Figure 1.
nr2f1a is conserved among vertebrates.
(A) Comparison of the amino acid sequence of zebrafish nr2fIa to orthologs from other species by using ClustalW2 software. Identical amino acids are marked (*) below the sequence. The yellow bar represents a putative ligand binding domain and green box area indicates a zinc-finger DNA binding domain. Squared box indicates a conserved LLLRLP motif. (B) Schematic drawing shows that zebrafish nr2f1a protein contains 411 amino acids with the DNA binding domain (DBD) and ligand binding domain (LBD). (C) Cladograms provided from phylogenetic analysis of nr2f1a orthologs in different species using ClustalW2-phylogeny with the neighbor-joining method. The number (bootstrap values) at the nodes represent the possibility of groupings between 2 different nr2f members from different species. (D) Percent amino acid identity matrix among mouse (Mus_musculus) Nr2f2 and zebrafish (Danio_rerio) nr2f1a, nr2f1b and nr2f2.
Figure 2.
Spatiotemporal expression of nr2f1a during development.
(A) nr2f1a expression can be observed at the 18S stage in the lateral plate mesoderm (lpm), the telencephalon (t), diencephalon (d) and hindbrain (h). (A’) Dorsal view of embryos show that nr2f1a is expressed at lpm and the axial vessels (av). (B, B’) At 24 hpf, nr2f1a is expressed in the telencephalon (t), diencephalon (d), hindbrain (h), as well as in vessels (v), and caudal vein plexus (CVP) of the trunk. B’ is an enlargement of B. (C, D) Cross sections of embryos from B’ show that nr2f1a is expressed in dorsal aorta (da), posterior cardinal vein (pcv), and caudal vein plexus (CVP). (E, E’) At 30 hpf, nr2f1a expression continues in the head, vessels (v), intersegmental vessels (ISV) and caudal vein plexus (CVP) of the trunk. E’ is an enlargement of E. Scale bars in all figures represent 100 µm.
Figure 3.
Knockdown of nr2f1a causes defects in zebrafish vascular development.
(A–H) Loss of nr2f1a showing ISV growth defect (yellow arrows in F) and mis-pattern plexus at the CVP (caudal vein plexus) (white arrows) compared to wild-type control (E) at 30 hpf. E and F is enlargement of C and D, respectively. At 48 hpf, the swallowed plexus becomes obvious (H). At 30 hpf, in uninjected control embryos, intersegmental vessels (isv) have reached the DLAV at the dorsal aspect of the embryo (C, E) and the caudal vein plexus (cvp) formed honeycomb-like structures at the tail (E, white arrows). At the same stage ISVs are stalled at mid-somite in nr2f1ae2i2 morphants (D, F). (I) Quantification of percentage of completed ISV shows a ∼50% increase compared to nr2f1a morphants (n = 26 in wt and n = 34 in nr2f1aMO) at 30 hpf. (J) Quantification of loop formation at CVP shows a 4-fold decreased in nr2f1a morphants (n = 10 in wt and in nr2f1aMO) at 48 hpf. (*** refers to p<0.0001 by an unpaired student’s t-test. Scale bars are 200 µm for A–D, and 100 µm for E–H.
Figure 4.
Morpholino-knockdown efficiency of nr2f1a in zebrafish embryos.
(A) Scheme shows nr2f1a morpholino targeting the pre-mRNA structure of nr2f1a and suggested mis-splicing fragment which can be detected by a nr2f1a_MO_f and nr2f1a_MO_r primer set. (B) cDNA from uninjected controls or nr2f1a morphants (injected with 4 ng morpholino) underwent PCR with primers for the housekeeping control gene (GAPDH), or for the nr2f1a gene. Morphants injected with nr2f1ae2i2, GAPDH levels are unchanged while the amount of wild-type nr2f1a product (683 bp) got diminished and a new lower molecular weight band appeared (120 bp), representing mis-splicing caused by the morpholino.
Figure 5.
Loss of nr2f1a results in circulation defects and pericardial edema.
