Abstract
Purpose :
The ability to spatially, temporally and reversibly regulate protein expression in vivo is a central goal for many gene therapy approaches, especially for therapeutic strategies involving proteins that may be detrimental to physiology if constitutively expressed. Accordingly, we develop retinal gene therapy approaches that use a destabilized E. coli dihydrofolate reductase (DHFR) to control the abundance, and therefore activity, of therapeutic proteins. This ‘destabilized domain’ (DD), along with any protein fused to it, undergoes rapid proteasome-mediated degradation in the absence of its stabilizer, the antibiotic trimethoprim (TMP). However, presence of TMP stabilizes the DHFR fusion protein in a rapid, reversible and dose-dependent manner. Herein we explored the ability of DDs to control protein abundance in mouse retina.
Methods :
We generated an AAV expression construct containing DHFR-YFP which concomitantly expressed mCherry. AAV 7m8 virus encoding for this construct or a PBS control was intravitreally delivered to C57BL6/J mice. Three weeks after injection, mCherry expression, which indicates the AAV-infected area, and DHFR-YFP expression were measured by fundus fluorescence imaging. After initial imaging, TMP was administered in the drinking water for select periods of time. YFP and mCherry expression were then determined by fundus imaging, flat mounts and immunohistochemistry. The effect of long-term TMP treatment on visual acuity or retinal structure was evaluated by optokinetic reflex (OKR), electroretinogram (ERG), and histology.
Results :
mCherry fluorescence demonstrated successful transduction of the entire neural retina. In the absence of TMP, DHFR-YFP was efficiently degraded and not detectable by fundus fluorescence or immunohistochemistry. After oral TMP treatment, DHFR-YFP was stabilized and detectable by 6 h. This stabilization is repeatedly reversible following removal or addition of TMP. Alternatively, constantly administered TMP can stabilize DHFR-YFP for long periods (>3 mo). OKR, ERG and histology show long-term TMP treatment has no detrimental impact on retina function/structure.
Conclusions :
This DD-based conditional system is a rapid, efficient and reversible tool to control protein expression in retina. These data provide the foundation for using DDs to control retinal gene therapies that require transient, cyclical or controlled gene expression.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.