Myeloid cell modulation by a GLP-1 receptor agonist regulates retinal angiogenesis in ischemic retinopathy

Abstract

Ischemic retinopathies including diabetic retinopathy are major causes of blindness. Although neurons and Müller glia are recognized as important regulators of reparative and pathologic angiogenesis, the role of mononuclear phagocytes (MPs) - particularly microglia, the resident retinal immune cells - is unclear. Here, we found MP activation in human diabetic retinopathy, especially in neovessels from human neovascular membranes in proliferative retinopathy, including TNF-α expression. There was similar activation in the mouse oxygen-induced retinopathy (OIR) model of ischemia-induced neovascularization. Glucagon-like peptide-1 receptor (GLP-1R) agonists are in clinical use for glycemic control in diabetes and are also known to modulate microglia. Herein, we investigated the effect of a long-acting GLP-1R agonist, NLY01. Following intravitreal administration, NLY01 selectively localized to MPs in retina with OIR. NLY01 modulated MPs but not retinal endothelial cell viability, apoptosis, and tube formation in vitro. In OIR, NLY01 treatment inhibited MP infiltration and activation, including MP expression of cytokines in vivo. NLY01 significantly suppressed global induction of retinal inflammatory cytokines, promoted reparative angiogenesis, and suppressed pathologic retinal neovascularization. Collectively, these findings indicate the important role of mononuclear phagocytes in regulation of retinal vascularization in ischemia and suggest modulation of MPs as a potentially new treatment strategy for ischemic retinopathies.

Keywords: Ophthalmology; Retinopathy.

Figure 1. Microglial/macrophage activation in human diabetic retina, PDR membrane, and mice OIR retina.

(A) MPs were immunostained with IBA1 antibody (blue) in nondiabetic and NPDR human eyes. In nondiabetic retina, ramified microglia (blue arrowhead) are presented in inner retinal layers. Hypertrophic MPs (green arrowhead) were observed in NPDR retina. n = 3. Scale bar: 100 μm. (B) Representative images of immunofluorescence (IF) staining of microglia/macrophages (red) in nondiabetic (left), NPDR (middle), and PDR (right) human eyes. In the nondiabetic control cases, IBA1 staining was weak. There were significantly more amoeboid microglia (yellow arrowheads) in both NPDR and PDR retinas. Blue: DAPI staining. (n = 6, nondiabetic eyes; n = 5 NPDR eyes; n = 6 PDR eyes). Scale bar: 100 μm. (C) Representative images of IHC staining of IBA1⁺ microglia/macrophages (blue) in membranes from patients with active PDR. Images were taken using a ×63 objective. Nuclear fast red was used for counterstaining. n = 3. Scale bar: 100 μm. (D) Representative IF staining of MPs (red) and TNF-α (green) in PDR membranes. Blue: DAPI staining. Scale bar: 100 μm. (E) Schematic diagram of OIR. Neonatal mice were exposed to 75% oxygen from P7 to P12 to induce vaso-obliteration and returned to room air from P12 to P17 to induce physiological revascularization and pathological NV, which reached maximum at P17. (F) Representative IF images of vasculature (IB4, red) and IBA1⁺ MPs (green) in retinal flat mounts from P17 OIR and room air (n = 3 mice). Scale bar: 100 μm.

Figure 2. NLY01 regulates LPS-stimulated cultured microglia activation.

(A) Representative images of NF-κB p65 (red) staining in LPS-stimulated primary brain microglia pretreated with vehicle or NLY01 (1 μM). Blue: DAPI staining. Scale bar: 50 μm. (B) Representative (top panel) and quantitation (bottom panel) Western blotting analysis of NF-κB p65 nuclear translocation in LPS-stimulated primary brain microglia with vehicle or NLY01 pretreatment. *P < 0.05. One-way ANOVA test was used for statistical analysis. Data are presented as mean ± SD (n = 4). The experiment was conducted 3 times with similar results. (C) Expression of inflammatory genes in LPS-stimulated BV2 cells with vehicle or NLY01 pretreatment. *P < 0.05. One-way ANOVA test was used for statistical analysis. Data are presented as mean ± SD (n = 6).

Figure 3. NLY01 does not exhibit a direct antiangiogenic effect on HRECs.

HRECs were treated with NLY01 (1 μM) for 24 hours under hypoxic conditions. Cell viability and apoptosis were quantified by (A) CellTiter-Glo (CTG) luminescent cell viability assay (n = 10) and (B) Caspase 3/7 assay (n = 3), respectively. (C) Representative images and (D) quantitation of HREC tube formation with vehicle or NLY01 treatment (n = 3). An unpaired, 2-tailed Student’s t test was used for statistical analysis. Data are presented as mean ± SD. Scale bar: 100 μm.

Figure 4. NLY01 colocalizes with MPs in OIR retina.

FITC-NLY01 was injected intravitreally on P12. Representative images show the staining of FITC-NLY01 (green), NLY01 (blue), and CD31 (red). Enlarged images show FITC-NLY01 (green) colocalized with IBA1⁺ (blue) microglia/macrophages in both avascular areas and around neovascular tufts on (A) P15 and (C) P17 (n = 4). Scale bar: 100 μm. (B) 3D reconstruction image of FITC-NLY01 colocalization with MPs around the neovascular tufts in OIR retina. White arrowhead: FITC-NLY01 (green). Scale bar: 20 μm.

Figure 5. NLY01 modulates MP activation in OIR retina.

(A) Representative images of vasculature (CD31, red) and microglia (IBA1, green) staining on retinal flat mounts from vehicle- or NLY01-treated eyes on P17 OIR (n = 3). Scale bar: 100 μm. (B) Representative flow cytometric plots and quantifications of CD11b⁺CD45^lo cells in P15 vehicle- and NLY01-treated retinas (n = 5). Each data point represents the value from 2 retinas. (C) Schematic diagram of RiboTag strategy. (D) Microglia/macrophage polyribosome RNA enrichment in the HA-pulldown samples. Input samples served as the control (n = 6). (E) Changes in inflammatory gene expression in MP polyribosome RNA enriched from vehicle- or NLY01-treated Cx3cr1CreER:Rpl22HA mice retinas (n = 7). Each data point represents the value for a single retina. *P < 0.05; **P < 0.01. An unpaired, 2-tailed Student’s t test was used for statistical analysis. Data are presented as mean ± SD.

Figure 6. NLY01 reduces inflammatory and angiogenesis-related gene expression in OIR retina.

(A) Inflammatory gene expression in vehicle- and NLY01-treated eyes. (B) Angiogenesis-related gene expression in vehicle- and NLY01-treated groups. (C) Gliosis-related gene expression (n = 7). Each data point represents the value for a single retina. *P < 0.05. Statistical analysis was done via unpaired, 2-tailed Student’s t test. Data are presented as mean ± SD.

Figure 7. NLY01 treatment reduces avascular retinal area and neovascular tufts in OIR.

(A) Representative retinal flat mounts from OIR P17 vehicle- or NLY01-treated retinas stained with IB4. (B) Quantitation of avascular area and NV tufts shown as a scatter plot–bar graph with unpaired analysis (n = 12). Each data point represents the value for a single mouse. **P < 0.01; ***P < 0.001. An unpaired, 2-tailed Student’s t test was used for statistical analysis. Data are presented as mean ± SD. (C) Quantitation of avascular area and NV tufts shown as a symbol-line graph with paired analysis (n = 12). Each data point represents the value for a single retina. **P < 0.01; ***P < 0.001. A paired, 2-tailed Student’s t test was used for statistical analysis. Scale bar: 1000 μm.