Files in this Data Supplement:
Fig. S1. Targeting of human and zebrafish SRGAP2 by miR-218. (A) Schematic of the miR-218 seed match in the SRGAP2 3′ UTR. An 8-mer seed is present in human SRGAP2 and a 7-mer seed is present in zebrafish srgap2. Very little sequence conservation exists outside of the seed sequence. (B) Overexpression of miR-218 in HeLa cells can repress the luciferase activity of constructs containing the human or zebrafish SRGAP2 3′ UTR. miR-126 overexpression had no effect on these constructs. *, P<0.05 compared with control. Data are mean + s.e.m.
Fig. S2. miR-218 knockdown in zebrafish. (A) Schematic indicating a portion of the precursor sequences of the three miR-218 genes in zebrafish (miR-218a-1, miR-218a-2 and miR-218b). Sequences overlapping the region targeted by miR-218 MO1 and miR-218 MO2 that differ among miR-218 genes are indicated in blue. The location of the mature microRNA is indicated beneath and is also indicated in red. The Drosha cleavage site, which is targeted by the miR-218 MOs is indicated by an arrow, as is the Dicer cleavage site. Watson-Crick base-pairing is indicated by vertical lines, and colons indicate base-pairing that is not Watson-Crick for all of the miR-218 isoforms. (B) Knockdown of miR-218 was quantified by real-time qRT-PCR in control embryos or embryos injected with 5 or 10 ng of miR-218 MO2. Expression of slit2 and slit3 was also monitored by real-time qRT-PCR. (C) The phenotype of embryos injected with miR-218 MO2 is shown. miR-218 morphants have edema and circulation defects and minor vascular defects. Scale bar: 100 µm.
Fig. S3. Quantification of Slit/Robo/miR-218 morphant phenotypes. Phenotypic analysis of miR-218 (MO1 and MO2), slit2, slit3, robo1 and robo4 morphants at 48 hpf. The percentage of embryos with the indicated defects was averaged across multiple independent experiments. Circulation defects includes reduced circulation in the axial vessels, head vessels or in the intersomitic vessels. The average + s.e.m. across multiple experiments is shown. The total numbers of embryos analyzed were as follows: control, n=355; slit2 MO, n=115; slit3 MO, n=212; miR-218 MO1, n=229; miR-218 MO2, n=100; robo1 MO, n=205; robo4 MO, n=350.
Fig. S4. Slit/Robo signaling does not appear to affect vessel sprouting and miR-218 does not control vascular integrity. (A) Images of the sprouting of the intersomitic vessels (isv) in morphants by confocal microscopy. Tg(kdrl:GFP) labels the endothelial cells. The somite boundaries are indicated by F59 staining (red). vegfa morphants had severe sprouting defects, but sprouting appeared unaffected in other morphants. Scale bar: 100 µm. (B) Sprouting phenotypes were quantified by determining the percentage of the intersomitic vessels that were present and the percentage of intersomitic vessels that crossed the myoseptum. The total numbers of embryos analyzed were as follows: control, n=29; slit2 MO, n=19; miR-218 MO1, n=14; robo1 MO, n=13; robo4 MO, n=17; vegfa MO, n=20. Data are mean + s.e.m. (C) Vascular integrity was assessed by confocal microscopy of transverse Vibratome sections of embryos at 48 hpf. The axial vasculature was lumenized in miR-218 morphants as determined by the presence of gata1:dsRed+ erythrocytes within the axial vessels. Scale bar: 25 µm.
Fig. S5. Phenotypic analyses of Slit/Robo morphants or mutants. (A) Phenotypic analysis of slit2, slit3, robo1 and robo4 morphants at 48 hpf. slit2, robo1 and robo4 morphants displayed circulation defects (gata1:dsRed) but normal vascular patterning (kdrl:GFP). Pericardial edema is indicated by a black arrow. slit3 knockdown results in defects in the intersomitic vessels and dorsal longitudinal anastomotic vessels, including broken or missing vessels (white arrows); however, pericardial edema does not occur in slit3 morphants. Whereas circulation is defective in the intersomitic vessels and dorsal longitudinal anastomotic vessels, circulation appears normal in the axial vessels. h, head; ht, heart; da, dorsal aorta; pcv, posterior cardinal vein; isv, intersomitic vessel; dlav, dorsal longitudinal anastomotic vessel. (B) robo2−/− (astray) mutants do not have obvious circulation defects or edema at 72 hpf. The pou4f3:GFP transgene marks the retinal ganglion cell axons (arrow), the migration of which is defective in astray mutants. Scale bars: 100 µm.
Fig. S6. Expression of Slit/Robo components during heart field migration. (A) Expression of slit2, robo1 and robo4 was determined by fluorescent in situ hybridization of 20-somite stage embryos. Shown are transverse sections, dorsal to the top. slit2, robo1 or robo4 are indicated in green in the merged images, whereas Tg(kdrl:GFP), which marks the endocardium, is indicated in red in the merged images. slit2 appears to be expressed in the floor plate of the neural tube, in the endoderm and also in the endocardium. robo1 appears to be expressed in the endoderm, myocardium and the endocardium, whereas robo4 appears to be expressed in the dorsal neural tube and at low levels in the endocardium. fp, floor plate; ec, endocardium; en, endoderm; nt, neural tube; m, myocardium. Scale bar: 50 µm. (B) Expression profile of miR-218, slit2, slit3, robo1, robo2 and robo4 in sorted cells from the indicated transgenic zebrafish lines at 18-20 somites. kdrl:GFP labels the endothelial/endocardium, myl7:GFP labels the myocytes and sox17:GFP labels the endoderm. etsrp (etv2) and hand2 were used as positive controls of endothelial/endocardial and myocardial lineages, respectively. The mean ± s.e.m. of three independent experiments is shown.
Fig. S7. Characterization of migration defects in slit2 morphants. (A) Migration tracks of individual kdrl:ras-mCherry-expressing endocardial cells in uninjected control embryos or slit2 morphants. Scale bar: 25 µm. (B) Myocardial directionality was quantified in uninjected controls and slit2 and robo4 morphants. No significant differences in directionality were noted. ns, not statistically significant. Data are mean + s.e.m. (C) Example of multiple lumens that form in slit2 morphants (arrows). Scale bar: 50 µm.
Movie 1. Migration of heart fields in wild-type zebrafish embryos. Maximum projection of fluorescent confocal time-lapse images. z-stacks were acquired every 5 minutes, beginning at ∼17 somites. Cardiomyocytes are labeled by Tg(myl7:GFP) expression (green) and endocardial cells by Tg(kdrl:ras-mCherry) expression (red). Video speed is ten frames/second. Both the myocardial and endocardial fields migrate collectively as coherent sheets.
Movie 2. Migration of heart fields in slit2 morphants. Maximum projection of fluorescent confocal time-lapse images. z-stacks were acquired every 5 minutes, beginning at ∼16 somites. Cardiomyocytes are labeled by Tg(myl7:GFP) expression (green) and endocardial cells by Tg(kdrl:ras-mCherry) expression (red). Video speed is ten frames/second. In the slit2 morphant endocardium, collective cell migration is impaired and individual endocardial cells extend multiple filopodial protrusions.