|
|
Blood, Vol. 113, Issue 8, 1794-1804, February 19, 2009
miR-451 regulates zebrafish erythroid maturation in vivo via its target gata2
Blood Pase et al.
113: 1794
Supplemental materials for: Pase et al
Files in this Data Supplement:
- Table S1. Oligonucleotides used in this study (PDF, 103 KB)
- Table S2. Potential miR-144 binding sites in mammalian GATA2-3′UTRs (PDF, 20 KB) -
Potential miR-144 binding sites in the human, mouse and rate GATA2-3′UTRs, recognised by computer prediction1 or manual examination, aligned vertically with columns to indicate homology (grey shading indicates high sequence homology). Alignment is shown against miR-144 itself (in red). Numbering is the position in the 3′UTR taking the first bp after the stop codon as position 1.
- Figure S1. Genomic organisation of the miR-144/miR-451 locus (JPG, 92.7 KB)
-
Manual alignment of mature miR-144 and miR-451 genomic sequences in six vertebrate species. Sequences derive from the miRBase database (Release 10).
- Figure S2. miR-144/451 locus expression levels by semi-quantitative RT-PCR (JPG, 176 KB)
-
RT-PCR for actin and miR-144/451 run for 25, 30 and 35 cycles on cDNA prepared from 24 hpf wild-type (WT), meunier (mnr) and cloche (clo) embryos, demonstrating equal actin loading (25 cycle result) and reduced miR-144/451 levels in mnr and clo mutants compared to WT (35 cycle result), concurring with the WISH data of Figs. 1 and 2. Two controls demonstrate no genomic DNA contamination: reverse transcriptase absent (RT-) lanes are clear, and the primers for actin span an intron, generating a larger PCR product from genomic DNA (gDNA) template. As this RT-PCR detects the pri-miRNA transcript (Fig. 1), the lack of accumulation of pri-miRNA transcript in mnr indicates that the defect is not a miR-144/451 transcript processing defect proximal to and including Drosha. Methods note: mnr and clo cannot be recognised at 24 hpf under simple brightfield microscopy. Hence, genotyping was necessary. All cDNA pools were prepared from RNA prepared from the posterior portions of 25 embryos transected coronally at the level of the start of the yolk extension. For mnr and WT pools, the embryos were collected from a mnr+/− incross, DNA prepared from the rostral part was PCR genotyped at SSLP Z3984, and the caudal part corresponding to 25 accurately-genotyped embryos were pooled for cDNA preparation. clo embryos were recognised at 24 hpf by their reduced vascular fluorescence in a line carrying the clo mutation with the Tg(fli1:EGFP) transgene.
- Figure S3. O-dianisidine staining for hemoglobin (JPG, 71.9 KB)
-
(A) Randomly-selected 48 hpf embryos of wild-type (WT+/−) or mnr genotypes showing normal degree of O-dianisidine staining in both groups. (B) Randomly-selected 48 hpf embryos injected with control morpholino (MO-Con) or morpholino antagonist of miR-451 (MO-451), showing normal degree of O-dianisidine staining in both groups.
- Figure S4. Expression of hematopoietic lineage markers in mnr by whole mount in situ hybridisation (JPG, 403 KB)
-
(A, B) Normal tal1 expression at 12 hpf was observed in all embryos from a mnr+/− incross, indicating normal tal1 expression in mnr. A – lateral view; B – flat mount view. (C–F) Normal spi1 expression at 14 and 20 hpf was observed in all and in 45/47 embryos respectively of a mnr+/− incross, indicating normal spi1 expression in mnr. C – lateral view; D – flat mount view. (G, H) Reduced expression of the myeloid marker lcp1 in mnr embryos at 48 hpf in a mnr+/− incross. E – lateral view; F – ventral view (I, J) Reduced expression of the myeloid marker lyz expression in mnr embryos at 48 hpf. (K, L) Reduced expression of the thrombocyte marker mpll in mnr embryos at 60 hpf. All embryos are from mnr+/− incrosses. For G–L, mnr embryos were recognized by their small eyes and head. G–L are lateral views. Wild-type controls are sibling embryos. In B, D, F, X/Y indicates number of embryos with phenotype shown in panels / total number of embryos. For G–L, X/Y indicates number of embryos with phenotype shown in panel / total number of embryos of that genotype.
- Figure S5. Tight genetic linkage between the mnr locus and loss of miR-144/451 expression (JPG, 156 KB)
-
(A) Individual embryos from a mnr heterozygote incross were first characterised for the miR-144 expression phenotype (bottom row, N=normal, =reduced) and then PCR genotyped at the SSLP marker z3984 (independently positioned to within 3 cM of mnr by mapping for its primary mpx-deficiency phenotype). Agarose gel electrophoresis of PCR genotyping reactions (upper panel) showed 100% linkage between mnr homozygosity and the reduction/loss of miR-144 expression. (B) Tabulated data demonstrating tight linkage of mnr and loss of miR-144 and miR-451 expression, generated as in (A). 3/64 embryos not genotyping as expected is consistent with z3984 lying 3 cM from mnr.
- Figure S6. Supplementary data corresponding to Fig. 2E–F (JPG, 333 KB)
-
(A) The nuclear:cytoplasmic (N:C) area ratio is a robust parameter of erythrocyte maturity during early zebrafish hematopoiesis. Scatterplots of the nuclear:cytoplasmic (N:C) area ratio in wild-type (WT) and mnr embryos from 30 hours post-fertilization (hpf) to 4 days post-fertilization (dpf); horizontal lines indicate mean values. The N:C ratio decreases in both genotypes as erythrocytes mature; p-values for comparisons within genotypes are shown here, cross-genotype comparisons are tabulated in Fig. 2F. For 2 dpf (48 hpf), the 4 colors (black, blue, red, green) indicate data from 4 independent assays of different groups of embryos on different days. This degree of inter-assay variability translates into the tight descriptive statistics for uninjected embryos in Fig. 2F (mean ± SD for all pooled datapoints) and Fig. 3E (mean ± SE for the 4 independent experiments). (B) The N:C ratio is one aspect of a range of morphological features of reflecting erythocyte immaturity in mnr. Representative fields showing erythrocytes from 2 days post fertilization (dpf) WT and mnr embryos, showing that mnr erythrocytes display several morphological features of immaturity: e.g. larger diameter, a more open pattern of nuclear chromatin, and a more basophilic cytoplasm. The N:C area ratio captured one aspect of these morphological differences, and proved to be a quantitative variable that served as marker of erythrocyte immaturity. Note that the for N:C area ratio calculations, erythrocytes whose morphology was influenced by neighbor effects (such as several shown here) were excluded. The single cells in panels of Fig. 2E are higher power views of single erythrocytes representative of cells in the fields shown here. May-Grünwald/Giemsa stain; scale bar=5 µm.
REFERENCE
1. Betel D, Wilson M, Gabow A, Marks DS, Sander C. The microRNA.org resource: targets and expression. Nucleic Acids Res. 2008; 36(Database issue):D149-53. Epub 2007 Dec 23.
|
|
