HHT2-related ALK1 mutants are defective in transcriptional activation and dorsoventral patterning of zebrafish embryos. (A) Plasmids encoding HA-tagged wild-type or mutant ALK1 were transfected into HEK293T cells (2 μg each plasmid per 60 mm dish of cells). Forty hours after transfection, the cells were harvested, and the protein expression levels of ALK1 were examined by immunoblotting with an anti-HA antibody. (B) Hep3B cells were transiently transfected with BRE₄-luciferase (0.5 μg) and the ALK1 constructs (0.03 μg each) as indicated. Forty hours later, the cells were harvested and luciferase activity was measured. (C) Wild-type (+: 0.03 μg; +++: 0.09 μg) and individual mutant ALK1 (0.06 μg) were cotransfected into Hep3B cells together with the BRE₄-luciferase reporter plasmid (0.5 μg) as indicated and the luciferase activity was measured 40 hours later. Reporter assays were performed in triplicate in the absence of TGFβ, and Renilla (20 ng) was cotransfected as an internal control. The data represent the mean plus or minus the standard deviation after normalization to Renilla activity. (D) mRNA encoding wild-type or various mutant ALK1 was injected into one-cell embryos with 1 ng per embryo. Because of severe embryo lethality, a lower amount of ca-ALK1 mRNA (150 pg) was used. The phenotype was observed at 24 hpf. The embryos shown are lateral views, with anterior to the left. The arrows indicate the intermediate cell mass. Control: no injection. (E) Expression of ventral ectoderm marker gata2 was examined by in situ hybridization at the shield stage. The embryos shown are ventral views with the animal pole at the top. Note that the ventralized embryos exhibit increased expression of gata2 (ca-alk1) whereas the dorsalized embryos show a reduction of gata2 expression (alk1(K221R)). Control: no injection.