spacer gif spacer gif spacer gif spacer gif spacer gif
 QUICK SEARCH:   [advanced]


spacer gif
     Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents    

First published online 19 May 2004
doi: 10.1242/dev.01156


Development 131, 2853-2864 (2004)
Published by The Company of Biologists 2004


This Article
Right arrow Summary Freely available
Right arrow Full Text
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fürthauer, M.
Right arrow Articles by Thisse, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fürthauer, M.
Right arrow Articles by Thisse, B.

Fgf signalling controls the dorsoventral patterning of the zebrafish embryo

Maximilian Fürthauer, Jeanne Van Celst, Christine Thisse and Bernard Thisse*

Institut de Génétique et de Biologie Moléculaire et Cellulaire, UMR 7104, CNRS/INSERM/ULP, 1 rue Laurent Fries, BP10142, CU de Strasbourg, 67404, Illkirch Cedex, France



View larger version (80K):

[in a new window]
 
Fig. 1. Dorsal clearance of Bmp gene transcripts and concomitant formation of a dorsoventral Fgf gradient. (A-C) Expression of bmp2b at sphere stage (A), 30% epiboly (B) and at shield (C) stages. (D-F) Expression of bmp7 at sphere (D), 30% epiboly (E) and at shield (F) stages. (G) Expression of bmp2b at 30% epiboly in embryo injected with a chd morpholino (mo-chd). (H) Expression of bmp2b at 30% epiboly in an embryo injected with chd RNA. (I) Expression of dharma (dha) in wild-type embryo at 30% epiboly (inset provides a lateral view of the same embryo). Expression of dha is restricted to the yolk syncytial layer (y). Expression of (J) fgf8 at sphere stage, (K) fgf3 at 30% epiboly and (L) fgf24 at 30% epiboly. (M-O) Expression of spry4 at sphere (M), 30% epiboly (N) and late blastula (O) stages. Embryos are oriented dorsal towards the right. A-I,O are animal pole views; J-N are lateral views.

 


View larger version (99K):

[in a new window]
 
Fig. 2. Fgf signalling dorsalises the embryo by inhibiting Bmp gene expression. (A-D) Inhibition of Bmp signalling or Fgf overexpression cause dorsalisation phenotypes. Early segmentation stage morphology of wild-type (A), nog1-injected (B), fgf3-injected (C) and fgf24-injected (D) embryos. (E-N) Blastula stage Bmp gene expression is unaffected by inactivation of Bmp proteins, but lost following Fgf overexpression. All embryos are at 30% epiboly. Expression of bmp2b in wild-type (E), nog1-injected (F) and fgf8-injected (G,H) embryos (H is a lateral view of G, dorsal toward the right). (I,J) bmp7 expression in wild-type (I) and fgf8-injected embryos (J). (K,L) bmp4 expression in wild-type (K) and fgf8-injected embryos (L). (M,N) Loss of bmp2b expression following injection of fgf3 (M) or fgf24 (N). (O,P) fgf8 overexpression dorsalises chd-depleted embryos. Characteristic dorsalised morphology (O) and loss of bmp2b expression at blastula stage (P) following co-injection of mo-chd and fgf8 RNA. (Q,S) Ectopic expression of spry4 expression after localised misexpression of RNA encoding Erm1 (Q) or Pea3 (S). (R,T) Local inhibition of bmp2b expression (asterisk) after localised misexpression of RNA encoding Erm1 (R) or Pea3 (T). (A-D,O) Dorsal views, anterior towards the left; (E-G,I-N,P-T) Animal pole views, dorsal towards the right.

 


View larger version (107K):

[in a new window]
 
Fig. 4. Spry2 is a feedback inhibitor of Fgf signalling. (A) Wild-type expression of spry2. (B) Inhibition of Fgf signalling by injection of RNA encoding dn-fgfr abolishes spry2 expression. (C) Ubiquitous or (D) localised (arrowhead) overexpression of fgf8 RNA induces ectopic spry2 expression. (F) spry2 expression is reduced in fgf8/ace mutant embryos compared with wild-type siblings (E). Compared with wild-type (G) embryos injected with fgf8 RNA (H) display dorsalised phenotypes. (I) Co-injection of fgf8 with spry2 RNA rescues embryos. (J,O,L) Injection of (J) mo-spry2 or (O) dn-spry2, or (L) co-injection of both induce dorsalised phenotypes. Compare with injection of fgf8 RNA (H,P). (K) Dorsalisation is abolished by co-injection of mut-spry2. (A-D) Late blastula stage, (E-P) early segmentation stage. (A-C) Animal pole views; (D-L) lateral views; (M-P) dorsal views. (E,F,M-P) Anterior is towards the left.

