HSF2 interferes with the stress-inducible binding of HSF1 to the hsp70 promoter in mitotic cells. (A) Representative immunoblot analysis of coprecipitated HSF1 and HSF2 (n = 3). Lysates from K562 cells, either unsynchronized (U) or synchronized into mitosis by a thymidine-nocodazole block (M), were left untreated (C) or heat shocked (HS; 30 min at 42°C), and immunoprecipitated with an anti-HSF1 antibody or normal rabbit IgG (NS). Input: whole-cell lysates were analyzed by immunoblotting with the indicated antibodies. (B) Representative EMSA analysis of HSF DNA-binding activity (n = 3). K562 cells were transfected with indicated shRNAs and either left unsynchronized (U) or synchronized into mitosis (M) as in A before they were left untreated or subjected to a 30-min heat shock at 42°C. Whole cell lysates were analyzed by EMSA, using a ³²P-labeled probe corresponding to the proximal HSE from the human hsp70 promoter. HSF-HSE indicates the specific DNA-binding complex, CHBA indicates the constitutive HSE-binding activity, and NS indicates nonspecific DNA–protein complexes. (C, top) ChIP analysis of stress-induced (30 min at 42°C) recruitment of HSF1 to the hsp70 promoter in K562 cells transfected and synchronized as in B. NS indicates nonspecific DNA binding. Input represents 10% of the material used in the ChIP assay. (C, bottom) There is an inverse relationship between HSF1 binding and hsf2 mRNA levels. qPCR quantification of HSF1 occupancy of hsp70 promoter and hsf2 mRNA levels of individual samples from heat-treated mitotic samples, transfected with either Scrambled or HSF2-specific shRNA, was plotted and the correlation was calculated (n = 8). (D) qPCR quantification of HSF2 occupancy at the hsp70 promoter in K562 cells transfected, synchronized, and treated as in B (n = 3). Values represent mean + SEM (error bars).