QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online 13 August 2003
doi: 10.1242/dev.00683

This Article Summary Full Text Full Text (PDF) An erratum has been published Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Scholpp, S. Articles by Brand, M. Search for Related Content PubMed PubMed Citation Articles by Scholpp, S. Articles by Brand, M.

Engrailed and Fgf8 act synergistically to maintain the boundary between diencephalon and mesencephalon

Department of Genetics, University of Dresden (TU), Pfotenhauer Strasse 108, 01307 Dresden, Germany
Max Planck Institute of Molecular Cell Biology and Genetics (MPI-CBG), Dresden, Germany

View larger version (75K): [in a new window]   Fig. 1. Anteroposterior patterning in the early neural tube. Morphological analysis and whole-mount double in situ hybridization of wild-type embryos with indicated markers and stages. (A-C) The transition between forebrain and midbrain is marked by altered morphology, the posterior pax6.1 gene expression boundary and location of the posterior commissure (PC). (A-C) Morphology of the diencephalic-mesencephalic boundary (DMB): confocal microscopy of a living Bodipy-FL-ceramide C5 stained embryo at 26 hpf (B); (A) scanning plane. The white arrow indicates the position of the biotinylated rhodamine-dextran labeled cell and white circles indicate the position of the PC as a morphological landmark of the DMB. (C) The same embryo after in situ hybridization with a pax6.1-probe and antibody staining against biotin to visualize the position of the labeled cell (red arrow). (D-R) Analysis of expression patterns at the territory of the DMB: expression pattern of pax6.1 and pax2.1. pax6.1 is first detectable at 90% of epiboly in a `salt and pepper' distribution (D; black arrows). A horseshoe-like pattern is visible at 100% (E). (G) Cross-section of the embryo in F visualizes the gap between pax6.1 and pax2.1 at tailbud stage (arrowheads). (H,I) Pattern of gap at 6 ss and 24 hpf. (J-N) Expression pattern of pax6.1 and eng2. From the onset (J) until midsomitogenesis (M) there is no gap between the expression domains of pax6.1 and eng2. At 24 hpf (N), the expression of eng2 is much broader on the MHB compared with pax2.1 (I) and fades out into the midbrain. (O-R) Expression pattern of pax6.1 and fgf8. fgf8 is first expressed in a wide band spreading from the MHB primordium over to the anterior hindbrain (O). At tailbud stage the expression is confined to the MHB territory and the presumptive rhombomere 4 (P). In the early phase of neural patterning, 10 cell wide gap is shown between the expression domains (O,P). At 6 ss the gap increases in size to 20 cells (Q). At 24hpf, the expression of fgf8 is refined to the posterior part of the MHB (R). Arrowheads demarcate the gap between the expression domains of pax6.1 and the indicated marker (D-R). Asterisks mark the pax6.1 free territory (E,J,O). dd, dorsal diencephalon; eye, eye anlage; fb, forebrain; fec, facial ectoderm; DMB, diencephalic-mesencephalic boundary; mb, midbrain; MHB, midbrain-hindbrain boundary; ost, optic stalk; PC, posterior commissures, somites per stage; tb, tailbud stage; tec, tectum opticum. Percentage specifications are percent of epiboly indicated in the figures.

View larger version (86K): [in a new window]   Fig. 2. Mis-expression of eng3 results in an anterior shift of the position of the DMB. Mis-expression of 500 pg eng3 mRNA in one cell at the two blastomere stage (A-D). The broken line indicates the midline; the upper side is the control side (con) and the lower side is the injected side (inj). At 1 ss, pax6.1 expression is strongly diminished in the injected side (black arrowheads). The expression of otx2 is not altered. To localize the experimental side, lacZ mRNA was co-injected and detected with a secondary, FITC-coupled antibody (B). At 26hpf, efna4 is suppressed at the injected side (C). The mis-expression of eng3 results in an altered formation (n=36, out of 84) or in a loss of the posterior commissure (n=23, out of 84; D, white arrowheads). To follow the fate of small cell clones in eng3-mRNA-injected embryos, caged fluorescein was co-injected with lacZ-mRNA in control embryos or together with eng3-mRNA (E-J). For this experiment, a microscope was used that was provided with an nitrogen laser unit (E) and a grid (raster size 0.05 mm). At shield stage, the embryos were orientated dorsal side up with the shield to the left. The fluorescein dye was uncaged with a laser beam (365 nm) at the position of the grid: x-axis, –4; y-axis, –3 (F). A picture of a living embryo at shield stage was superimposed on a dark field picture to visualize the clone, and a grid was overlaid. `sh' marks the position of the shield. At 15 ss, the cell clones are visible in the eye or in the posterior diencephalon visualized by a superimposed pseudo lateral picture of the same living wild-type embryo with a darkfield-fluorescence picture (G). In situ hybridization for pax6.1 expression on the same embryo shows that the clone is located in the pax6.1-positive domain (encircled by red dots, H). Small cell clones that were uncaged in embryos injected with 500 pg eng3-mRNA are located in the pax6.1-negative region, suggesting a transformation of cell identity (I,J). (K-M) Injection of different amounts of eng3-mRNA (300 pg and 600 pg, indicated by the red triangle below) show concentration-dependent reduction of pax6.1 expression and of the size of the eye anlage (arrowheads). After the injection of 600 pg eng3-mRNA, the forebrain expression of pax6.1 is completely abolished and the eye anlage was not detectable (M). DMB, diencephalic-mesencephalic boundary; eye, eye anlage; hb, hindbrain; mb, midbrain; PC, posterior commissure; sh, shield.