 


View larger version (113K):

[in a new window]
 
Fig. 3. spry2 is a novel member of the fgf8 synexpression group. Comparison of the expression pattern of spry2 (A,B,D,F,H,J,L,N,P,R,T) with fgf8 (C,E,G,I,K,M,Q,S), fgf3 (O), spry4 (U), sef (V), erm1 (W) and pea3 (X). (A,B) spry2 transcripts in 32-cell stage embryos. (C,D) Late blastula stage expression of fgf8 (C) and spry2 (D). (E,F) Midgastrula stage expression of fgf8 (E) and spry2 (F); black arrowhead indicates margin; black arrow indicates presumptive midbrain-hindbrain region; white arrow indicates axial hypoblast; white arrowhead indicates presumptive forebrain. (G-J) Five-somite stage expression of fgf8 (G,I) and spry2 (H,J). (K,L) Expression of fgf8 (K) and spry2 (L) at the 16-somite stage. (M-P) Expression of fgf8 (M), fgf3 (O) and spry2 (N,P) at 36 hours. Arrowheads indicate rhombomere boundaries; arrows indicate anterior otic vesicle. (Q-T) Expression of fgf8 (Q,S) and spry2 (R,T) at 48 hpf in the neurohypophysis (Q), adenohypophysis (R), the apical ectodermal ridge (S) and mesenchyme of the pectoral fin (T). (U,V) Five-somite stage expression of spry4 (U), sef (V), erm1 (W) and pea3 (X). (A,C,D,G,H,K-R, U-X) Lateral views; (B) animal pole view; (I,J,S,T) dorsal views. Anterior is upwards in C-F and towards the left in G-X. ba, branchial arches; dd, dorsal diencephalon; hpm, heart primordia; mhb, midbrain hindbrain boundary; os, optic stalk; r2/r4, rhombomeres 2 and 4; som, somites; tb, tail bud; tel, telencephalon.

 


View larger version (119K):

[in a new window]
 
Fig. 5. Inhibition of spry2 alters dorsoventral patterning. (A-D) Embryos co-injected with mo-spry2 and dn-spry2 display severely reduced bmp2b expression. (E-L) Increased Bmp signalling levels cause alterations of DV patterning in spry2 loss-of-function experiments. (F) Injection of mo-spry2 causes an expansion of cyp26a-expressing dorsal anterior neurectoderm, compared with (E) wild type (wt), which is abolished by co-injection of bmp2b (G) similar to inhibition of neural fate after injection of bmp2b alone (H). (I,J) mo-spry2-injected embryos display a reduction of the foxi1-expressing ventral epidermis. Co-injection of bmp2b rescues and even enlarges the epidermal territory (K) similar to a single bmp2b injection (L). (A-D) Shield stage; (E-L) midgastrula stage. (A-D,E-L, top) Lateral views. (E-L, bottom) Animal pole views; all embryos are dorsal towards the right.

 


View larger version (47K):

[in a new window]
 
Fig. 9. DV patterning of the early zebrafish embryo. (A) Two levels of regulation of the morphogenetic Bmp activity gradient. First, the action of Bmp proteins (dark blue squares) is inhibited by chordin and noggin (red), which bind Bmps and prevent them from interacting with their receptors. Second, Fgfs affect the DV patterning by restricting the domain of Bmp gene expression (light blue). Fgf signals are modulated by feedback-inhibitors such as Spry2. The combined regulation of Bmp gene expression and Bmp protein activity results in the generation of a Bmp activity gradient (dark blue) that determines the identity of cells along the DV axis. Cells that experience high levels of Bmp activity will adopt a ventral (e.g. epidermal, green; ventral, V) fate, cells that experience a low Bmp activity a more dorsal (e.g. anterior neurectodermal, yellow; dorsal, D) fate. (B) Noggin overexpression does not affect early Bmp gene expression but abolishes Bmp activity by complexing all available Bmp molecules. As a result, the ventral epidermis is lost, to the benefit of the dorsal neurectoderm. (C) Fgf overexpression abolishes Bmp gene expression and therefore also Bmp activity. As for Noggin-injected embryos, the ventral epidermis is lost at the expense of the dorsal neurectoderm. (D) Loss of function of the Fgf-signalling antagonist spry2 causes an upregulation of endogenous Fgf signalling and a decrease of the Bmp transcription domain. The dorsal neurectoderm is expanded at the expense of the epidermis. (E) Following inhibition of Fgf signalling, Bmp gene transcripts are expressed throughout the embryo with the exception of the dorsalmost marginal blastomeres. As a consequence, the ventral epidermis expands while the dorsal neurectoderm is severely reduced in size. (F) Overexpression of Noggin in Fgf-depleted embryos inactivates Bmp proteins. Consequently, all cells adopt a dorsal neurectodermal fate.

