Fig 1. The regulation of Cebpa in PUER cells.

A. A 78kb region surrounding the Cebpa TSS is shown. The boxes represent putative CRMs previously identified using evolutionary conservation and analyzed using sequence-based thermodynamic modeling [21]. Green and magenta boxes represent enhancers and silencers respectively. The activators and repressors inferred by the model are indicated. B. Wright Giemsa stains of PUER cells in uninduced IL3 (top), 7-day OHT-induced IL3 (bottom left), and 7-day OHT-induced GCSF (bottom right) conditions. Uninduced cells have a blast morphology with high nucleocytoplasmic ratio. Cells induced in IL3 conditions have a vacuolated cytoplasm and low nucleocytoplasmic ratio, while induction in GCSF results in cells with segmented nuclei. C. Time series of the ratio of Cebpa and Hprt expression measured by RT-RTPCR during the differentiation of PUER cells. Relative expression has been normalized to average relative expression in uninduced PUER cells. -48 hours and 0 hour points are both measurements from uninduced cells. With the exception of 96 hours GCSF+OHT, for which N = 2, N ≥ 3. Error bars show standard error. D. Schematics of reporter vectors, based on the pGL4 backbone (Promega), which contain the Cebpa promoter immediately upstream of luc2 either with (below) or without (above) a distal CRM located downstream of the SV40 Poly(A) signal. E. Normalization of Firefly luminescence against Renilla luminescence to correct for sample-to-sample variation in transfection efficiency. Points are independent Firefly and Renilla luminescence measurements for Cebpa(0) (blue) and Cebpa(7) (red). Luminescence is reported in relative luminescence units (RLUs). The ratio of Firefly and Renilla luminescence was estimated as the slope of the best-fit line (solid) determined by robust errors-in-variable (EIV) regression. Dashed lines represent the 95% confidence interval for slope determined by bootstrapping (see Methods).