A p21-GFP zebrafish model of senescence for rapid testing of senolytics in vivo.

Senescence drives the onset and severity of multiple ageing-associated diseases and frailty. As a result, there has been an increased interest in mechanistic studies and in the search for compounds targeting senescent cells, known as senolytics. Mammalian models are commonly used to test senolytics and generate functional and toxicity data at the level of organs and systems, yet this is expensive and time consuming. Zebrafish share high homology in genes associated with human ageing and disease. They can be genetically modified relatively easily. In larvae, most organs develop within 5 days of fertilisation and are transparent, which allows tracking of fluorescent cells in vivo in real time, testing drug off-target toxicity and assessment of cellular and phenotypic changes. Here, we have generated a transgenic zebrafish line that expresses green fluorescent protein (GFP) under the promoter of a key senescence marker, p21. We show an increase in p21:GFP+ cells in larvae following exposure to ionising radiation and with natural ageing. p21:GFP+ cells display other markers of senescence, including senescence-associated ß-galactosidase and IL6. The observed increase in senescent cells following irradiation is associated with a reduction in the thickness of muscle fibres and mobility, two important ageing phenotypes. We also show that quercetin and dasatinib, two senolytics currently in clinical trials, reduce the number of p21:GFP+ cells, in a rapid 5-day assay. This model provides an important tool to study senescence in a living organism, allowing the rapid selection of senolytics before moving to more expensive and time-consuming mammalian systems.

ProNodal acts via FGFR3 to govern duration of Shh expression in the prechordal mesoderm.

The secreted glycoprotein sonic hedgehog (Shh) is expressed in the prechordal mesoderm, where it plays a crucial role in induction and patterning of the ventral forebrain. Currently little is known about how Shh is regulated in prechordal tissue. Here we show that in the embryonic chick, Shh is expressed transiently in prechordal mesoderm, and is governed by unprocessed Nodal. Exposure of prechordal mesoderm microcultures to Nodal-conditioned medium, the Nodal inhibitor CerS, or to an ALK4/5/7 inhibitor reveals that Nodal is required to maintain both Shh and Gsc expression, but whereas Gsc is largely maintained through canonical signalling, Nodal signals through a non-canonical route to maintain Shh. Further, Shh expression can be maintained by a recombinant Nodal cleavage mutant, proNodal, but not by purified mature Nodal. A number of lines of evidence suggest that proNodal acts via FGFR3. ProNodal and FGFR3 co-immunoprecipitate and proNodal increases FGFR3 tyrosine phosphorylation. In microcultures, soluble FGFR3 abolishes Shh without affecting Gsc expression. Further, prechordal mesoderm cells in which Fgfr3 expression is reduced by Fgfr3 siRNA fail to bind to proNodal. Finally, targeted electroporation of Fgfr3 siRNA to prechordal mesoderm in vivo results in premature Shh downregulation without affecting Gsc. We report an inverse correlation between proNodal-FGFR3 signalling and pSmad1/5/8, and show that proNodal-FGFR3 signalling antagonises BMP-mediated pSmad1/5/8 signalling, which is poised to downregulate Shh. Our studies suggest that proNodal/FGFR3 signalling governs Shh duration by repressing canonical BMP signalling, and that local BMPs rapidly silence Shh once endogenous Nodal-FGFR3 signalling is downregulated.