View larger version (104K): [in a new window]   Fig. 3. The DMB shifts posteriorly in embryos deficient for eng2/eng3. The forebrain expression domain of pax6.1 expands posteriorly in embryos deficient for eng2/eng3 at 15 ss (A,B). noi mutants are recognized by the absence of pax5 expression (C). The normal AP extent of the forebrain is marked by a white arrow from the anterior tip of the telencephalon to the posterior border of the forebrain; this arrow is of the same length in A,B,D,E. The red arrow indicates the expansion of pax6.1 expression observed in embryos lacking eng2/eng3 or noi function. A comparable phenotype is observed in the noi mutant embryos (C,F). Diencephalic expression of efna4 is similarly expanded in eng2/eng3 morphant embryos (D,E) and noi mutant embryos (F). Proliferation of forebrain cells is unaffected, as visualized by -Phosphohistone 3 antibody and in situ hybridization with pax6.1 probe of wild-type and eng2/eng3 morphant embryos at the indicated stages (G-L). The expanded pax6.1 territory is marked by red arrows. (M,N) In eng2/eng3 morphants Engrailed protein is not detectable by the -Engrailed antibody 4D9. (O,P) The expansion is paralleled by a posterior shift of the posterior commissure (arrowheads) at 32 hours, stained with antibodies against acetylated tubulin. Asterisks mark the epiphysis. DMB, diencephalic-mesencephalic boundary; fb, forebrain; hb, hindbrain.

View larger version (81K): [in a new window]   Fig. 6. Eng2/Eng3 and Fgf8 act synergistically in positioning of the DMB. The positioning of the DMB is dependent on Fgf signalling in early somitogenesis. Embryos were orientated dorsal side upwards and anterior towards the left. Inhibition of Fgf signalling via the pharmaceutical Fgfr inhibitor SU5402 (8 µm) in noi mutant embryos from 90% of epiboly to 15 ss leads to a strong expansion of the anterior pax6.1 domain (A,B). A weaker expansion is visible when the treatment starts at 5 ss and lasts until 15 ss (C). After 10 ss, inhibition of Fgf signalling does not lead to a posterior shift of the forebrain expression of pax6.1 (D). (H,J) Eng2/Eng3- and Fgf-deficient embryos show a stronger expansion of the forebrain expression of pax6.1 than do single deficient embryos (F,G). (E-G) Embryos were orientated laterally and anterior is towards the left. The forebrain expression domain of pax6.1 expands in wild-type embryos (E), noi mutant embryos (F) and in MO-fgf8 morphant embryos (G). In noi mutant embryos, injected with MO-fgf8, the forebrain and hindbrain expression domain fuses (H). A comparable phenotype is observed in noi mutant embryos treated with Fgfr inhibitor SU5402 from 90% of epiboly until 15 ss (I). In ace mutant embryos injected with MO-eng2/eng3 forebrain and hindbrain domain fuses similarly (J). DMB, diencephalic-mesencephalic boundary; fb, forebrain; hb, hindbrain.

