Sequence and expression analyses of zebrafish k-ras.
(A) Zebrafish k-ras nucleotide and putative amino acid sequence. k-ras-MO1 and k-ras-MO2 binding sites are underlined. (B) Alignment of zebrafish K-ras with known Ras proteins of human, mouse and zebrafish. (C) Phylogenetic analysis of Ras proteins. (D) k-ras expression during embryonic development. k-ras transcripts were detectable from one cell stage (i) and then persist throughout the whole embryos (ii–vi). (i), one cell stage, side view; (ii), 3 hpf (hours-post fertilization), top view; (iii), 10 hpf, dorsal view, anterior to the top; (iv), 10 hpf, bottom view, dorsal to the top; (v), 20 hpf, lateral view, anterior to the left and dorsal to the top; (vi), 20 hpf, dorsal view, anterior to the left.

Disruption of zebrafish K-ras signaling resulted in the defective hematopoiesis.
(A) k-ras-MO injected embryo showed empty heart, without or with few red blood cells inside (indicated by arrows in ii and iii), in comparison to wild type embryo, which showed plenty of red blood cells inside the heart (indicated by arrow in i). Embryos at 3 dpf (days-post fertilization). (B) Plenty of circulating red blood cells inside dorsal aorta (DA), posterior cardinal vein (PCV), inter-segmental vessels (Se), caudal artery (CA) and caudal vein (CV) in wild type embryos (i and iii), while no or less circulating red blood cells were found in k-ras-MO injected embryos inside DA, PCV, CA, CV and Se (ii and iv). Embryos at 3 dpf. (C) k-ras-MO injected embryos showed accumulated red blood cells in some sites that away from the circulation. Embryos at 3 dpf. (D) o-Dianisidine staining for wild type embryo showed hemoglobin positive cells inside branchial arches (indicated by arrow in i) and heart chambers (indicated by arrow in ii ), while k-ras-MO injected embryo showed less/negative o-Dianisidine staining for branchial arches (indicated by arrow in iii) and heart (indicated by arrow in iv). Embryos at 6 dpf. (E) o-Dianisidine staining for wild type embryo showed hemoglobin positive cells inside Se (indicated by arrows in i ), dorsal longitudinal anastomotic vessels (DLAV), CA and CV (indicated by arrows in ii), while k-ras-MO injected embryo failed to give the positive o-Dianisidine staining in these corresponding positions (indicated by arrows in iii and iv). Embryos at 6 dpf. All embryos shown are lateral view, with anterior to the left and dorsal to the top.

Disruption of zebrafish K-ras signaling resulted in the disruption of gata-1 and Βe3-globin expression.
(A) K-ras knockdown or over-expression of K-ras-N17 resulted in the disruption of gata-1 expression. Embryos at 20 hpf, lateral view, with anterior to the left and dorsal to the top. (B) K-ras knockdown or over-expression of K-ras-N17 resulted in the disruption of Βe3-globin expression. Embryos at 24 hpf, lateral view, with anterior to the left and dorsal to the top.

Disruption of zebrafish K-ras signaling resulted in the defective angiogenesis.
(A) Both un-injected and k-ras-MO injected fli1-GFP embryos (22 hpf) showed normal development of dorsal aorta and caudal artery (indicted by solid arrows), posterior cardinal vein and caudal vein (indicated by empty arrows). Higher magnifications of the square area in (i) and (ii) were shown in (iii) and (iv) respectively. (B) Un-injected fli1-GFP embryo at 3 dpf showed well-organized inter-segmental vessels (i), while k-ras-MO1 injected (ii, v and vi), k-ras-MO2 injected (iii) or k-rasN17 injected (iv) embryos at 3 dpf showed aberrant and irregularly organized inter-segmental vessels. (C) Alkaline phosphatase staining for k-ras-MO1 injected embryos (3 dpf) showed aberrant trunk blood vessels. (D) Alkaline phosphatase staining showed well-organized SIV (sub-intestinal vein, indicated by arrow) in wild type embryo at 3 dpf (i), while disorganized SIV (indicated by arrow) in k-ras-MO1 injected embryo (ii). Inserted figures in i and ii showed the anterior part of the embryos. All embryos shown in lateral view, with anterior to the left and dorsal to the top.

PI3K-Akt are crucial mediators for K-ras signaling in zebrafish hematopoiesis and angiogenesis.

