Figure 7.

Elavl1 binds to the gata1 transcript and Elavl1a-binding sites in the 3′-UTR affect gene expression. (A) Association of Elavl1 protein and gata1 transcript demonstrated by RNA immunoprecipitation. Transcripts including the gata1 3′-UTR (WT) or with the putative Elavl1a-binding sites mutated along with Elavl1a expression vectors were transfected into HEK293 cells. IP experiments used no antibody, IgG control antibody, or Elavl1a-specific antibodies, as indicated, to precipitate RNA, and relative levels of the gata1 3′-UTR in the precipitate were quantified by qPCR. Fold-change is shown normalized to the no-antibody control group. **P < .01; ***P < .001. (B) Lysates from MEL cells were likewise immunoprecipitated with ELAVL1-specific antibody and murine Gata1 RNA quantified in the pull-down. **P < .01. (C) Reporter gene assay in zebrafish embryos. Luciferase reporters contained no UTR, zebrafish gata1 3′-UTR, or the mutated isoform 3′-UTR. Relative firefly luciferase activity was normalized by Renilla luciferase activity. All data are averaged from at least 3 experiments. *P < .05; **P < .01. Error bars represent SD. (D) Erythroid lineage-specific reporter assay in zebrafish embryos. Embryos were injected with transgenes regulated by the gata1 erythroid-specific promoter. Relative firefly luciferase activity was normalized by Renilla mRNA luciferase activity. All data are averaged from at least 3 experiments. *P < .05; **P < .01. Error bars represent SD. (E) Quantification of phenotypes for embryos in each study group, including: lane 1, control embryos; lane 2, morphant embryos; or lane 3, morphant embryos, that were also injected with 12.5 ng of gata1 mRNA, which can partially rescue the blood cell phenotype. Embryos were scored as relatively normal, strongly positive for o-Dianisidine–stained cells, even though the cells fail to circulate (o-D+) or were negative for o-Dianisidine (o-D−). χ², P < .001. (F) Results from qPCR experiments measuring hbbe3 transcripts (in lane 1, control Mo-injected; lane 2,elavl1a Mo-injected; or lane 3, elavl1a Mo-injected embryos) that were coinjected with gata1 mRNA. Transcript levels in control and RNA-injected morphants are plotted relative to those from elavl1a morphants; **P < .01. (G) Shown are representative 48 hpf embryos stained for hemoglobin (o-Dianisidine, top panels) or processed by in situ hybridization for globin transcripts (hbbe3, bottom panels). For each representative sample, n > 20. MT, mutated; ns, not significant.