(A–E) nr2f1ae2i2 MO was injected into transgenic Tg (fli:eGFPy1; gata1:dsRed) embryos with GFP-labeled endothelial cells (A, C) and dsRed-labeled blood cells (B, D, E). Loss of nr2f1a showed a mis-pattern at ISV and at the caudal vein plexus (CVP) and results in circulation defects in severe (D, no circulation and blood stock) or mild (E, blood stock at CVP and shorter circulation) at 52 hpf compared to wild-type fish (B). (F) Circulation defects at the ISV/DLAV (∼68% in morphants) and slow to lose axial circulation of the aorta/vein in the trunk region (∼58% in morphants) are quantitated in wt (n = 37) and nr2f1aMO (n = 42) at 48–52 hpf. (G, H) Representative edema fish and quantitative results from three independent experiments showed 65% of nr2f1a morphants (n = 69) with mild to severe pericardial edema compared to wt (n = 30). The scale bar in A–E represent 200 µm and in G is 500 µm.
Figure 6.
Knockdown of nr2f1a can be rescued by overexpression of nr2f1a.
In uninjected control embryos, intersegmental vessels (isv) have reached the DLAV at the dorsal region from 24–30 hpf (A, white arrowheads) and caudal vein plexus (cvp) were formed honeycomb-like structures at the tail around 48 hpf (E, white arrows). At the same stage, ISVs are stalled at mid-somite (B, hollow arrowheads) and less/no honeycomb structure is formed at CVP (F with no arrows) in nr2f1ae2i2 morphants. Overexpression of nr2f1a driven by fli promoter had no obvious defect in vasculature (C, G), but rescued the defect of ISV stalling (solid arrowhead) (D) and restored the honeycomb structure at CVP (H). (I) Quantification of percentage of completed ISV at 26 hpf shows a ∼40% increase in rescued embryos compare to nr2f1a morphants. Wild-type embryos had 62±3.1% complete ISV; nr2f1aMO had 18±8.0% complete ISV; fli:nr2f1a overexpression embryos had 63±2.6% complete ISV and rescued embryos had 57±2.7% complete ISV. (J) Quantification of loop formation at CVP showed a 2-fold increased in rescued embryos compared to nr2f1a morphants at 48 hpf. Wild-type embryos (19.6±0.9); nr2f1aMO (4.2±1.5); fli:nr2f1a overexpression embryos (22.6±1.3) and rescued embryos (10.6±1.6). (*** refers to p<0.0001, ** refers to p<0.001 and * refers to p<0.05 by an unpaired student’s t-test. Data are represented as means ± SEM). Scale bars are 100 µm for A–H.
Figure 7.
nr2f1a is required for the growth of ISV cells.
(A–D) TUNEL labeling and AO staining was used to detect apoptotic cells in wild type, and nr2f1ae2i2 morphants. Although some apoptotic cells were observed on the skin, cell death in vascular regions was not elevated compared to wild type controls. (E–H) The number of cells forming each ISV counted in wild type control Tg(kdrl:mCherry; fli1a:negfp)y7 (E) and nr2f1a morphant embryos (F) at 32 hpf. (G) Proportional distribution of ISVs containing 1–5 cells and (H) average ISV cells per ISV counted in both wild type and nr2f1a morphant. (*** refers to p<0.0001 by an unpaired student’s t-test).
Figure 8.
nr2f1a modulates the expression of vascular markers.
(A–H) Compared to wild type controls (A, C, E, G), expression of the venous markers mrc1 (D) and flt4 (F) were reduced in the trunk of nr2f1a morphants at 24 hpf. In addition, expression of the arterial marker ephrinb2 (H) and pan-vascular marker stabilin decreased (B). (I) Quantification of the relative expression level by qPCR assay showed a ∼25% to 50% reduced expression in vascular markers flt4 (0.66±0.14), EphB4 (0.46±0.08), gridlock (0.73±0.06) and ephrinb2 (0.59±0.05) in nr2f1a morphants. *** refers to p<0.0001, ** refers to p<0.001 and * refers to p<0.05 by an unpaired student’s t-test. Data are represented as means ± SEM. Scale bars represent 100 µm in A–H.
Figure 9.
Regulatory relationships between nr2f1a and Notch.
(A–B) nr2f1a expression is downregulated in rbpsuh morphants (B) as compared to untreated control embryos (A). (D–E) nr2f1a expression is downregulated after treatment with DAPT (E) as compared to DMSO control embryos (D). (C, F) Quantification of the relative nr2f1a expression level by real-time qPCR assay shows a ∼50% reduced expression in rbpsuh morphants (C) and DAPT-treated embryos (F) compared to uninjected embryos and DMSO-treated controls. Scale bars represent 100 µm in A, B, D and E.