We sought to characterize the regenerated cells, if any, when photoreceptor ablation was mostly limited to a particular cone subtype. This allowed us to uniquely assess whether the remaining cells influence specification of regenerating photoreceptors. The ability to replace lost photoreceptors via stem cell therapy holds promise for treating many retinal degenerative diseases. Zebrafish are potent for modelling this because they have robust regenerative capacity emanating from endogenous stem cells, and abundant cone photoreceptors including multiple spectral subtypes similar to human fovea. We ablated the homolog of the human S-cones, the ultraviolet-sensitive (UV) cones, and tested the hypothesis that the photoreceptors regenerating in their place take on identities matching those expected from normal cone mosaic development. We created transgenic fish wherein UV cones can be ablated by addition of a prodrug. Thus photoreceptors developed normally and only the UV cones expressed nitroreductase; the latter converts the prodrug metronidazole to a cell-autonomous neurotoxin. A significant increase in proliferation of progenitor cell populations (p<0.01) was observed when cell ablation was primarily limited to UV cones. In control fish, we found that BrdU primarily incorporated into rod photoreceptors, as expected. However the majority of regenerating photoreceptors became cones when retinal cell ablation was predominantly restricted to UV cones: a 2-fold increase in the relative abundance of cones (p = 0.008) was mirrored by a 35% decrease in rods. By primarily ablating only a single photoreceptor type, we show that the subsequent regeneration is biased towards restoring the cognate photoreceptor type. We discuss the hypothesis that, after cone death, the microenvironment formed by the remaining retinal cells may be influential in determining the identity of regenerating photoreceptors, though other interpretations are plausible. Our novel animal model provides control of ablation that will assist in identifying mechanisms required to replace cone photoreceptors clinically to restore daytime vision.

Retinal dystrophies are predominantly caused by mutations affecting the visual phototransduction system and cilia, with few genes identified that function to maintain photoreceptor survival. We reasoned that growth factors involved with early embryonic retinal development would represent excellent candidates for such diseases. Here we show that mutations in the transforming growth factor-β (TGF-β) ligand Growth Differentiation Factor 6, which specifies the dorso-ventral retinal axis, contribute to Leber congenital amaurosis. Furthermore, deficiency of gdf6 results in photoreceptor degeneration, so demonstrating a connection between Gdf6 signaling and photoreceptor survival. In addition, in both murine and zebrafish mutant models, we observe retinal apoptosis, a characteristic feature of human retinal dystrophies. Treatment of gdf6-deficient zebrafish embryos with a novel aminopropyl carbazole, P7C3, rescued the retinal apoptosis without evidence of toxicity. These findings implicate for the first time perturbed TGF-β signaling in the genesis of retinal dystrophies, support the study of related morphogenetic genes for comparable roles in retinal disease and may offer additional therapeutic opportunities for genetically heterogeneous disorders presently only treatable with gene therapy.

No abstract

Hurdles in the treatment of retinal degeneration include managing the functional rewiring of surviving photoreceptors and integration of any newly added cells into the remaining second-order retinal neurons. Zebrafish are the premier genetic model for such questions, and we present two new transgenic lines allowing us to contrast vision loss and recovery following conditional ablation of specific cone types: UV or blue cones. The ablation of each cone type proved to be thorough (killing 80% of cells in each intended cone class), specific, and cell-autonomous. We assessed the loss and recovery of vision in larvae via the optomotor behavioural response (OMR). This visually mediated behaviour decreased to about 5% or 20% of control levels following ablation of UV or blue cones, respectively (P<0.05). We further assessed ocular photoreception by measuring the effects of UV light on body pigmentation, and observed that photoreceptor deficits and recovery occurred (p<0.01) with a timeline coincident to the OMR results. This corroborated and extended previous conclusions that UV cones are required photoreceptors for modulating body pigmentation, addressing assumptions that were unavoidable in previous experiments. Functional vision recovery following UV cone ablation was robust, as measured by both assays, returning to control levels within four days. In contrast, robust functional recovery following blue cone ablation was unexpectedly rapid, returning to normal levels within 24 hours after ablation. Ablation of cones led to increased proliferation in the retina, though the rapid recovery of vision following blue cone ablation was demonstrated to not be mediated by blue cone regeneration. Thus rapid visual recovery occurs following ablation of some, but not all, cone subtypes, suggesting an opportunity to contrast and dissect the sources and mechanisms of outer retinal recovery during cone photoreceptor death and regeneration.

Functional vision restoration is within reach via stem cell therapy, but one of the largest obstacles is the derivation of colour-sensitive cone photoreceptors that are required for high-acuity daytime vision. To enhance progress made using nocturnal murine models, we instead utilize cone-rich zebrafish and herein investigate relationships between gdf6a and tbx2b in cone photoreceptor development. Growth/differentiation factor 6a (gdf6a), a bone morphogenetic protein family ligand, is an emerging factor in photoreceptor degenerative diseases. The T-box transcription factor tbx2b is required to specify UV cone photoreceptor fate instead of rod photoreceptor fate. Interactions between these factors in cone development would be unanticipated, considering the discrete phenotypes in their respective mutants. However, gdf6a positively modulates the abundance of tbx2b transcript during early eye morphogenesis, and we extended this conclusion to later stages of retinal development comprising the times when photoreceptors differentiate. Despite this, gdf6a−/− larvae possess a normal relative number of UV cones and instead present with a low abundance of blue cone photoreceptors, approximately half that of siblings (p<0.001), supporting a differential role for gdf6a amongst the spectral subtypes of cone photoreceptors. Further, gdf6a−/− larvae from breeding of compound heterozygous gdf6a+/−;tbx2b+/− mutants exhibit the recessive lots-of-rods phenotype (which also shows a paucity of UV cones) at significantly elevated rates (44% or 48% for each of two tbx2b alleles, χ2 p≤0.007 for each compared to expected Mendelian 25%). Thus the gdf6a−/− background sensitizes fish such that the recessive lots-of-rods phenotype can appear in heterozygous tbx2b+/− fish. Overall, this work establishes a novel link between tbx2b and gdf6a in determining photoreceptor fates, defining the nexus of an intricate pathway influencing the abundance of cone spectral subtypes and specifying rod vs. cone photoreceptors. Understanding this interaction is a necessary step in the refinement of stem cell-based restoration of daytime vision in humans.