View larger version (71K): [in a new window]   Fig. 4. FGF8-bead represses pax6.1 expression in the anterior neural plate. (A) FGF8-soaked unilaterally implanted beads repress pax6.1 expression in the posterior diencephalon of a wild-type embryo (arrowhead). An ectopic induction of eng3 expression was observed around the bead (B). In the noi mutant embryos, pax6.1 is also repressed by FGF8 (C) and eng3 is ectopically activated in the absence of functional Pax2.1 protein. In the noi mutant embryos pax2.2, a pax2.1-independent marker at the MHB, is not induced in the same timeframe (E,F). All pictures show dorsal views with anterior to the left, (B-D) show darkfield pictures of embryos in (A,C,D); stages as indicated; arrowheads indicate the posterior border of the expression domain of pax6.1 in the forebrain, circles indicate the position of the bead. DMB, diencephalic-mesencephalic boundary; eye, eye anlage; fb, forebrain; hb, hindbrain.

View larger version (63K): [in a new window]   Fig. 5. Midbrain cells that cannot respond to Fgf signalling acquire forebrain character. Embryos were injected with 400 pg XFD-mRNA (A), grafts of these donor embryos were transplanted to a wild-type host embryos (B) and fixed at indicated stages prior to staining procedures (C). Arrows indicate transplanted cells. (D,F) Cell clones in the midbrain express the forebrain markers pax6.1 and efna4. Transplanted cells contain rhodamine dextran (E,G), but do not express the midbrain marker Engrailed, as visualized by -Engrailed antibody 4D9 staining (H-J). Control transplantations without XFD mRNA show co-localization of rhodamine dextran and -Engrailed staining (K-M). (J,M) Superimposed pictures of H,K and I,L. Transplanted cells do not express any of the hindbrain markers gbx1, fgf8 or krox20 (N,O). a, anterior; fb, forebrain; hb, hindbrain; p, posterior; r3, rhombomere 3.

View larger version (80K): [in a new window]   Fig. 7. Morphological analysis of the function of fgf8 and eng2/eng3 at the DMB at late stages. In comparison with the control embryo at 26 hpf (A), the expression of eng2 is absent in fgf8-MO morphant embryos (B), in noi mutant embryos (C) and in fgf8-MO-injected noi mutant embryos (D). The expression domain of pax6.1 weakly expands in fgf8-MO embryos (B), strongly in noi mutant embryos (C) and it fuses in the double treated embryos (D). The border of the pax6.1 expression is marked by arrowheads (A-D). The position of the posterior commissure (PC) is altered in eng2/eng3- and fgf8-deficient embryos (E-L). Embryos were orientated dorsal side upwards and anterior towards the top. At 28 hpf the isl1 staining labels the neurons of the epiphysis and interneurons of the PC (E-H). In addition, acetylated tubulin is marking the outgrowing axons (I-L). The position of the neurons of the PC is shifted posteriorly in fgf8-MO-injected embryos (F,J); a stronger expansion and an increased number of neurons were found in the noi mutant embryos (arrowheads; G,K). Furthermore, the axon bundle of the PC is fanned out. The strongest phenotype was observed in embryos deficient for Engrailed and Fgf8 (H,L). The number of interneurons is strongly increased (H), and the PC is not formed at all (L). Single branches project into the territory of the misspecified hindbrain (white arrows). Red arrows indicate the most lateral position of the PC and asterisks indicate the position of the epiphysis. Number of isl1 positive epiphyseal neurons in wild-type, MO-fgf8, noi mutant embryos and noi mutants injected with MO-fgf8 (M). Values are the average of the total number of neurons of 10 embryos per treatment. Error bars show the standard deviation and asterisks indicate significant differences (*P=0.01) when compared with wild-type siblings.

View larger version (28K): [in a new window]   Fig. 8. Function of eng2/eng3 and fgf8 in AP patterning of the anterior neural plate. Pax6.1 (blue) demarcates the forebrain and hindbrain, whereas expression of eng2/eng3 (yellow) and fgf8 (red) can be observed in the midbrain/MHB organizer territory (A). Eng2/Eng3 displays the cell-autonomous signal maintaining midbrain fate, whereas Fgf8 acts non-autonomous in the midbrain territory visualized by red arrows. (B) Loss of Eng2/Eng3 causes a posterior shift of the DMB (blue arrow). Fgf8 expression is unaffected at least until 10 ss, suggesting a partial repression of the forebrain expansion. Loss of both signal Eng2/Eng3 and Fgf8 causes in a loss of midbrain identity and in a fusion of forebrain and hindbrain domain (C). Only a dorsal patch of cells remains free of pax6.1 expression. DMB, diencephalic-mesencephalic boundary.