Characterization of gata1 mutations: T301K, K333R, and vlt. (A) Schematic representation of Gata1 protein with N- and C-zinc fingers marked as NF and CF, respectively. Multispecies alignment of the region containing the mutations: T301K (blue box), K333R (green box), and vlt (R339X, red box). Dr indicates zebrafish; Hs, human; Mm, mouse. (B) Shown is a 3-dimensional protein model of the zebrafish Gata1 C-terminal zinc finger domain. The ribbon structure depicts a 60 amino acid fragment of the zebrafish Gata1 protein (purple) bound to DNA (gray). The location of substituted residues (T301 and K333; yellow ball-and-stick) and the region truncated in vlad tepes (vlt; red) are indicated. Residue T301 is in close proximity to the DNA backbone of the major groove, suggesting its missense mutation may alter DNA binding. In contrast, the side chain of residue K333 points outward from an extended loop, an orientation that may be tolerant of conservative substitution. (C) DNA-binding activity of wild-type (WT) and mutant forms of Gata1 as determined by electrophoretic mobility shift assay with a ³²P-labeled probe for a β/ϵ globin promoter sequence and in vitro–translated proteins. The DNA–protein complexes are seen as slower migrating bands marked by black and white arrowheads. The first lane on the left contains only the probe, and the next lane also contains lysate but no Gata1 protein. Similar amounts of the Gata1 proteins were used in the experiments, as demonstrated by the Western blot (bottom). (D) Transcriptional activity of mutant forms of Gata1 relative to the wild-type (WT) Gata1 protein as measured by a Dual Luciferase Assay. T301K substitution reduced the ability of Gata1 to activate transcription in vitro, whereas K333R substitution did not affect Gata1 activity in vitro.

Effect of lower Gata1 DNA binding activity on erythropoiesis in gata1^T301K/vlt and gata1^T301K/T301K embryos. (A) o-dianisidine staining for hemoglobin (black arrows) to show the reduction or absence of erythrocytes in gata1^T301K/T301K (n = 10/10) and gata1^T301K/vlt (n = 10/10) embryos, respectively, compared with the wild-type (n = 0/20) embryos at 2 dpf. (B) Percentage of GFP⁺ cells in 3 dpf wild-type and gata1^T301K/T301K embryos carrying Tg(gata1:GFP). Fifty to 60 embryos from each genotype were pooled, dissociated, and subjected to flow cytometric analysis. gata1^T301K/T301K embryos showed a 38% reduction in the percentage of GFP⁺ cells compared with the wild-type. (C) Tg(gata1:GFP) as a marker for erythrocytes (red arrows point to the GFP⁺ cells seen in circulation) to show the absence of erythrocytes in the gata1^T301K/vlt embryo (bottom, n = 25/25) compared with the wild-type embryo (top, n = 0/30) at 4 dpf. Original magnification ×50 (A,C).

Expression of hematopoietic genes in gata1 mutant embryos by whole-mount RNA in situ hybridization and qRT-PCR. Lateral views of embryos with the tails to the left and dorsal to the top. (A) In situ hybridization showing the differential expression pattern of the erythroid markers band3, α-globin embryonic forms 1 (αe1) and 2 (αe2), draculin, and gata1 in 15-18 somites gata1^T301K/vlt (band3, n = 16/16; globins, n = 6/6; draculin, n = 5/5; gata1, n = 9/9), gata1^vlt/vlt (band3, n = 8/8; globins, n = 6/6; draculin, n = 5/5; gata1, n = 8/8) and wild-type (band3, n = 6/6; globins, n = 10/10; draculin, n = 13/13; gata1, n = 10/10) embryos. Original magnification ×80. Individual embryos were imaged and genotyped by PCR followed by Taq1 digestion (vlt) or sequencing (T301K). (B) qRT-PCR of Gata1 target genes band3 and βe2 in 40 hpf wild-type, gata1^+/T301K, gata1^T301K/T301K, and gata1^vlt/vlt embryos. Data represents fold change in expression relative to gata1^vlt/vlt. Twenty embryos from each genotype were pooled for RNA extraction and qRT-PCR.

