Rapamycin prevents retinal neovascularization by downregulation of cyclin D1 in a mouse model of oxygen-induced retinopathy

The experimental procedures performed on mice were approved by Tianjin Medical University, Laboratory Animal Care and Use Committee; and the study protocol followed the Association for Research in Vision and Ophthalmology (ARVO) for the Use of Ophthalmic Animals. Forty-two 7-day-old C57BL/6 J mice (Academy of Military Medical Science, Beijing, China) were randomly divided into normoxia control group (CON) (14 mice), OIR group (14 mice), and RAPA group (14 mice). OIR model was induced in OIR and RAPA groups according to method described by Smith et al. [8]. Briefly, 7-day-old C57BL/6 mice were exposed to 75% oxygen for 5 days, then abruptly returned to room air. Mice in RAPA group were treated with RAPA (dissolved in 2% carboxymethylcellulose, 2 mg/kg/d) by intraperitoneal injection every day from postnatal day 12(P12) to P17. And mice in OIR group were treated with the same volume of the vehicle (carboxymethylcellulose). Mice were fed commercial mouse food and were allowed access to water freely in a room with a 12 h light/12 h dark cycle. The experiments were performed on P17.

Retinal flat mounts were used to show the non-perfused areas and neovascularization in retina. Four animals from each of the three groups were anesthetized with pentobarbital sodium (50 mg/Kg) by intraperitoneal injection. Mice were perfused with fluorescein isothiocyanate (FITC)-dextran (Sigma, St. Louis, MO, USA) through left ventricle. Then eyes were enucleated after euthanasia (intraperitoneal injection with pentobarbital sodium, 800 mg/Kg) and fixed in 4% paraformaldehyde at 4 °C for 12 h. Retinas were isolated, flat-mounted on glycerol/gelatin-coated glass slides, and viewed by fluorescent microscope (Zeiss, Oberkochen, Germany), and photographed. Areas of retinal nonperfusion were quantified by Image-Pro plus 6.0 analysis software for statistical analysis.

The severity of RNV was quantified by counting the number of vascular cell nuclei broke through the internal limiting membrane (ILM) into the vitreous. For the orientation, two eyes (one eye of each animal) were selected from each group then enucleated and placed in 4% paraformaldehyde at 4 °C for 24 h, after that they were embedded in paraffin. Serial 5-μm sections (each separated by at least 30 μm) through the cornea and parallel to the optic nerve were prepared, stained with hematoxylin and eosin (H&E), and viewed by light microscopy (OLYMPUS Optical Co., Ltd., Japan), for the assessment of the retinal vasculature.

Using Trizol reagent (Invitrogen, Carlsbad, CA), total retinal RNA was isolated (four eyes from four mice from each group) according to the manufacturer’s instructions. Then RNA was reverse transcribed with reverse transcriptase (Promega, Madison, WI, USA) to generate cDNA, and the relative amounts of cyclin D1 transcript were determined by real-time quantitative PCR (qRT-PCR). The primers used were: 5′-TGC CAT CCA TGC GGA AAA TCG T-3′ and 5′-GCT CCT CGA CGA CGT TTA CCT T-3′ for Cyclin D1, and 5′-ATG GAT GAC GAT ATC GCT GCG C-3′ and 5′-TAC CTA CTG CTA TAG CGA CGC G-3′ for β-actin. The conditions of PCR were 94 °C for 30 min followed by 40 cycles at 95 °C for 20 s, 57 °C for 20 s, and 72 °C for 20 s. Measurements were performed three times independently.

Western blot was performed using the documented standard methods as we have done before [9]. Retinas (four eyes from four mice from each group) were dissected, immersed and homogenized in 0.1% triton X-100 extraction buffer (Solarbio, Beijing, China) containing phenylmethylsulfonyl fluoride (PMSF) and dithiothreitol (DTT), then the lysates were centrifuged to remove insoluble material, after which total retinal protein was extracted. Protein concentration was measured by a protein assay (Bradford Protein Assay; Bio-Rad, Hercules, CA, USA), and adjusted to allow equal total protein loading on the gels. Proteins were transfer to polyvinylidene fluoride (PVDF) membranes (Millipore, Billerica, MA, USA) using standard electroblotting procedures. Then membranes were blocked with 5% skim milk and then incubated with cyclin D1(1:500 in 5% BSA) (AC853, Beyotime Biotechnology, Jiangsu, China), or β-actin (1:5000 in 5% BSA) (Santa Cruz Biotechnology, Santa Cruz, CA, USA) primary antibodies at 4 °C overnight. Then membranes were incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies (ZSGB-BIO, Beijing, China) at 37 °C for 1 h, and visualized by the ChemiDoc™ MP System (Bio-Rad, Hercules, CA, USA). Protein bands were quantified using Image-Pro plus 6.0 analysis software.

All data are presented as the means ± SD. Data were analyzed by the independent sample student’s t-test, or one-way ANOVA followed by LSD test. Differences were considered statistically significant when p < 0.05.

RNV was examined by FITC-dextran in retinal flat mounts at P17. Control group retinas had both superficial and deep vascular layers extended from optic nerve to periphery. Vessels in superficial retinal layers formed a fine radial branching pattern and in deeper retinal layers formed a polygonal reticular pattern (Fig. 1a). OIR group retinas were characterized by non-perfusion areas and neovascularization. Compared to OIR group, the ratio of non-perfused area to total retinal area was lower in RAPA group (0.018 ± 0.007 VS 0.059 ± 0.08; t = 8.780, p < 0.001), which demonstrates a strong inhibitory effect of RAPA on RNV in OIR model. (Fig. 1).

H&E stained retina sections are shown in Fig. 2, and the degree of RNV was quantified in serial 5-μm cross sections by counting the number of vascular cell nuclei on vitreous side of ILM. Large clusters of blood vessels were observed in OIR group, while fewer blood vessels were observed in RAPA group retinas. The average number of nuclei in CON, OIR and RAPA groups was 0.67 ± 0.52, 12.83 ± 2.64 and 4.83 ± 1.33, respectively (F = 76.463, p < 0.001) (Fig. 2).

Cyclin D1 is the most important positive regulatory factor of the cell cycle, and the effect of RAPA on the expression levels of both Cyclin D1 mRNA and protein were determined in the retina of OIR mice. The Cyclin D1 mRNA levels in the retina of OIR group was 289.8% ± 48.5% of CON, and RAPA treatment significantly reduced Cyclin D1 mRNA level (147.1% ± 33.1% of the control) (F = 22.641, p < 0.001). The protein expression level of Cyclin D1 demonstrated the same trend. Cyclin D1 protein level significantly increased in OIR group, and RAPA prevented the change, at least in part (246.3% ± 47.4 and 157.1% ± 33.4% of CON, respectively) (Fig. 3).

Mice in the three groups all got weight (P17) and there was no significant difference in mean body weight among three groups, indicating that no obvious systemic side effect was associated with RAPA treatment (Fig. 4).