During animal development, the separation of cells of different identity is crucial for the formation and function of tissues and organs. Tissues are compartmentalised to ensure organisation remains during dynamic processes, such as cell division and cell rearrangement. At Drosophila compartment boundaries, supracellular actomyosin cables mechanically prevent the mixing of cells of different identity between compartments. We know some of the signalling pathways and transcription factors that control actomyosin enrichment at these boundaries. However, the effectors downstream of these signals, such as cell surface receptors, have only begun to be identified. Further, there is generally a poor understanding of how cell identity information is read to promote actomyosin enrichments. Beyond cell segregation, the role of mechanical boundaries during tissue morphogenesis is unclear.

In this thesis, I studied the compartmental boundaries, called parasegment boundaries (PSBs), responsible for cell sorting in the early Drosophila embryo. I focused on the role of the PSBs during axis extension at gastrulation, when the boundaries first become mechanically active. With the aim of identifying new cell surface molecules required to recruit actomyosin at PSBs, I undertook an in silico screen using publicly available datasets. I screened numerous resources to identify genes that: i) encode cell surface proteins, ii) are expressed in D-V oriented stripes in the early embryo and iii) are under pair-rule gene network regulation. I then undertook an in situ hybridisation screen to ask which of the candidate genes were expressed in stripes that border PSBs. For this work I took advantage of in situ Hybridisation Chain Reaction (HCR) methods, which allowed me to locate mRNA expression with cellular resolution.

I found several genes matching the above criteria and chose to focus on a gene that abuts both odd and even-numbered PSBs throughout germ-band extension, tartan. tartan encodes a transmembrane protein that is regulated by the pair-rule gene network. To assess the role of Tartan, I developed a novel method to locate PSBs in live Drosophila embryos based on the MS2-MCP system to image nascent mRNA transcripts under the control of an engrailed enhancer region. By combining this method with cell tracking, I demonstrated that in tartan null mutant embryos, both odd and even-numbered PSBs lose their straightness during axis extension. This indicates that Tartan is required for recruiting actomyosin to PSBs during axis extension. I further showed that cell intercalation is reduced in tartan null mutant embryos, indicating that functional compartment boundaries is important for morphogenesis.

The pair-rule gene, fushi-tarazu, genetically regulates tartan expression. Therefore, I also tested, using the same methods, whether loss of ftz impacts compartment boundary straightness. I found that ftz is required to straighten even-numbered, but not odd-numbered, PSBs during axis extension. This reveals that the ftz mutant does not fully recapitulate the boundary phenotype of tartan mutants. I discuss several hypotheses to explain this unexpected finding.

In conclusion, I have developed methods that can monitor boundary activity in live embryos, both at the cell and tissue scale. My findings contribute to our understanding of the cell surface code that recruits actomyosin to compartment boundaries during Drosophila embryogenesis. Identifying cell surface factors functioning to straighten PSBs helps us better understand the processes that maintain distinct cell populations on either side of compartment boundaries during embryonic development.