Fig. 5: PPARG induction requires CDK4/6 and CDK2, as well as ERK, inactivation.

a, b Proliferating OP9 preadipocytes expressing APC/C-reporter cell cycle reporter were treated with a combination of Palbociclib and CDK2i (1 μM each) for 24 h, fixed and immunostained for phosphorylated RB protein levels. Plots show the mean and the individual datapoints from 3 to 5 replicate wells in the experiment. Representative of two biological replicates (also see Supplementary Fig. 4). a Percent of proliferating cells (i.e., cells in S/G2/M phases) defined by the number of cells positive for nuclear APC/C reporter fluorescence (also see Fig. 1c and d). b Immunofluorescence staining was used to measure the percentage of cells positive for phosphorylated RB (pRB). c mCitrine-PPARG cells were imaged live during a standard 96 h DMI-induced differentiation protocol, and CDK inhibitors (or vehicle) were added and kept in starting 24 h after the DMI stimulus (experimental scheme, left panel). Dual CDKi refers to co-inhibition with Palbociclib and Tagtociclib, 1 μM each. Percent of cells with PPARG levels higher than the different commitment thresholds at 96 h was calculated (see Fig. 1b and “Methods” section) for control and Dual CDKi-treated conditions (right panel). Plots show the mean and individual datapoints points from 6 replicate wells in the experiment. Representative of three biological replicates. a–c Unpaired two-tailed t-test was used to compare sample means, a ****P = 1.59 × 10⁻⁵, b ****P = 7.5 × 10⁻⁵, c ****P = 7.75 × 10⁻⁸. d Western blot analysis for phospho-ERK (pERK) and total ERK and PPARG levels during adipogenesis. Beta-actin was used as a loading control. Data is representative of two biological replicates. e Scatter plot of pERK levels (nuclear + cytosolic) versus local cell density from single unstimulated OP9 cells in culture, measured 24 h after they were plated at different densities. Line on each plot shows a non-linear least-squares regression fit to the datapoints. Representative of two biological replicates. f Distribution of pERK from single OP9 cells at similar densities (cell density index of 30–45) before, at 24 and 48 h of DMI stimulation. One-way ANOVA with Tukey’s multiple comparisons was used to compare the differences between conditions. ****P = 3.27 × 10⁻¹¹ (pre-stimulus vs. DMI 24 h), ****P = 3.27 × 10⁻¹¹ (pre-stimulus vs. DMI 48 h), ****P = 7.58 × 10⁻⁶ (DMI 24 h vs. DMI 48 h). More than 500 cells were analyzed per condition. g mCitrine-PPARG cells were imaged live during a standard 96 h DMI-induced differentiation protocol, and MEK inhibitor (PD0325901—100 nM) or vehicle was added and kept for the first 24 h after the DMI stimulus. Percent of cells with PPARG levels higher than the different commitment thresholds at 96 h was calculated (see Fig. 1b and “Methods” section) for control and MEKi-treated conditions. Plots show the mean and individual datapoints points from 4 replicate wells in the experiment. Representative of three biological replicates. Unpaired two-tailed t-test was used to compare differences between sample means. ****P = 1.04 × 10⁻⁵ (Control vs. MEKi). h Scheme for the doxycycline-inducible MEK-CA construct (top left). Immunofluorescence images (bottom left) and distribution of HA-tagged MEKCA expression (right) in control and doxycycline-induced cells after 24 h of induction. Scale bar—25 μm. i–k mCitrine-PPARG cells were treated with doxycycline (2 μg/mL) to induce the expression of MEK-CA (or vehicle) during a standard 96 h DMI-induced differentiation protocol, DMI stimulation, and imaged over 96 h. i Representative single cell time courses of PPARG and APC/C cell cycle reporter from control and MEK-CA overexpressing cells. Yellow circle shows the differentiation commitment point. j Percent of cells with PPARG levels higher than the different commitment thresholds at 96 h was calculated (see Fig. 1b and “Methods” section) for control and MEK-CA overexpression conditions. Plot shows means and individual datapoints from 6 replicate wells in the experiment. Representative of two biological replicates. Unpaired two-tailed t-test was used to compare differences between sample means. ****P = 1.36 × 10⁻¹¹. k MEK-ERK signaling may suppress the rise in PPARG both through promoting CDK activity and by directly phosphorylating and suppressing PPARG transcriptional activity, and preventing any positive feedback on its expression levels. l Besides doxycycline for MEK-CA expression, cells were also treated with vehicle, Dual CDKi (Palbociclib—4 μM + CDK2i—1 μM) or MEKi (PD0325901—100 nM) or Dual CDKi + MEKi. Inhibitors (or vehicle) were added and kept in starting 24 h after the DMI stimulus. Percent of cells with PPARG levels higher than the different commitment thresholds at 96 h was calculated (see Fig. 1b and “Methods” section) for control and inhibitor(s)-treated conditions. Plot shows means and individual datapoints from 6 replicate wells in the experiment. Representative of two biological replicates. One-way ANOVA with Tukey’s multiple comparisons test was used to compare differences between conditions. ****P = 3.81 × 10⁻¹³ (Control vs. MEK-CA), ****P = 2.77 × 10^-10 (Control vs. MEK-CA + Dual CDKi), ns—not significant, P = 0.2590 (Control vs. MEK-CA + MEKi). m Representative single cell time courses of PPARG and APC/C cell cycle reporter from control and MEK-CA cells treated with Dual CDKi (Palbociclib—4 μM + CDK2i—1 μM) as described for l, showing variability in the timing of commitment to differentiation (yellow circle) after completion of mitosis (dashed line). n Distribution of the time taken after mitosis to undergo differentiation commitment from single differentiating cells for the experiment represented in (m). Horizontal line shows the median of each distribution. Mann–Whitney test used to compare medians of control and MEK-CA + CDKi conditions, ****P < 0.0001 (approximate), control—770 cells, MEK-CA + CDKi—655 cells. Representative of two biological replicates. Example traces in i and l from the same experimental dataset.