Effect of T/C substitution 5′ to the mGATA and hGATA sequences as assessed by EMSA and transfection analysis.

(A) EMSA. DNA binding activities of wild-type mGATA and hGATA probes are compared with that of mutated mGATA*C and hGATA*T probes in EMSA, using K562 nuclear extracts. A and B complexes are indicated by arrows. Probe sequences are described (right panel). (B) Competitive gel mobility shift assay. Nuclear extracts from K562 cells were incubated with labeled mGATA probe, in the absence (−), or in the presence of 100- and 200-fold molar excess of the unlabeled indicated competitor, and 200-fold molar excess of SP1 cold competitor. Bands A and B are indicated by arrows. (C) Titration of GATA-1 and B complex with mGATA*C and hGATA*T probes in EMSA. Binding affinity studies of mGATA*C and hGATA*T probes were performed by Scatchard plot analysis as described for wild-type mGATA and hGATA DNA probe in Figure 5. The Kd values were calculated for each DNA/protein complex. (D) Functional studies by transfection. In the murine −899 mGATA*C construct, the murineTGATAA sequence was replaced by humanCGATAA sequence. In the human −813 hGATA*T construct, the human CGATAA sequence was replaced by the murineTGATAA. The murine 899 mGATA*C with hGATA site (open box *C) and the human 813 hGATA*T construct with mGATA site (gray-filled box *T) were transfected in HEL, K562, and LIN-175 cells. CAT activities are compared with the wild-type murine and human GPIIb promoter constructs.