Rescue of hematopoiesis in gata1^T301K/vlt embryos. (A) Recovery of blood circulation in gata1^T301K/vlt embryos between 7 and 13 dpf. gata1^vlt/vlt embryos were bloodless and never recovered blood circulation. Twenty embryos of each genotype were monitored daily. (B) Survival curves of gata1^T301K/vlt and gata1^T301K/T301K (blue) and gata1^vlt/vlt (red) embryos from fertilization to 30 dpf. Twenty embryos of each genotype were monitored daily. (C) Lateral views of the tails of 48 hpf embryos after RNA in situ hybridization with erythroid markers α-globin embryonic forms 1 (αe1) and 2 (αe2), showing the absence of globin expression in the CHT region of gata1^vlt/vlt embryos (n = 14/14) and relatively normal globin expression in gata1^T301K/vlt (n = 7/9), gata1^T301K/T301K (n = 4/4), and wild-type (n = 8/8) embryos. (D) Noncirculating cd41-GFP^low cells in the CHT (white arrows, right inset) of 4 dpf embryos. No significant difference was detected in this cell population in gata1^vlt/vlt (n = 15/15) and gata1^T301K/vlt (n = 25/25) embryos compared with the wild-type embryos (n = 30/30). Original magnifications ×80 (C) and ×50 (D).

Gata1 DNA binding level–dependent myeloid expansion. Lateral views of whole embryos with the head to the right and dorsal to the top. (A) Expression pattern of the myeloid markers l-plastin and mpo in 24-26 hpf gata1^T301K/vlt (l-plastin, n = 9/9; mpo, n = 5/5), gata1^vlt/vlt (l-plastin, n = 2/2; mpo, n = 5/5), and wild-type embryos (l-plastin, n = 6/6; mpo, n = 7/7) showing an expansion of l-plastin expression in the gata1 mutant embryos. (B) Double in situ hybridization showing l-plastin (purple staining, indicated by purple arrows) and mpo (orange staining, indicated by orange arrows) in the ICM/AGM regions of 26-30 hpf gata1^T301K/vlt (n = 4/4), gata1^vlt/vlt (n = 12/12) and wild-type (n = 12/12) embryos. Original magnifications ×50 (A) and ×115 (B). Individual embryos were imaged and genotyped by PCR followed by Taq1 digestion (vlt) or sequencing (T301K).

Normal hematopoiesis in adult gata1^T301K/vlt fish. (A-B) FACS analysis of adult kidney cells. (A) Representative blood cell profile of whole kidney marrow from wild-type and gata1^T301K/vlt adult fish. Blood cell lineages are represented as: R, red blood cells; M, myelomonocytes; P, precursors; and L, lymphocytes. (B) Average cell numbers of each blood cell lineage calculated from 8 wild-type, 5 gata1^T301K/T301K, and 5 gata1^T301K/vlt fish. The differences in lineage cell numbers observed between wild-type and the mutant fish were not statistically significant. RBC indicates red blood cells. (C) Cytospin preparation from wild-type and gata1^T301K/vlt adult fish whole kidney marrow, showing the presence of multilineage cells in the gata1^T301K/vlt fish.

Gata2 knockdown causes a decrease in blood circulation in gata1^T301K/T301K embryos during primitive hematopoiesis. (A) gata2 expression in gata1 mutant embryos analyzed by whole-mount RNA in situ hybridization. Lateral views of the embryo tails show gata2 expression in the posterior ICM. Embryos were harvested at 24 hpf. Individual embryos were imaged and genotyped (wild-type, n = 4/4; gata1^T301K/vlt, n = 5/7; gata1^vlt/vlt, n = 3/3). Original magnification ×50. (B) Bar graphs showing the percentage of wild-type, gata1^+/T301K, and gata1^T301K/T301K embryos with reduced blood circulation 72 hours after gata2 morpholino injection. There was a significant reduction of gata1^T301K/T301K embryos with normal circulation after injection (*). UI indicates uninjected embryos; INJ, gata2 MO injected embryos. (C) By χ² analysis, the number of injected embryos maintaining normal blood circulation was significantly lower for the gata1^T301K/T301K embryos but not for the gata1^+/T301K embryos. The numbers shown in parentheses are the numbers of embryos expected for each phenotype based on wild-type data.