Self-complementary AAV5 vector facilitates quicker transgene expression in photoreceptor and retinal pigment epithelial cells of normal mouse

Abstract

To clarify whether transduction efficiency and cell type specificity of self-complementary (sc) AAV5 vectors are similar to those of standard, single-stranded AAV5 vectors in normal retina, one micro liter of scAAV5-smCBA-GFP vector (1 x 10(12) genome-containing particles/ml) and AAV5-smCBA-GFP vector (1 x 10(12) genome-containing particles/ml) were subretinally or intravitreally (in both cases through the cornea) injected into the right and left eyes of adult C57BL/6J mice, respectively. On post-injection day (PID) 1, 2, 5, 7, 10, 14, 21, 28 and 35, eyes were enucleated; retinal pigment epithelium (RPE) wholemounts, neuroretinal wholemounts and eyecup sections were prepared to evaluate green fluorescent protein (GFP) expression by fluorescent microscopy. GFP expression following trans-cornea subretinal injection of scAAV5-smCBA-GFP vector was first detected in RPE wholemounts around PID 1 and in neuroretinal wholemounts between PID 2 and 5; GFP expression peaked and stabilized between PID 10-14 in RPE wholemounts and between P14 and P21 in neuroretinal wholemounts with strong, homogeneous green fluorescence covering the entire wholemounts. The frozen sections supported the following findings from the wholemounts: GFP expression appeared first in RPE around PID 1-2 and soon spread to photoreceptors (PR) cells; by PID 7, moderate GFP expression was found mainly in PR and RPE layers; between PID 14 and 21, strong and homogenous GFP expression was observed in RPE and PR cells. GFP expression following subretinal injection of AAV5-smCBA-GFP was first detected in RPE wholemounts around PID 5-7 and in neuroretinal wholemounts around PID 7-10; ssAAV5-mediated GFP expression peaked at PID 21 in RPE wholemounts and around PID 28 in neuroretinal wholemounts; sections from AAV5 treated eyes also supported findings obtained from wholemounts: GFP expression was first detected in RPE and then spread to the PR cells. Peak GFP expression in RPE mediated by scAAV5 was similar to that mediated by AAV5. However, peak GFP expression mediated by scAAV5 in PR cells was stronger than that mediated by AAV5. No GFP fluorescence was detected in any retinal cells (RPE wholemounts, neuroretinal wholemounts and retinal sections) after trans-cornea intravitreal delivery of either scAAV5-GFP or AAV5-GFP. Neither scAAV5 nor AAV5 can transduce retinal cells following trans-cornea intravitreal injection. The scAAV5 vector used in this study directs an earlier onset of transgene expression than the matched AAV5 vector, and has stronger transgene expression in PR cells following subretinal injection. Our data confirm the previous reports that scAAV vectors have an earlier onset than the standard, single strand AAV vectors (Natkunarajah et al., 2008; Yokoi et al., 2007). scAAV5 vectors may be more useful than standard, single-stranded AAV vector when addressing certain RPE and/or PR cell-related models of retinal dystrophy, particularly for mouse models of human retinitis pigmentosa that require rapid and robust transgene expression to prevent early degeneration in PR cells.

Figure 1

Diagram of trans-cornea and trans-sclera subretinal/intravitreal injections. This diagram clearly shows that there is no penetration of the retina (and hence damage) via the standard trans-cornea intravitreal injection.

Figure 2

Time course of GFP expression in RPE wholemounts of C57BL/6J mice after subretinal injection of scAAV5-smCBA-GFP or AAV5-smCBA-GFP vectors. Fluorescent micrographs of the RPE wholemounts from PID 2 to PID 21 showing a panoramic view of GFP expression in RPE cells from the beginning to the peak time of expression. A-D): RPE whole mounts at PID 2, 5, 14 and 21 following subretinal delivery of scAAV5-smCBA-GFP vectors; E-H: RPE wholemounts at PID 2, 5, 14 and 21 following subretinal delivery of AAV5-smCBA-GFP vectors. The PID is labeled as “d” in each picture.

Figure 3

Time course of GFP expression in neuroretinal wholemounts of C57BL/6J mice after subretinal injection of scAAV5-smCBA-GFP or AAV5-smCBA-GFP vectors. Fluorescent micrographs of the neuroretinal wholemounts from PID 2 to PID 28 showing a panoramic view of GFP expression in neuroretinal wholemounts from the beginning to the peak time of expression. A-D: neuroretinal wholemounts at PID 2, 7, 14 and 21 following subretinal delivery of scAAV5-smCBA-GFP vectors; E-H: neuroretinal whole mounts at PID 2, 7, 14 and 28 following subretinal delivery of AAV5-smCBA-GFP vectors. The large dark areas in 3E and 3F represent areas where the neuroretina has no GFP fluorescence.

Figure 4

Time course of GFP expression in eyecup cross-sections of C57BL/6J mice after subretinal injection of scAAV5-smCBA-GFP or AAV5-smCBA-GFP vectors. Fluorescent micrographs of the cross-sections showing GFP expression from PID 7 to PID 28. A-C: Cross-sections from mice at PID 7, 14 and 21 after subretinal injection of scAAV5-smCBA-GFP. D-F: Cross-sections from mice at PID 7, 14 and 28 after subretinal injection of AAV5-smCBA-GFP vectors. RPE: retinal pigmented epithelium; OS: outer segments of PR cells; IS: inner segments of PR cells; ONL: outer nuclear layer; INL: inner nuclear layer; GCL: retinal ganglion cell layer.

Figure 5

GFP expression in wholemounts of C57BL/6J mice after trans-cornea intravitreal injection of scAAV5-smCBA-GFP vectors. Fluorescent micrographs of the whole mounts at PID 35 showed no GFP expression either in RPE (A) or neuroretinal (B) wholemounts.