Control of brain patterning by Engrailed paracrine transfer: a new function of the Pbx interaction domain.
Bottom Line: In light of recent reports on the paracrine activity of homeoproteins, including Engrailed, we asked whether Engrailed intercellular transfer was also involved in brain patterning and boundary formation.Both zebrafish Eng2a and Eng2b are competent for intercellular transfer in vivo, but only extracellular endogenous Eng2b, and not Eng2a, participates in DMB positioning.In addition, disruption of the Pbx-interacting motif in Engrailed, known to strongly reduce the gain-of-function phenotype, also downregulates Engrailed transfer, thus revealing an unsuspected participation of the Pbx interaction domain in this pathway.
Affiliation: Université Paris Diderot, Sorbonne Paris Cité, Paris 75205, Cedex 13, France Center for Interdisciplinary Research in Biology (CIRB) - CNRS UMR 7241, INSERM U1050, Labex MemoLife, PSL Research University, Collège de France, Paris F-75005, France.Show MeSH
Mentions: In the course of the ex vivo experiments, we noticed that the two antibodies, 4D9 and 4G11, behaved differently on the two zebrafish Engrailed 2 proteins. Although 4D9 recognised both Eng2a and Eng2b, 4G11 recognised only Eng2a on western blots and by immunohistochemistry (Fig. 4A,B). This observation offered us the opportunity to address the role of the paracrine activity of each of the two endogenous Eng2 proteins in normal brain patterning. Zebrafish embryos were injected into the intercellular space at the blastula stage with either 4G11, 4D9 or control (anti-myc) antibodies. Embryos were fixed at 24 hpf, and the size of the mesencephalon was measured by in situ hybridisation against pax6 or wnt1, which delineate the midbrain (Fig. 4C,D). Upon injection, 4D9 (recognising both Eng2a and Eng2b) reduced the size of the mesencephalon (Fig. 4E,F, in a dose-dependent manner; supplementary material Fig. S4), whereas 4G11 (recognising only Eng2a) had no effect (Fig. 4G,H). To confirm the specificity of action of 4D9, we verified that the effect induced by 4D9 injection was abolished upon pre-incubation with its cognate epitope peptide (Fig. 4E,F). It is worth noting that 4D9 in this assay had no effect on eye size (supplementary material Fig. S5), rhombomere formation or oligodendrocyte migration (supplementary material Fig. S6), corroborating the specificity of its action on anterior brain patterning. Given that 4D9 and 4G11 were both able to rescue the reduced eye phenotype induced by En2 activation (Fig. 2E), the absence of the phenotype in 4G11-treated embryos cannot be simply attributed to its inability to block intercellular transfer. Although both Eng2a and Eng2b have the capacity to transfer between cells, only Eng2b paracrine signalling activity is required for correct brain patterning. This result led us to reconsider the gain-of-function phenotypes induced by the two zebrafish Engrailed proteins and, more precisely, their sensitivity to blocking antibody treatment. Surprisingly, treatment with 4D9, which equally recognises Eng2a and b (Fig. 4A,B), decreased the frequency of eye phenotypes induced by Eng2b but not by Eng2a mRNA injection (Fig. 4I). If Eng2a paracrine action did not account for the eye phenotype, we asked whether the intracrine action of the protein differs from that of Eng2b or En2. The transcriptional activity of the Eng2a tested on the MAP promoter in HeLa cells was increased almost twofold compared with that of the other proteins (Fig. 4J), suggesting a more efficient intracrine action of Eng2a.Fig. 4.
Affiliation: Université Paris Diderot, Sorbonne Paris Cité, Paris 75205, Cedex 13, France Center for Interdisciplinary Research in Biology (CIRB) - CNRS UMR 7241, INSERM U1050, Labex MemoLife, PSL Research University, Collège de France, Paris F-75005, France.