Extended Data Fig. 5: Live cell single-molecule imaging experiments to study the chromatin-bound NuRD complex.

(a) Distribution of the four biophysical parameters from 500 ms exposure tracking of: (i) chromatin bound CHD4 in wild-type ES cells, in the absence of MBD3, and in the presence of DRB (an inhibitor of transcriptional elongation); (ii) chromatin bound MBD3 in wild-type ES cells, and in the presence of the HDAC1/2-specific inhibitor FK228; and (iii) JF549 dye bound to the coverslip. The grey dotted lines indicate the upper bound (at the 95 % confidence interval) of the different biophysical parameters determined for stationary JF₅₄₉ dye molecules. (b) Table showing the proportions (K) and estimated values of the apparent diffusion coefficients (D), as well as the number of slow (S) and fast (F1+F2) diffusing sub-trajectories obtained from the total number of trajectories analysed. (NB – many trajectories were discarded as they were either too short for analysis or because they had a low probability of being classified as slow or fast.) (c) Table summarising the changes in anomalous exponent of the slow and fast chromatin bound NuRD complex subunits in the presence and absence of MBD3, or in the presence of specific inhibitors. Errors given are for 95 % confidence intervals. (d) (Left) Fitting of 1, 2 or 3 Gaussians to the anomalous exponent distributions for fast moving chromatin bound CHD4 in wild-type ES cells – the R² values indicate the goodness of fit. (Right) The Bayesian information criterion (BIC) was calculated for all the datasets shown in (a) and shows that two populations (Gaussians) are the best model to account for the data – that is that model has the lowest BIC value (light blue box). (e) Table showing the proportions of trajectories containing either slow (S), fast (F1), or fast (F2) chromatin bound sub-trajectories (or combinations thereof).