Investigating non-canonical vertebral development in the zebrafish model system

• Main Author: Kishida, Marcia Gruppi
• Format: Thesis
• Language: English
• Published: Apollo - University of Cambridge Repository 2018
• Subjects: zebrafish; notochord; vertebra; The Arches; Atlantic salmon
• Online Access: https://dx.doi.org/10.17863/cam.24109; https://www.repository.cam.ac.uk/handle/1810/276830

A segmented vertebral column is one of the major innovations vertebrates. In mice and chicks – amniotes – a subpopulation of the somites, the sclerotome, is sole source of vertebral tissue. It is unclear, however, how applicable this amniote-based ‘canonical’ mechanism is across the vertebrates. In fact, the vast majority and diversity of vertebrates are not amniotes, but are members of ‘fish’ groups where there has been relatively little investigation into vertebral development. Indeed, there is great diversity in vertebra form throughout ‘fish’ groups and fossil evidence suggests that the components of the vertebra, the neural arches and the vertebral bodies, arose separately and that vertebrates have evolved multiple ways of building vertebral bodies. In teleosts fish, the vertebral bodies initially form as mineralised rings within the notochord sheath (chordacentra) and then secondarily, bone is deposited around this (perichordal centra and arches). Notochord cells (chordoblasts) have been implicated in chordacentrum mineralisation and patterning in zebrafish and Atlantic salmon, though the question of how the overtly unsegmented notochord could direct segmental mineralisation still remains. My project first aims to address this dual mechanism in the zebrafish model, by testing whether the chordoblasts can mineralise and pattern the chordacentra. The second aim is to elucidate the role of the sclerotome in teleost vertebral development. To do this, I explored CRISPR knock-in tools to label the sclerotome and used a Gal4 gene trap line to investigate sclerotome ablation. I characterised the chordacentra and chordoblasts in our model system and verified the specificity of a promoter as a chordoblast marker. With this promoter, I established a method to target the chordoblasts for KillerRed-induced phototoxicity. I demonstrated that intact chordoblasts are necessary for chordacentrum formation, but that vertebral arches are unaffected. Fused perichordal centra are still able to form, but the underlying sheath has a very different structure. This supports the ‘duality’ hypothesis that in teleosts the role of the sclerotome in vertebra formation is limited to the arches and perichordal centra, whereas the chordoblasts are responsible for the chordacentra.