(a) Representative PMT traces from EpH4 cells in 2D/3D following transfection with RhoA plasmid. Co-expression with a constitutively active RhoA (Q63L) in EpH4 cells cultured in 3D produces a twofold reduction in the amplitude of circadian oscillations. Black, transfection with empty vector (pcDNA3), control; Red, transfection with Q63L plasmid. Student's t-test, mean±s.e.m., *P<0.05, n=3 replicates. (b) Co-expression with a dominant-negative form of RhoA (T19N plasmid, as shown in red) in EpH4 cells cultured in 2D increases the amplitude of circadian rhythm by approximately 1.7-fold. Black, transfection with empty vector control. Student's t-test, mean±s.e.m., *P<0.05, n=3 replicates. (c) Knockdown of shRhoA in EpH4 cells confirmed by western blot. N=3. (d) Representative PMT traces from EpH4 cells in 2D following lentiviral transduction of an shRhoA vector (red) or an empty vector, control (black). Knockdown of RhoA in EpH4 cells in 2D increases the robustness of the rhythm. The amplitude of the oscillations is increased threefold following knockdown of RhoA. Student's t-test, mean±s.e.m., **P<0.01, n=3 replicates. (e) Representative Luc Assay in EpH4 mammary epithelial cells in 3D culture. Co-expression with a constitutively active RhoA vector in EpH4 cells cultured in 3D reduces transactivation of E-box targets. (f) Co-expression with a dominant-negative form of RhoA in EpH4 cells cultured in 2D increases activation of E-box targets. Data were normalized to β-galactosidase reporter gene and expressed relative to pCDNA3 control. Student's t-test, mean±s.e.m., *P<0.05; **P<0.01 versus Clock+BMAL1 group, n=3 replicates.