As more genes were identified to cause IRDs a
As more genes were identified to cause IRDs, a relatively large proportion were found to either cause multiple phenotypes or multiple inheritance patterns. Out of the 112 autosomal genes that are listed in RETNET (as of July 22, 2016) that are known to cause non-syndromic IRDs (RP, LCA, and CRD), 16% were reported to cause disease in both R547 and recessive inheritance patterns and 46% were reported to cause multiple phenotypes. Interestingly, 16 of the genes (14.3%) cause multiple diseases in multiple inheritance patterns, usually with no clear genotype-phenotype correlation and no available functional assay to predict the mutation effect on protein function. The GUCY2D gene, encoding for retinal guanylate cyclase, is a unique member of this group. Different GUCY2D mutations cause either arLCA1 or adCRD and characterization of patients with GUCY2D mutations has been reported. In addition, functional expression and biochemical analyses of representative mutations from each group allows us to establish genotype-phenotype correlation in vitro and in mouse models. In this review, we will describe and summarize the large number of reported GUCY2D mutations, the functional analysis of some of these mutations, the established genotype-phenotype correlation, and the development of animal models and gene augmentation therapy for GUCY2D. A few previous excellent reviews on the photoreceptor guanylate cyclase were recently published (Boye, 2016, Koch and Dell'Orco, 2013, Sharma and Duda, 2014, Wen et al., 2014) and the current one aims to integrate recent genetic, functional and clinical studies for a broader understanding of mutation effect on protein function and disease type.
Guanylate cyclases in humans
IRD mutations affecting the GC-E/GCAP complex seem to be restricted to the GUCA1A and the GUCY2D genes (to date only one exception is known). The exception is the p.G157R substitution in GCAP2 (GUCA1B) reported to cause ad retinal degeneration (Sato et al., 2005). Disease-causing mutations in the GUCA1B gene seem to be rare occasions based on screening tests on heterogeneous groups of patients (Kitiratschky et al., 2011, Payne et al., 1999).
Animal models with mutations in retinal guanylate cyclase Seven animal models have been reported so far to harbor mutations in GUCY2D as summarized in Table 5. This topic has been previously reviewed (Boye, 2016) and will be only briefly mentioned here.
It is a very exciting time in the field of inherited retinal disease research. This includes advances in treatment modalities that are not specific to the causative gene (i.e. stem cell therapy and retinal implants) which will not be covered in this review. The area with perhaps the greatest momentum is gene-based therapy (Lipinski et al., 2013). Most of these approaches are based on subretinal injection of adeno-associated virus (AAV) that is packed with the gene of interest and are currently being studied in animal models. A few (including CHM, CNGA3, CNGB3, RPE65, and XLRS) have been tested in human subjects and some of which show both promising results as well as some limitations that have been reviewed elsewhere (Barnard et al., 2014, Dalkara and Sahel, 2014, Pierce and Bennett, 2015, Trapani et al., 2014). Two proof of concept studies addressing GUCY2D gene therapy in models of LCA1 were reported in 2006. The first was performed on the GUCY1*B chicken model (section 4.1.2), which is homozygous for a null mutation, is blind at hatching and serves as a model for LCA1. An in ovo treatment strategy was used by injecting a lentivirus vector carrying the bovine Gucy2d gene into 7 embryos, prior to the appearance of retinal pathology. Six of the injected animals showed improved visual function, ERG responses, and a slower pace of retinal degeneration compared to wt animals (Williams et al., 2006). The authors noted that although gene transfer was at a very early stage and was successful at delivering the gene to the retina, it did not prevent the process of retinal degeneration, but rather a slower degeneration process was evident and ultimately, the rescue effects were transient.