Optimising Gene Therapy for X-linked Retinitis Pigmentosa

Abstract

Purpose Mutations in RPGRORF15 cause 70 to 90 % of the monogenetic disease X-linked retinitis pigmentosa (XLRP), making this gene a high-yield target for causal treatment with gene therapy. Due to the purine-rich, repetitive nature of the terminal ORF15 exon, maintaining transgene sequence fidelity has proven to be a road-block in translational efforts. This thesis contributes to the optimisation of a gene therapy for RPGR-XLRP in two ways: firstly, it aims to investigate codon optimization and use of mutant AAV capsids as a means to overcome the inherent instability of RPGRORF15 and increase transgene expression. Secondly, analysis of pre-treatment characteristics in a cohort of 50 RPGR-XLRP patients will assist both future prospective observational and interventional trials by determining symmetry of disease, rate of progression and suitability of outcome measures as endpoints for clinical trials.

Methods In the first part of the thesis, Western Blot was used to quantify transgene expression in HEK293T cells transfected with codon optimised (co) or wild type (wt) RPGR plasmids as well as to detect transgene expression in mice unilaterally injected with AAV2/8.coRPGR. Immunolabeling was used to show correct localisation of codon optimised transgene to the photoreceptor cilium and to compare transduction efficiency between wild type and single mutant AAV8Y733F capsids. In the thesis’ second part, a retrospective, cross-sectional analysis of 50 patients extracted visual acuity, visual fields (I4e and III4e targets), foveal thickness and ERG data points (ISCEV standard protocol) alongside molecular genetic data. Symmetry and progression were assessed using linear regression and cross-sectional analysis, respectively. Kaplan-Meyer Curves were used to estimate cumulative ‘survival’ of three important levels of visual function (full vision, reading ability, threshold to legal blindness) with age.

Results HEK293T cells transfected with p.coRPGR showed an increase in protein expression (p < 0.005) and demonstrated a superior transgene stability compared to the wild type control. Three different mouse lines, C57BL/6J, C57BL/6J Rd9/Boc and Rpgr-/y, treated with AAV2/8.coRPGR showed a reliable, albeit variable transgene expression and demonstrated co-localisation with RPGR interacting protein (RPGRIP) in the connecting cilium. Mutant capsid (AAV8Y733F) failed to show a significant increase in transduction of 661W cone-like photoreceptor cells (p = 0.058). In the retrospective analysis of clinical data from XLRP patients, 73 % of exonic mutations occurred in ORF15. Yet no clear genotype-phenotype relationship could be established between mutations located in these two parts of the RPGR gene and patients with ORF15 mutations did not have a significantly different visual acuity (p = 0.9) or visual field (III4e; p = 0.6) than those with mutations in exons 1-14. Comparison of both eyes revealed a strong symmetry of degeneration in all outcome measures, with visual fields (I4e ρ = 0.99; III4e ρ = 0.96) and ERG (30 Hz flicker ρ = 0.95) exhibiting the highest symmetry. Disease progression eluded description by a simple function. Kaplan-Meier curve (KMC) analysis predicts the most severe decline in vision between the third and fourth decade of life.

Conclusions Codon optimisation of RPGR significantly increased transgene levels in HEK293T cells compared to a wild type RPGR expression cassette. AAV2/8.coRPGR injected mouse eyes reliably expressed RPGR protein that correctly localised to the photoreceptor connecting cilium in mouse models of RPGR-XLRP. High symmetry in all outcome measures confirm that the contralateral eye can be used as an internal control in an RPGR-XLRP gene therapy trial. The variability between patients makes an intra-individual control preferable to an inter-individual control. According to these findings, the most sensitive parameter to measure disease progression and treatment success in an interventional RPGR-XLRP trial seems to be kinetic visual field using the III4e target. Overall, these two pillars of research contribute to the foundation enabling translation of RPGRORF15 gene therapy into a clinical trial.