GATA-1 binds and activates the cyclin D1 promoter, while FOG-1 represses its transcription. (A) The pGL3-basic construct containing 2 kb of the cyclin D1 promoter upstream of the luciferase gene. Putative GATA-1–binding sites are indicated as diamonds, and dotted vertical lines indicate serial deletions made of the cyclin D1 promoter. D1p-1 is the full 2-kb fragment, and D1p-7 is the shortest 76-bp fragment. (B) Luciferase assay performed by transient transfection of NIH3T3 cells with the indicated micrograms of pcDNA3 vector containing GATA-1, GATA-1s, GATA-1 V205G, and FOG-1. Luciferase levels are shown as fold change relative to cells transfected with reporter construct alone. The total amount of DNA transfected was kept constant by transfection with empty pcDNA3 vector. One representative of 2 experiments, each in triplicate, is shown. (C) Luciferase assay performed as in panel B with luciferase vectors containing mutations in the 3 putative GATA-binding sites within the last 76 base pairs. Mutation of the first GATA site (D1p-7 mut-1) showed reduced response to GATA-1, whereas mutation of 2 other putative GATA sites (D1p-7 mut-2 and D1p-7 mut3) does not alter GATA-1 activation of the cyclin D1 promoter. D1p-1 and D1p-7 are described in panel A. One representative of 3 experiments, each in triplicate, is shown. (D) Chromatin immunoprecipitation assay performed in primary mouse fetal liver–derived megakaryocytes with primer sets directed toward the functional GATA-binding site identified in luciferase assays, as well as, downstream (control primers 1) and upstream (control primers 2) primer sets. Control rat IgG and anti–GATA-1 (N6) antibodies were used for immunoprecipitation and indicate GATA-1 enrichment at the functional GATA-1–binding site in the cyclin D1 promoter. P values were generated using the Student t test and are as follows: *P ≤ .04; #P ≤ .01. One representative of 2 experiments, each in triplicate, is shown. (E) Real-time quantitative PCR data showing expression levels of cyclin D1 mRNA following reconstitution of GATA-1 KD megakaryocytes with the empty MigR1 retroviral vector or containing wild-type GATA-1, GATA-1s, or GATA-1 V205G. Values are shown relative to empty MigR1-infected GATA-1 KD expression levels. P values were generated using the Student t test and are as follows: *P ≤ .006; #P ≤ .007. Changes between expression values of MigR1 and MigR1-G1s were not statistically significant. One representative of 3 experiments, each in triplicate, is shown. For each study in B-E, the means ± 1 SD is shown.