Establishment of Notch1-IC-, Notch2-IC-, and EBNA2-expressing EREB2-5 cells. (A) Expression vectors for Notch1-IC, Notch2-IC, and CAT (negative control) were stably transfected in EREB2-5 cells, a conditionally immortalized lymphoblastoid cell line, in which EBNA2 function depends on the presence of estrogen. Notch1-IC, Notch2-IC, and CAT were cloned downstream of a bidirectional doxycycline-dependent (Tet-O7) promoter. In the opposite orientation, truncated ΔNGFR was cloned to detect or sort Notch1/2-IC– and CAT-expressing cells. The tet-on cassette is coding for the reverse transactivator and the KRAB repressor. The hygromycin expression cassette (Hyg) allows selection of stably transfected cell clones, and the EBNA1 gene ensures episomal maintenance of the expression plasmids. Transfection of the expression plasmids into EREB cells resulted in the following cell lines: Notch1-IC/EREB, Notch2-IC/EREB, and CAT/EREB. Addition of estrogen results in the activation of EBNA2, and addition of doxycycline leads to the expression of Notch1-IC, Notch2-IC, and CAT as well as ΔNGFR. (B) Experimental design that was valid for all kinetic experiments: (i) before induction, cells were deprived of estrogen for 3 days. Cells were harvested directly after estrogen withdrawal (0 hours), 8 hours and 24 hours after doxycycline addition (Notch1-IC, Notch2-IC, CAT) and 4 hours and 24 hours after estrogen addition to CAT/EREB cells (EBNA2). (ii) To examine long-term effects of Notch1/2-IC expression, doxycycline was added immediately after estrogen depletion and cells were harvested 3 days later. The control cell line CAT/EREB was treated in the same way as Notch1-IC/EREB and Notch2-IC/EREB cells. Before the preparation of RNA or protein isolation, ΔNGFR+ cells were purified by MACS separation to enrich the cells with a transcriptional response to doxycycline. (C) Expression of Notch1-IC and Notch2-IC in stably transfected Notch1-IC/EREB (Notch1-IC), Notch2-IC/EREB (Notch2-IC), and CAT/EREB (CAT) cell lines was analyzed by Western blotting. Cells were treated as outlined in panel B. Proteins were harvested at the indicated time points after doxycycline addition. Notch1-IC and Notch2-IC were detected by specific antibodies raised against the intracellular part of Notch1 (left part) or Notch2 (right part). Equal protein loading was controlled by an α-actin antibody. The weak signal of approximately the same molecular weight as Notch2-IC in Notch2-IC/EREB and in CAT/EREB cells in the absence of doxycycline results from endogenous Notch2 expression in B cells and corresponds to the intracellular membrane-anchored part of the Notch2 receptor (Notch2™).