Figure 1.
Dorsoventral and laterality defects in zygotic and maternal zygotic lin mutant embryos.
(A–D) Wild-type zygotic lin siblings with normal ventral tail fin (A), left-positioned heart tube (B) and normal organ situs with liver on the left, pancreas on the right and left looped gut tube (C). (D) Quantification of heart position (n = 14) and direction of gut looping (n = 40). (E–H) Zygotic lin (Zlin) mutant embryos displayed a mild reduction of the ventral tail fin (n = 100/108) (E). In addition, in almost 30% of Zlin mutant embryos, the heart tube was positioned at the midline (F). Gut laterality was unaffected in Zlin mutant embryos (G). Quantification of heart position (n = 108) and direction of gut looping (n = 100). (I–L) Maternal zygotic lin (MZlin) mutant embryos derived from a cross of a homozygous lin mutant female and male showing the more severe posterior truncation (I) compared to a Zlin mutant embryo (E). In addition, most MZlin mutant embryos displayed a laterality defect in the heart (J), liver (bilateral, K) and in looping of the gut (K). (L) Quantification of heart positioning (n = 151) and direction of gut looping (n = 16).
Figure 2.
Genetic variations found in zebrafish bmpr1aa and bmpr1ab genes.
(A) The lin mutation was mapped to a region on chromosome 13 that includes the bmpr1aa gene. (B) T>A basepair change that was found in all lin mutant embryos results in a Leu to Arg change at position 337 (L337R). (C) Crystal structure of human BMPR1B. The kinase domain from the human BMPR1B with the kinase inhibitor LDN-193189 (ball-and-stick representation) bound to the ATP binding site (pdb entry 3MDY). Leu 312 (corresponding to Leu 337 in fish) is shown in red. Structural elements providing residues to the hydrophobic core surrounding Leu 312 are highlighted in dark blue. (D) Detailed view of the hydrophobic core surrounding Leu 312 (in red). Black labels refer to the structure of human BMPR1A, the corresponding residues in fish are indicated by grey italic labels. Consequences of the L312R mutation are analyzed by replacing the leucine side chain in the structure model with arginine, of which five typical rotamers are shown (yellow to green). All rotamers cause serious clashes with surrounding residue, which are highly conserved in fish. (E) C>A basepair change in the bmpr1ab gene that results in a premature stop codon at position 84 in the extracellular domain of the receptor. (F) A MZbmpr1ab mutant embryo at 2 dpf with no obvious phenotypes in the heart or tail region (magnified). (G) Bmpr1a dose-dependent effect on dorsoventral and left-right patterning. Embryos derived from an incross of bmpr1aa+/−;bmpr1ab+/− double carrier fish was analyzed and quantified for the dorsoventral phenotypes (classified as C1 (mild) to C4 (strong)) and position of the heart (left or midline) if present. n/a, not applicable since no heart tissue was present.
Table 1.
Injection studies.
Figure 3.
Dose-dependent effect of Bmpr1a on the expression of laterality genes.
In situ hybridisation at 18-somites for spaw (in LPM) and no tail (ntl) (in midline) (A–C), lefty1 at 23-somites (heart field and midline) (D–F) and pitx2 at 23-somites (in LPM) (G–I). (A,D,G) Embryos selected for normal ventral tail fin or C1 dorsalization (genotypes: bmpr1aa +/+ or +/−; bmpr1ab +/+ or +/− or −/−). (B,E,H) Embryos selected for C3 dorsalization (genotype bmpr1aa−/−;bmp1ab+/−). (C,F,I) Embryos selected for C4 dorsalization (genotype bmpr1aa−/−;bmpr1ab−/−). All embryos are shown as dorsal views with anterior to the top and left to the left. Number of embryos examined is presented in Table 2.
Table 2.
Expression pattern of spaw, lefty1, and pitx2 in bmpr1aa/ab genotypes.
Figure 4.
Rescue of Bmp-related cardiac laterality defects by Nodal beads.
