Drugless peptide-based nanohybrids alleviate diabetic retinopathy by suppressing microglial activation and endothelial inflammation

Abstract

Background: Diabetic retinopathy (DR) is a vision-threatening microvascular complication of diabetes mellitus. Chronic inflammation and endothelial dysfunction are critical factors in the disease's pathogenesis. Consequently, interventions developed to reduce retinal inflammation are anticipated to be beneficial for both the prevention and treatment of DR. In the present study, we developed a unique class of drugless peptide-based nanohybrids with potent anti-inflammatory activities and investigated their therapeutic efficacy for treating DR in an oxygen-induced retinopathy (OIR) mouse model and a streptozotocin (STZ)-induced diabetic mouse model. Methods: Hexapeptides were applied to modify gold nanoparticles to form the drugless peptide-based nanohybrids (P12). We then examined the physicochemical properties and anti-inflammatory activities of P12 in HUVECs and BV2 cells and identified the critical amino acids for this novel bioactivity. The intravitreal and retro-orbital injections were applied to determine the optimal retinal delivery route for P12. The therapeutic efficacy of P12 in treating DR were investigated using both the OIR model and STZ-induced diabetic model. Through immunohistochemistry and flow cytometry analyses, we identified the major cells that internalize P12 in the retina. Furthermore, in vitro experiments were used to explore the underlying molecular mechanisms for the anti-inflammatory activities of P12. Results: We found that P12 exhibited potent anti-inflammatory effects in both HUVECs and BV2 cells. In addition, P12 can be efficiently delivered to the retina via intravitreal injection. Intravitreally injected P12 significantly improved early DR symptoms including vascular leakage and pericyte loss in STZ-induced diabetic mice. It also suppressed pathological neovascularization and retinal hemorrhage in OIR mice. Importantly, we found that intravitreally injected P12 was mainly taken up by microglial and endothelial cells, leading to reduced retinal endothelium inflammation and microglial activation in DR animal models. Mechanistic studies revealed that P12 potently inhibited several TLR4 downstream signaling pathways, such as NF-κB, JNK, and P38 MAPK, in both endothelial and microglial cells. This effect is due to the capacity of P12 in blocking the endosomal acidification process that governs the endosomal TLR signaling transduction. Conclusions: Our findings suggest that local injection of properly designed, drugless, peptide-based nanohybrids can serve as a safe and effective anti-inflammatory nanomedicine for treating DR.

Keywords: Toll-like receptors; bioactive nanoparticles; diabetic retinopathy; endothelial inflammation; microglia activation.

Figure 1

Screening of anti-inflammatory nanoparticles in microglia and endothelial cells. (A) The construction of the peptide-GNP hybrids P12. (B) TEM image of Bare GNP and P12. Scale bar = 50 nm. (C) The size distribution of the bare GNP and P12 measured by DLS. (D) The zeta-potential of the bare GNP and P12. N ≥ 3, *p < 0.05. (E, F) ELISA measurement of VCAM-1 and IL-6 levels in the supernatant of HUVEC (E) and BV2 (F) cells with various peptide-GNP hybrids treatment. (G, H) Intravitreal vs. retro-obital delivery of Cy5-labeled P12. Fluorescence of the eyeballs was imaged at 12 h after indicated injection. (I) TEM images showing the distribution of P12 in mouse retina at 6 h, 12 h, 14 d, and 28 d after intravitreal injection. P12 was found in nerve fiber layer (black star) at 6 h,12 h and even 28 d after injection, and was found in pericyte (white star) surrounding the basement layer (black arrowhead) of microvascular in the retina at 14 d after injection. Scale bar = 200 nm.

Figure 2

ivP12 ameliorated retinal vascular leakage and retinal inflammation in STZ-induced diabetic mice. (A) Schematic depiction of the induction of STZ-induced diabetic mouse model. (B, C) Measurement of blood glucose levels (B) and body weight (C) of PBS and P12 treated mice for 20 weeks post-STZ treatment. (D) Retinal vascular permeability was assessed by FITC-dextran assay in PBS and P12 treated mice at 24 weeks post-STZ injection. Scale bar=500 μm. (E) Quantification of FITC-dextran permeation in PBS and P12 treated mice. (F) Pericyte coverage of retinal vessels was assessed by immunostaining of desmin in retinal flatmounts from PBS and P12 treated mice at 24 weeks post-STZ injection. The black arrows denote the regions of the retinal vessels where pericytes were lost. Scale bar=200 μm. (G) The protein levels of the inflammatory cytokines and adhesion molecules in retina samples of STZ mice with or without P12 treatment.

