Bicistronic lentiviruses containing a viral 2A cleavage sequence reliably co-express two proteins and restore vision to an animal model of LCA1.

The disease processes underlying inherited retinal disease are complex and are not completely understood. Many of the corrective gene therapies designed to treat diseases linked to mutations in genes specifically expressed in photoreceptor cells restore function to these cells but fail to stop progression of the disease. There is growing consensus that effective treatments for these diseases will require delivery of multiple therapeutic proteins that will be selected to treat specific aspects of the disease process. The purpose of this study was to design a lentiviral transgene that reliably expresses all of the proteins it encodes and does so in a consistent manner among infected cells. We show, using both in vitro and in vivo analyses, that bicistronic lentiviral transgenes encoding two fluorescent proteins fused to a viral 2A-like cleavage peptide meet these expression criteria. To determine if this transgene design is suitable for therapeutic applications, we replaced one of the fluorescent protein genes with the gene encoding guanylate cyclase-1 (GC1) and delivered lentivirus carrying this transgene to the retinas of the GUCY1*B avian model of Leber congenital amaurosis-1 (LCA1). GUCY1*B chickens carry a null mutation in the GC1 gene that disrupts photoreceptor function and causes blindness at hatching, a phenotype that closely matches that observed in humans with LCA1. We found that treatment of these animals with the 2A lentivector encoding GC1 restored vision to these animals as evidenced by the presence of optokinetic reflexes. We conclude that 2A-like peptides, with proper optimization, can be successfully incorporated into therapeutic vectors designed to deliver multiple proteins to neural retinal. These results highlight the potential of this vector design to serve as a platform for the development of combination therapies designed to enhance or prolong the benefits of corrective gene therapies.

Expression characteristics of dual-promoter lentiviral vectors targeting retinal photoreceptors and Müller cells.

PURPOSE: Growing evidence suggests that successful treatment of many inherited photoreceptor diseases will require multi-protein therapies that not only correct the genetic defects linked to these diseases but also slow or halt the related degenerative phenotypes. To be effective, it is likely that therapeutic protein expression will need to be targeted to specific cell types. The purpose of this study was to develop dual-promoter lentiviral vectors that target expression of two proteins to retinal cones and rods, rods only, or Müller cells. METHODS: Dual-promoter lentivectors were constructed using the following promoters: Xenopus opsin promoter (XOPS)1.3, murine opsin promoter (MOPS), interphotoreceptor retinoid binding protein promoter (IRBP156), rhodopsin kinase (RK), neural retina leucine zipper (NRLL), vimentin (VIM), cluster differentiation (CD44), and glial fibrillary acidic protein (GFAP). Vectors were packaged and injected into the neural tubes of chicken embryos. The activities of the promoters alone, in duplicate, or when paired with a different promoter were analyzed in transduced, fully-developed retinas, using direct fluorescent and immunofluorescent microscopy. RESULTS: IRBP156, NRLL, and RK were active in cones and rods while XOPS1.3 was active only in rods. Of the glial promoters, only GFAP activity was restricted to Müller cells; both VIM and CD44 were active in Müller and neural cells. Dual-promoter vectors carrying IRBP156 and RK or XOPS1.3 and MOPS, in the order listed, exhibited robust expression of both reporter transgenes in cones and rods or rods only, respectively. Expression of the upstream transgene was much lower than the downstream transgene in dual-promoter vectors constructed using two copies of either RK or IRBP156. Analyses of the expression of a dual-promoter vector carrying CD44 and VIM in the order listed showed that the activity of the VIM promoter was more restricted to glial cells when paired with the CD44 promoter, while the activity of the CD44 promoter was inhibited to the extent that no CD44-driven reporter protein was detected in transduced cells. CONCLUSIONS: We have identified two dual-promoter vectors, one that targets cones and rods and one that targets rods alone. Both vectors reliably express the two proteins encoded by the transgenes they carry. When two well matched promoters are not available, we found that it is possible to target expression of two proteins to single cells using dual-promoter vectors carrying two copies of the same promoter. These vectors should be useful in studies of retina when co-delivery of a reporter protein with an experimental protein is desired or when expression of two exogenous proteins in targeted cells is required.