CLL patients show heterogeneity of NF-κB DNA binding. (A) NF-κB DNA binding activity in cell nuclear extracts was measured using electrophoretic mobility shift assays. An oligonucleotide corresponding to the consensus sequence to NF-κB was radiolabeled and incubated with 2 μg nuclear extract. The DNA-polynucleotide complex was restored by electrophoresis in a 4% native polyacrylamide gel in 0.5× Tris-borate/EDTA buffer. The gels were dried, and protein binding was visualized by autoradiography. To demonstrate specificity, a cold competitor assay was performed on 2 μg nuclear extracts of CLL patient k (B). Cold, nonradiolabeled NF-κB and a nonspecific oligonucleotide, AP1, was added at 10× and 25× the concentration of radiolabeled NF-κB and incubated for 30 minutes, before radiolabeled NF-κB incubation. (C) For qualitative analysis of NF-κB subunits, super-shift analysis was performed on NF-κB using p50, Rel A, p52, Rel B, and c-Rel antibodies and normal rabbit sera; 2 μg nuclear extracts from CLL nuclear extracts were used for these experiments. The different lanes marked, none, p50, Rel A, p52, Rel B, c-Rel, and Rab ser (pre-immune rabbit sera) represent incubation with different antibodies. They were then incubated with a radiolabeled oligonucleotide corresponding to the consensus sequence of NF-κB for 30 minutes. Ab indicates the different antibodies used; None, no antibody was incubated. White arrows indicate the antibody-protein-DNA complexes; black arrows, the protein-DNA complexes. Free DNA has been omitted. (D) Nuclear extracts from sample a through sample e were Western blotted for β-actin to assess the relative integrity of the nuclear protein extracts between samples and to confirm equal loading of nuclear proteins in the experiments.