(A) K-ras knockdown could be rescued by k-ras mRNA, but this rescue was suppressed by wortmannin at lower dose (250 nM). At this concentration, wortmannin itself could not induce hematopoietic defects in controlled phenol-red injected group. Embryo numbers n1 = 37, n2 = 38, n3 = 43 and n4 = 46, from two independent sets of experiments. (B) Hematopoietic defects caused by K-ras knockdown could be rescued by wild type K-ras and K-ras mutant k-rasC40 respectively. Embryo numbers n1 = 475, n2 = 344 and n3 = 80, from >4 independent sets of experiments. (C) Hematopoietic defects caused by K-ras knockdown could be rescued by Akt2 effectively. Embryo numbers n1 = 156 and n2 = 129, from 4 independent sets of experiment. * indicates p<0.05. (D) Angiogenic defects caused by K-ras knockdown could be rescued by wild type K-ras and K-ras mutant k-rasC40 respectively. Embryo numbers n1 = 133, n2 = 92 and n3 = 62, from >2 independent sets of experiments. (E) Angiogenic defects caused by K-ras knockdown could be rescued by Akt2. Embryo numbers n1 = 36, n2 = 20, each group from 2 independent sets of experiments. ** indicates p<0.10. All data are means±SD (standard deviation). Values indicated by the same letter are not significantly different at p<0.01 for (A) and (B), and at p<0.05 for (D).

Expression analyses of zebrafish k-ras in tissues.

RT-PCR analysis of zebrafish k-ras, n-ras and BC048875 expression in adult zebrafish tissues. Most tissues examined, except spleen, show high or medium level of k-ras expression. Zebrafish n-ras and zebrafish BC048875 transcripts were also detectable in all tissues examined at variant levels.

Reduced heart beat rate was induced by K-ras knock-down, and it was able to be rescued by k-ras mRNA co-injection.

The observed heart beat rate (per 30 seconds) at 30 hpf (hours-post fertilization) of wild type embryos, k-ras-MO injected embryos, and k-ras-MO plus k-ras mRNA co-injected embryos respectively, showing the reduced heart beat rate caused by K-ras knock-down and the rescue by k-ras mRNA co-injection. Embryo numbers, n1 = 30, n2 = 26 and n3 = 38. Data are means±SD (standard deviation), *p<0.05.

Values indicated by the same letter are not significantly different at p<0.05.

Determination of K-ras, N-ras and H-ras protein level between wild type and k-ras-MO injected embryos.

K-ras-MO injected embryos (1 dpf, one day post fertilization) showed reduced K-ras protein expression compared to its wild type control, while the expression of N-ras and H-ras was not affected significantly, indicating the specificity and efficiency of K-ras knock-down.

RFP expression analysis at 20 hpf for k-ras-5′UTR-RFP injected embryos, indicating the targeting specificity of k-ras morpholino antisense oligo.

(A) Embryo injected with k-ras-5′UTR-RFP/PCS (red fluorescent protein reporter was down stream of K-ras 5′UTR and was cloned into PCS2 vector) construct, showing strong RFP signal.

(B) Embryo co-injected with k-ras-5′UTR-RFP/PCS and k-ras-MO1, showing very weak RFP signal, indicating the blockage of RFP protein expression by k-ras-MO1.

(C) Embryos from different treatments, showing the different RFP strength under the same exposure. These embryos were (i), injected with k-ras-5′UTR-RFP/PCS alone; (ii), co-injected with k-ras-5′UTR-RFP/PCS and k-ras-MO1; and (iii), wild type embryo with no injection.

k-ras mRNA could not rescue the gastrulation defects induced by RhoA knock down.

RhoA-MO injection can induce gastrulation defects [12] and these defects could not be rescued by the co-injection of k-ras mRNA. Embryos were observed at 1-somite stage. Embryo numbers n1 = 113, n2 = 112, from 2 sets of independent experiments. Data are means±SD. Values indicated by the same letter are not significantly different at p<0.05.

PI3K inhibitor wortmannin or MEK inhibitor U0126 could induce hematopoietic and angiogenic defects similar to the defects induced by K-ras knock-down.

(A) Either wortmannin or U0126 treatment were able to cause the hematopoietic defects. These defects include empty heart, with no or few red blood cells inside heart (indicated by arrows in ii and iii, compared to wild type in i), reduced or lack of normal circulation and reduced number of circulating red blood cells (indicated by arrows in v and vi, compared to wild type in iv), and accumulation of blood cells in some sites away from the circulation (indicated by arrows in vii and viii). All embryos were observed at 4 dpf (days-post fertilization), lateral view, anterior to the left and dorsal to the top.

(B) o-Dianisidine staining for wortmannin or U0126 treated embryos, showing loss or reduction of hemoglobin positive cells overall, especially inside heart and in yolk sac (indicated by empty arrows and block arrows respectively in ii, iii, v and vi, compared to wild type embryos in i and iv). Except for grouped embryos, all other embryos are lateral view, anterior to the left and dorsal to the top. Embryos were observed at 6 dpf.

(C) Either wortmannin or U0126 treatment were able to cause angiogenic defects. Inhibitor treatment for fli1-GFP embryos resulted in disorganized blood vessels, including the missing segmental vessels and/or bearing ectopic vessel sprouts (indicated by arrows in i and ii), similar to the defects caused by K-ras knock-down. Embryos were observed at 4 dpf, lateral view, anterior to the left and dorsal to the top.