Flik, a chick follistatin-related gene, functions in gastrular dorsalisation/neural induction and in subsequent maintenance of midline Sonic hedgehog signalling.
We have targetted the chick gene Flik with antisense oligodeoxynucleotide treatment at gastrular stages, when it is expressed in organiser-derived structures of the midline (K. Patel et al., 1996, Dev. Biol. 178, 327-342). A specific syndrome of deficient axial patterning and holoprosencephaly is produced. Most aspects of this syndrome can be understood as due to attenuation of dorsalising and neural-inducing signals during gastrulation, followed by failure to maintain the later signals from chordamesoderm/neural midline that pattern the mesodermal and neural cross sections during subsequent stages. Anatomical effects are first apparent at early neurula stages and correspond with what might be expected from a reduced counteraction of the ventralising Bone morphogenetic protein (BMP) pathway at the earlier stages, coupled with inadequate Sonic hedgehog (Shh) signalling subsequently. Delay in the clearing of BMP-4 RNA expression from the presumptive neural region at gastrulation is indeed seen, though chordin RNA expression within organiser derivatives remains normal. Subsequently, specific attenuation of chordamesoderm and neural midline Shh expression is observed. Brief preincubation of stage 4 chick blastoderms in supernatant from Xenopus oocytes that have been injected with Flik RNA prolongs and enhances the competence of their peripheral epiblast to respond to neural inductive signals from grafted Hensen's nodes. This effect specifically mimics that recently observed using microg/ml solutions of recombinant Follistatin (D. J. Connolly et al., 1999, Int. J. Dev. Biol., in press), further suggesting that Flik protein might act in vivo by somehow modulating activity of signalling pathways through BMP or other TGFbeta-related ligands. We discuss the significance of the observations in relation to recent ideas about neural induction, about possible redundancy in gene action, and about subsequent patterning of the axial cross section, suggesting that a Flik function in autocrine/paracrine maintenance of later midline Shh signalling represents a role of the gene separate from that in primary dorsalisation/neural induction.