: Given the evidence of a link between methamphetamine (METH) exposure and retinal vascular abnormalities, this study aims to investigate the molecular and cellular mechanisms underlying METH-induced retinal angiogenesis using a unique self-administration rat model and primary rat retinal microvascular endothelial cells (RRMECs).

Method: To model the impact of compulsive use of METH, rats underwent an 8-week METH long-access self-administration protocol, with retinal tissues analyzed using whole retinal flatmount imaging and vascular network quantification. Proteomic analysis via liquid chromatography/tandem mass spectrometry identified differentially expressed proteins, while RRMECs were treated with METH to assess molecular changes through immunoblotting and quantitative RT-PCR.

Results: Consistent with compulsive use of METH in humans and our previous experience with this model, rats self-administered high levels of METH. METH self-administration elevated dopamine levels in the vitreous humor and increased vascular density in both superficial and deep capillary layers across central, mid-peripheral, and peripheral retina regions. Proteomic analysis revealed 148 differentially expressed retinal proteins, with gene ontology enrichment highlighting pathways related to abiotic stimuli, hypoxia, and ischemia. Increased hypoxia inducible factor-1α (HIF-1α) and vascular endothelial growth factor a (VEGFa) expression confirmed a hypoxia-driven angiogenesis process, further supported by in vitro experiments showing enhanced endothelial cell proliferation and HIF-1α/VEGFa expression. Additionally, TAAR-1 upregulation in both the retina and endothelial cells was observed, with TAAR-1 antagonism reducing METH-induced endothelial cell proliferation and modulating HIF-1α/VEGFa signaling.

Conclusions: METH self-administration leads to significant retinal vascular changes and angiogenesis, driven by upregulation of hypoxia-related pathways. TAAR-1 plays a critical role in endothelial cell proliferation through the HIF-1α/VEGFa pathway, potentially contributing to pathological retinal conditions.

Figure 1.

METH self-administration rats. (A) Weekly number of infusions in METH self-administration rats. (B) Weekly amount of METH intake in METH self-administration rats. (C) Comparison of average number of infusions during the initial and subsequent 4-week periods in METH self-administration rats. (D) Comparison of average METH intake during the initial and subsequent 4-week periods in METH self-administration rats. (E) Total number of infusions in self-administration rats. (F) Dopamine level in vitreous humor of self-administration rats. n = 4; mean ± SE. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 2.

Retinal vascular density in self-administration rats. The fixed eyes were stained with GSL-1. The vascular density in the superficial vascular layer (A–C) and the deep capillary layer (D–F) of central (A, D), mid-peripheral (B, E), and peripheral (C, F) retina was measured by AngioTools. n = 3; mean ± SE. *P < 0.05 and **P < 0.01 vs. saline.

Figure 3.

Different protein expressions of retina in METH self-administration rats. (A) A total of 1688 proteins were identified by LC-MS/MS spectrometry, and 148 proteins showed significant changes in METH self-administration rats (n = 4). (B) Principal component analysis plot of significantly differential expressed proteins in METH self-administration rats (blue circles) and saline self-administration rats (red circles). (C) Heatmap for the 148 significantly changed proteins (P < 0.05) in the retina samples. (D) GO biological process categories and (E) GO Process-Protein network associated with retinal angiogenesis were analyzed using the STRING Functional Enrichment Analysis database.

Figure 4.

Protein levels of HIF-1α and VEGFa in retina in response to METH versus saline. The relative protein expression levels of HIF-1α (A) and VEGFa (B) were normalized to β-actin expression using ImageJ (n = 4). Mean ± SE. **P < 0.01 vs. saline.

Figure 5.

Role of HIF-1α in METH-administrated RRMECs. (A) The effect of METH in cell proliferation was measured by the MTS cell proliferator assay (n = 5–6; mean ± SE, *P < 0.05 vs. control). (B) The relative protein levels of HIF-1α and VEGFa in METH-administrated RRMECs were normalized to β-actin expression using ImageJ (n = 3; mean ± SE, *P < 0.05 vs. control). (C) The relative mRNA expression levels of Hif1α and Vegfa in METH-administrated RRMECs were normalized to Mapk1 expression (n = 5–6; mean ± SE, *P < 0.05 vs. control). The effect of GN44028 on the cell proliferation in RRMECs (D) and METH-administrated RRMECs (E) was measured by the MTS cell proliferator assay (n = 6; mean ± SE, *P < 0.05, **P < 0.01, and ***P < 0.001 vs. comparison). (F) The effect of GN44028 on the Hif1α and Vegfa mRNA expressions in METH-administrated RRMECs was analyzed by quantitative RT-PCR (n = 5–6; mean ± SE, ***P < 0.001 vs. comparison). (G) The effect of GN44028 on the HIF-1α and VEGFa protein expressions in METH-administrated RRMECs was measured by immunoblot using ImageJ (n = 3; mean ± SE, *P < 0.05 and **P < 0.01 vs. comparison).

Figure 6.

Role of TAAR-1 in METH self-administration rats and METH-administrated RRMECs. (A) The effect of METH in retinal TAAR-1 protein expression in RRMECs was normalized to β-actin expression using ImageJ (n = 4; mean ± SE, *P < 0.05 vs. control). (B) The relative protein level of TAAR-1 in METH-administrated RRMECs was normalized to β-actin expression using ImageJ (n = 3; mean ± SE, *P < 0.05 vs. control). (C) The relative mRNA expression levels of Taar1 in METH-administrated RRMECs were normalized to Mapk1 expression (n = 6; mean ± SE, ***P < 0.001 vs. control). The effect of EPPTB on the cell proliferation in RRMECs (D) and METH-administrated RRMECs (E) was measured by the MTS cell proliferator assay (n = 5–6; mean ± SE, *P < 0.05 and ***P < 0.001 vs. comparison). (F) The effect of EPPTB on the Hif1α and Vegfa mRNA expressions in METH-administrated RRMECs was normalized to Mapk1 expression (n = 5–6; mean ± SE, ***P < 0.001 vs. comparison). (G) The effect of GN44028 on the HIF-1α and VEGFa protein expressions in METH-administrated RRMECs was normalized to β-actin expression using ImageJ (n = 3; mean ± SE, *P < 0.05 and **P < 0.01 vs. comparison).