Figure 3

P12 suppressed retinal neovascularization and vascular leakage in the OIR mouse model. (A) Schematic depiction of intravitreal P12 treatment in OIR pups. (B-D) P12 markedly decreased the retinal neovascularization and avascular area in the retina of OIR mice. (E) Gross examination of retinas isolated from OIR eye-cups. Black arrowhead depicts the area of vascular leakage (blood island). Scale bar = 500 μm.(F) Quantification of the area of blood island in PBS and P12 treated retinas. (G) Immunostaining of the erythrocyte marker TER119 in retinal flatmount of PBS and P12 treated mice with OIR. (H) Quantification of retinal vascular leakage by calculating the area of extravascular TER119⁺ cells in retinal flatmount. (I, J) P12 markedly decreased the degree of FITC- dextran leakage in the retina of OIR mice. Scale bar = 300 μm.

Figure 4

Identification of P12 targeted cells in the retina of OIR mice and STZ mice. (A) The construction of Cy5-labelled P12 nanoparticles. (B-D) Co-immunostaining of P12-Cy5 with F4/80 (B), Iba-1 (C), and CD31 (D) revealed colocalization of P12-Cy5 with microglia/macrophage and endothelial cells in the retina of OIR mice. (E) Gating strategy to identify microglia and endothelial cells within the retina of OIR and STZ mouse. (F-I) Uptake of P12 nanoparticle by microglia/macrophage (F, H) and endothelial cells (G, I) in the retina of OIR mice. (J-M) Uptake of P12 nanoparticle by microglia/macrophage (J, L) and endothelial cells (K, M) in the retina of STZ mice. Scale bar = 20 μm.

Figure 5

P12 treatment inhibited microglia activation and endothelial inflammation in the retina of OIR mice. (A, B) The fluorescence images of OIR retinal tissues revealed the reduced infiltration of F4/80 stained microglial/macrophages under the P12 treatment when compared with the untreated ones. Scale bar = 500 μm. (C, D) Flow cytometry analysis showed reduced levels of CD11b⁺CD45⁺ cells in retina treated with P12 compared to ctrl. (E, F) Western blot and densitometry analysis of the activation of MAPK (p38, JNK, and ERK) signaling in OIR retinas upon PBS or P12 treatment. (G) ELISA measurement of inflammatory cytokines and adhesion molecules in OIR retinas with indicated treatments. The assays shown are representative of at least 3 experiments with similar results, each experiment includes 1 retina sample per group. (H) P12 attenuated TNFα-induced gene expression of adhesion molecules in HUVEC as determined by qPCR. (I) Representative images of fluorescent THP-1 monocytes adherence to HUVEC with the indicated treatments. Scale bar = 100 μm. (J) Adherent THP-1 cells were counted in four randomly selected fields (20× objective) for each treatment group.

Figure 6

P12 inhibited the activation of NF-kB and MAPK signaling in HUVECs and BV2 cells. (A, B) Representative Western blots and densitometry analysis of phospho-NF-kB p65 and NF-kB p65 in HUVECs upon LPS and/or P12 treatment. (C-F) P12 mitigated LPS-induced phosphoactivation of JNK (D), ERK (E), and p38 (F) MAPK signaling in HUVECs. (G, H) Representative Western blots and densitometry analysis of phospho-NF-kB p65 and NF-kB p65 in BV2 cells upon LPS and/or P12 treatment. (I-L) P12 mitigated LPS-induced phosphoactivation of JNK (J), ERK (K), and p38 (L) MAPK signaling in BV2.

Figure 7

P12 exerts its anti-inflammatory effect in HUVECs and BV2 cells via inhibiting endosome acidification. (A) Immunofluorescence image showing the uptake of Cy5-labeled P12 in HUVECs. (B) Dose-dependent uptake of P12-Cy5 in HUVECs. (C) Immunofluorescence image showing the uptake of Cy5-labeled P12 in BV2. (D) Dose-dependent uptake of P12-Cy5 in BV2. (E, F) Confocal images of HUVECs (E) and BV2 (F) cells treated with the nanoparticles and a well-known pH modulator chloroquine (CQ). Endosomal pH was probed with pHrodo red- and fluorescein-labeled dextrans. Scale bar = 20 μm. (G, H) Changes in endosomal pH in HUVECs (G) and BV2 cells (H) were quantified as the fluorescence intensity ratios of green-to-red signals. (I-L) qPCR analysis of inflammatory gene expression in HUVECs (I, J) and BV2 (K, L) cells treated with the nanoparticles and CQ.

Figure 8

Toxicity assessment of intravitreal injection of P12. (A, B) Representative images of TUNEL assay of retinal sections from PBS (Ctrl) or P12-treated mice after 24 h of indicated treatments. Positive control was retinas treated with DNase I. (C) Histopathological analysis of other organs by HE staining in mice intravitreally injected with PBS or P12. Scale bar = 100 μm.