Abstract
Purpose:
Zebrafish are a premier animal model of cone photoreceptor development and physiology, being the only diurnal cone-rich species with tractable genetics. Zebrafish cone function, morphology and development are conserved with humans. One aspect that is distinct from mammals is the zebrafish cone mosaic: an obvious spatial pattern exists amongst the cone spectral subtypes, resulting in rigid neighbor relationships. Here we document the connectivity pattern of the four cone types (ultraviolet-, blue-, green- and red-sensitive cones = UV-, B-, G- and R-cones) at the level of the synapse, knowledge of which impinges upon understanding how/why cone mosaics are formed and how signals are processed towards colour vision.
Methods:
Cones were labelled using a combination of transgenic fish expressing GFP or mCherry in cone subtypes, along with immunohistochemistry to cone subtypes, including use of opsin antibodies whose specificity had not previously been characterized in zebrafish. Antibody specificity was determined using double-labelling to compare with established markers.
Results:
Zebrafish UV- and B-cones are known to have single-cone morphology and have their cell bodies packed evenly into contiguous rows. The R- and G-cones are fused along their apical-basal axis into a double-cone pair. Pedicles of double cones are confirmed here to be fused. UV- and B-cone synaptic pedicles have an unexpected relationship that does not match their contiguous spacing observed at the level of the cell body. Instead, UV- and B- cone pedicles are adjoined in distinguishable pairs, akin to the fusion of R- and G-cones into double cone pairs. Telodendria are observed to form connections beyond these UV-/B-cone pedicle pairs.
Conclusions:
The function of both double cone photoreceptors and cone mosaics remains debatable, as do the mechanisms underpinning their development. Speculatively, the fusion of UV- and B-cone pedicles identified here adds weight to the hypothesis that cone packing is adaptive for signal processing, perhaps increasing processivity speed during comparison of signals from cone spectral-subtypes prior to (or at the synapse onto) second order neurons. The data is also interpreted in light of the hypothesis that cone mosaics are adapted to detect moving stimuli. Further, this work serves as encouraging baseline data for mapping connectivity of the regenerated cone photoreceptors that we observe following cone subtype-specific ablation.
Keywords: 698 retinal development •
648 photoreceptors •
689 retina: distal (photoreceptors, horizontal cells, bipolar cells)