Figure 5

The dynamics of hormone-responsive gene expression is consistent with FFL network motifs. (A) BMMФ were treated with 100 nM Dex for indicated time and the expression levels of Dex-responsive genes were determined by RT-qPCR. (B) BMMФ were treated with Dex either alone or in the presence of the protein synthesis inhibitor Chx for 1–3 h; the expression of indicated genes were determined by RT-qPCR and expressed relative to transcript levels in the absence of Dex (‘Control’), with or without Chx (=1). (C) The dynamics of Klf2 expression is consistent with the I1-FFL with a strong repressor (R) as an intermediate regulator. Klf2 expression data (black circles) collected over 9 h was subjected to a global least square analysis (red line) using the equation (1) (Additional file 1). The quality of the fit as determined by calculating coefficient of determination R² (Additional file 1) improved considerably when the expression data were limited to initial 4 h (green line). The numerical solution of the equation (3), which allows for variation in degradation rates of both “R” and Klf2, yields a better fit (blue line) to the Klf2 expression data. The R² for curve fitting analyses are shown in the legend. (D) Glucocorticoid induction of Klf2 in BMMФ derived from Klf9-KO mice (red squares) loses peak-like kinetics, characteristic of I1-FFL controlled genes (blue diamonds). WT and Klf9-KO BMMФ were cultured in the presence of Dex for indicated time and the expression of Klf2, Tsc22d3 and Nr3c1 (GR) was assessed by RT-qPCR as in A. Basal levels of each transcript were set to 1 for untreated BMMФ of each genotype. Error bars are standard errors of the mean.