 


View larger version (111K):

[in a new window]
 
Fig. 6. Fgf-mediated restriction of Bmp gene expression is essential for dorsoventral patterning. bmp2b expression in wild type (A), or following overexpression of spry2 (B), injection of RNA encoding a dn-fgfr (C) or treatment with SU5402 (D). (E,F) Following dn-fgfr injection, the expression of the Bmp target gene ved expands dorsally. (G-J) Inhibition of Fgf signalling reduces or abolishes the expression of dorsal mesodermal markers goosecoid (gsc, G,H) and sonic hedgehog (shh, I,J). (K,L) Conversely, the expression territory of draculin (drl), a marker of the ventral hypoblast, is expanded compared with wild type. (M-P) Compared with wild type (M,O) dn-ras injection causes a severe reduction of the expression domain of the pan-neural marker zic2 (N,P). (Q,R) Two-colour in situ hybridisation with bmp2b (blue) and the anterior neural marker otx2 (green) or (S,T) with ved (blue) and cyp26a (green). Arrowheads (Q-T) indicate the border between the gene expression domains. (A-H) Shield stage, (I-T) 70% epiboly. (A-H) Animal pole views; (I,J,M,N,Q-T) lateral views; (K,L) optical cross-section at the level of the margin; (O,P) dorsal views. (A-N,Q-T) Dorsal towards the right.

 


View larger version (68K):

[in a new window]
 
Fig. 7. Inhibition of Bmp signalling restores the ectodermal DV patterning in Fgf-depleted embryos. Analysis of the DV patterning of the ectoderm with foxi1 (A-D), cyp26a (E-H) and vhnf1 (I-L). Black arrows indicate the space between the blastoderm margin and the ectodermal gene expressions. (B) Microinjection of dn-ras causes an expansion of foxi1 expression, (F) a reduction of the neural expression domain of cyp26a (white arrowhead) and (J) a complete loss of vhnf1. (C) Co-injection of nog1 with dn-ras abolishes epidermal cell fates, (G) expands the anterior neurectoderm but (K) does not rescue vhnf1 expression. (D) nog1 injection abolishes foxi1 expression, while the neural expression domains of cyp26 (H) and vhnf1 (L) expand ventrally. (A-H) 75% epiboly stage; (I-L) 85% epiboly stage. All embryos are shown in lateral view, dorsal towards the right.

 


View larger version (120K):

[in a new window]
 
Fig. 8. Fgf8 and chordin interact genetically. (A-D) Expansion of the ventral mesoderm (arrowheads) in embryos injected with chordin morpholino (mo-chd) (C) is enhanced by co-injection of mo-fgf8 (D) when compared with injection of mo-fgf alone (B) or with wild type (A). (E-H) Injection of mo-fgf8 causes a reduction of head size in chordino (din) mutant embryos. mo-fgf8 injection leads to loss of the cerebellum (arrowheads). (I-L) ace/fgf8 mutation enhances the expansion of the ventral hematopoietic mesoderm induced by mo-chd injection. (I,K) Loss of en3 expression (arrowheads) identifies ace homozygous mutants (K). (J,L) Expression of hemoglobin (hem) (arrows) after injection of mo-chd in heterozygous ace+/– embryo (J) and in homozygous ace–/– mutant (L). (M,N) Injection of mo-fgf8 does not affect cyp26a expression (N) compared with wild type (M). (O,P) The reduction of neural cyp26a expression caused by mo-chd (O) is further enhanced by co-injection of mo-fgf8 (P). (Q,R) Injection of mo-fgf8 does not affect drl expression (R) compared with wild type (Q). (S,T) The dorsal expansion of the expression of ventral mesodermal marker draculin (drl) caused by mo-chd (S) is further enhanced by co-injection of mo-fgf8 (T). Arrowheads in S,T indicate the dorsal limit of the drl expression domain. (A-E,G,I-L) Lateral views, anterior towards the left. (F,H) Frontal views, dorsal towards the top. (M-P) Dorsal views, anterior upwards. (Q-T) Optical sections through the margin, dorsal towards the right. (A-L) 30 hpf, (M-T) 75% epiboly.

 





© The Company of Biologists Ltd 2004