Research Highlights

Interphase cell shape defines the mode, symmetry and outcome of mitosis:

tip stalk
(Fig. 1) Endothelial tip and stalk cells display distinct behaviours and functions: Time-lapse images of mosaic labelled (lyn-mCherry) individual tip or stalk cells and quantification of the dorsal movement of tip or stalk cell nuclei and the top, bottom and centre of migrating cells. Tip cells are highly motile. Stalk cells are less motile and undergo cell elongation (Costa et. al., 2016. Nat. Cell Biol.).

Angiogenesis is coordinated by the collective migration of specialised leading “tip” and trailing “stalk” endothelial cells. Exploiting in vivo live single-cell imaging in the zebrafish model, my lab previously demonstrated that these tip and stalk cells exhibit very distinct motile behaviours and cell shape dynamics (Fig.1). Moreover, using an interdisciplinary approach integrating in silico computational modelling studies, we demonstrated that dividing tip cells undergo asymmetric divisions (Costa et. al., 2016. Nat. Cell Biol.). These asymmetric divisions generate daughters of distinct size and tip-stalk identities, effectively self-generating the tip-stalk hierarchy that drives angiogenesis.

( (Fig. 2) A switch to isomorphic division breaks the symmetry of daughter cell size: (a) Schematic of classical cell division. (b) Still images of a human EC in 2D culture undergoing mitotic rounding in division. (c) Still images of a zebrafish ‘tip’ endothelial cell undergoing isomorphic division to generate daughter of differential size (adapted from Lovegrove et. al., 2025. Science.).

More recently, we demonstrated that a fundamental trigger for these asymmetric divisions is the shape of cells in interphase (Lovegrove et. al., 2025. Science). Most cells are thought to adopt a spherical geometry in division, a process termed mitotic rounding (Fig.2a). Indeed, endothelial cells in 2D culture exhibit classic mitotic rounding (Fig.2b). In contrast, we find that endothelial cells and other cell types in vivo can switch to a novel “isomorphic” mode of division, which uniquely preserves pre-mitotic morphology throughout mitosis (Fig.2c). We further identify that distinct shifts in interphase morphology act as conserved instructive cues triggering this switch to isomorphic division. Moreover, in isomorphic divisions, we find that maintenance of interphase cell morphology throughout mitosis ultimately enables cell shape to act as a geometric code defining mitotic symmetry, identity determinant partitioning, and daughter state (Fig.3). Thus, morphogenetic shape change both sculpts tissue form and generates the cellular heterogeneity driving tissue assembly.

(Fig. 3) Interphase cell shape defines the mode, symmetry and outcome of mitosis: As cells elongate, they switch to a newly defined isomorphic mode of division. In isomorphic divisions, cells uncharacteristically retain pre-mitotic asymmetries in morphology and fate determinant positioning throughout mitosis. Consequently, isomorphic division fundamentally couples dynamic changes in interphase shape to the induction of asymmetric cell divisions (adapted from Lovegrove et. al., 2025. Science.).

mRNA localisation as a key determinant of cell shape remodelling and sensing:

mRNA
(Fig. 4) RAB13 mRNA is targeted to endothelial cell motile processes: Endothelial cell protrusions were fractionated to identify protrusion-enriched mRNAs by RNAseq. Endogenous RAB13 mRNA is localised to motile protrusions versus the cell body-localised control mRNA, GAPDH (Costa et. al., 2019. EMBO J).

The polarised targeting of diverse mRNAs to the front of motile cells is a hallmark of cell migration, yet, precise functional roles for such targeting of mRNAs has long remained unknown. Likewise, their relevance to modulation of tissue dynamics in vivo is unclear. Using cell fractionation to isolate motile cell protrusions and RNAseq we previously defined distinct clusters of mRNAs that have unique spatial distributions in migrating endothelial cells, both in vitro and in vivo (Fig.4). Moreover, using single-molecule analysis, endogenous gene-edited mRNAs and zebrafish in vivo live-cell imaging, my lab previously demonstrated that such compartmentalisation RAB13 mRNA uniquely acts to promote local filopodia extension. Moreover, this local cell shape remodelling ultimately acts as a molecular compass that orients motile cell polarity and spatially direct tissue movement (Costa et. al., 2020. EMBO J – see movie below).

More recently, we’ve demonstrated that another one of these targeted mRNAs directs the cell-size-scaling of mitochondria distribution and function (Bradbury et. al., 2025. bioRxiv). We find that mRNA encoding TRAK2, a key determinant of mitochondria retrograde transport, is targeted to distal sites of cell protrusions in a cell-size-dependent manner. This cell-size-scaled mRNA polarisation in turn scales mitochondria distribution by defining the precise site of TRAK2-MIRO1 retrograde transport complex assembly (Fig.5a). As a result, excision of a 29bp 3’UTR motif that underpins this cell-size-scaling eradicates size-scaling of mitochondria positioning, triggering distal accumulation of mitochondria (Fig.5b) and progressive hypermotility as cells increase size. As such, we find an RNA-driven mechanistic basis for the cell-size-scaling of organelle distribution and function that is critical to homeostatic control of motile cell behaviour (Bradbury et. al., 2025. bioRxiv).

(Fig. 5) TRAK2 mRNA localisation spatially modulates TRAK2-MIRO1 interactions: (a) Proximity ligation assay (PLA) reveals that loss of a 29bp 3’UTR motif that underpins the cell-size-scaling of TRAK2 mRNA localisation disrupts the spatial distribution of TRAK2-MIRO1 interactions. (b) Consequently, in the absence of TRAK2 mRNA localisation, mitochondria accumulate at distal sites in a cell-size-dependent manner (adapted from Bradbury et. al., 2025. bioRxiv.).