In situ hybridisation for myl7 to highlight the position of the linear heart tube at 30 hpf. Tg(hsp70l:nog3) embryos with no heat-shock (A,D) or heat-shocked at 16 hpf (B,E). Beads (blue) preincubated with recombinant Nodal protein placed in the right ALPM of non-heat-shocked (D) or heat-shocked (E) Tg(hsp70l:nog3) embryos at 17–18 hpf. Control siblings (G,J) or MZbmpr1aa mutant embryos (H,K). Beads (blue) preincubated with Nodal protein placed in the right ALPM of siblings (J) or MZbmpr1aa mutant embryos (K). Position of the inflow pole of the linear heart tube was determined for embryos without a Nodal bead (C) and for embryos in which a Nodal bead was placed on the right side (F). Embryos are shown as dorsal views with anterior to the top and left to the left.
Figure 5.
Bmp via Bmpr1a regulates lefty1 expression in the midline.
(A) In situ hybridisation for lefty1 at 15-somites on embryos from an incross of bmpr1aa+/−;bmpr1ab+/− double carrier fish. Embryos were analysed for lefty1 expression and classified as robust (blue boxed panel) or reduced (red boxed panel) expression after which the embryos were genotyped. Quantification of the results is shown in the stacked area graph (blue, robust lefty1; red reduced lefty1). (B) In situ hybridisation for lefty1 at 10-somite stage. Embryos shown are Tg(hsp70l:bmp2b) embryos either heat-shocked at 10 hpf to induce bmp2b expression (left panel) or without heat-shock (middle panel) and Tg(hsp70l:nog3) embryos heat-shocked at 10 hpf to inhibit Bmp signalling (right panel). Lateral view of 10-somite stage embryos with dorsal to the right and anterior up.
Figure 6.
Bmp and Nodal induce lefty1 independently.
(A–D) Tg(hsp70l:bmp2b) embryos were left at 28°C (A,C) or heat-shocked at 10 hpf to induce bmp2b expression (B,D). A subset of embryos were incubated in the presence of the Nodal inhibitor SB431542 directly after the heat-shock. Embryos were analysed by in situ hybridisation for lefty1 expression at 15-somites. (E–H) Tg(hsp70l:nog3) embryos were left at 28°C (E,G) or heat-shocked at 10 hpf to induce noggin3 expression (F,H). In a subset of embryos a bead preincubated with recombinant Nodal was placed in the ALPM. Embryos were analysed by in situ hybridisation for lefty1 expression at 18-somites. All embryos are shown as lateral views with dorsal to the right and anterior to the top. Arrows point to most anterior lefty1 expression domain. Numbers in lower right represents the number of embryos that displayed the phenotype represented in the panels.
Figure 7.
Lefty1 is required for Bmp induced repression of spaw.
In situ hybridisation of spaw (in LPM) and ntl (in midline) at 18-somites. Wild-type (A,C) or Tg(hsp70l:bmp2b) (B,D) embryos were heat-shocked at 10 hpf to induce bmp2b expression (B,D). Ectopic expression of bmp2b resulted in the loss of spaw expression in the LPM (B). A subset of embryos were injected with a lefty1 MO (C,D), which resulted in bilateral spaw expression even in the presence of ectopic bmp2b (D). Embryos are shown as dorsal views with anterior to the top and left to the left.
Figure 8.
Schematic representation of lefty1 regulation during LR axis specification.
Two phases of lefty1 regulation can be distinguished. i) At the 1–3 somite stage the KV (shown in red) is formed but the nodal flow (indicated by blue arrows) has not yet been initiated. While at this stage lefty1 is already expressed in midline (shown in blue) spaw expression is still absent from the embryo. Thus, this early lefty1 expression is induced independent of Nodal but does depend on Bmp activity. Most likely, robust lefty1 expression is required prior to the initiation of LR axis specification to prevent ectopic activation of spaw in the right LPM later on. ii) At the 5-somite stage spaw expression becomes apparent in the perinode region (yellow area flanking the KV) and from the 10-somite stage onward (up to the 25-somite stage) spaw is expressed unilateral in the left LPM (yellow-boxed area). Our results demonstrate that at this second phase both Spaw/Nodal and Bmp activity are required independently to maintain lefty1 expression in the midline. Lefty1 in the midline antagonises Spaw and prevents it from crossing the midline where it would induce its own expression in the right LPM.