A nuclear factor binds to the critical element in the human CD34 5′ UTR. (A) Sequence of the oligonucleotide +48/+80 used as a probe in EMSA. Putative Sp1 and μE4 sites are overlined. The potential Y box motif is boxed. (B) Nuclear extracts (10 μg) from CD34⁺ (KG1a and RPMI-8402; lanes 2 and 4, respectively) and CD34⁻ (Jurkat and HeLa; lanes 3 and 5, respectively) were incubated with radiolabeled +48/+80 probe. The protein/DNA complexes (marked with an arrowhead) were resolved on native acrylamide gel. No protein was added to the reaction in lane 1. (C) EMSA with the +48/+80 probe and 10 μg of nuclear extract prepared from cell lines indicated above lanes. The gel contained 4% acrylamide:bis-acrylamide at the ratio 29:1 to allow the separation of 2 closely migrating complexes (marked with arrows). The electrophoresis proceeded for 4 hours at 4°C, thus the unbound probe migrated out of the gel. (D) Sp1 is not involved in the binding to the +48/+80 region. Probe shown in (A) was incubated with 10 μg of nuclear extracts of KG1a (lanes 1 to 6) or Jurkat (lanes 7 to 12) with electrophoresis conditions as in (C). Lanes 2 and 8 contained unlabeled self-competitor, lanes 3 and 9 contained the Sp1-binding oligonucleotide from the human CD34 promoter (base pairs −68/−44) as competitor, and lanes 4 and 10 contained the Sp1-binding oligonucleotide from the murine PU.1 promoter (base pairs −114/−90).36 An antiserum raised against Sp1 or nonimmune serum were added to the binding reactions shown in lanes 5 and 11, or 6 and 12, respectively.