Extended Data Fig. 6: Hardened Klf4 droplets can incorporate new material and Klf4 droplets harden within 40 min.

a, In order to exclude the possibility that Klf4 droplets stopped growing on DNA because of hardening, we tested whether hardened droplets retain the ability to incorporate new material. Right, for this, Klf4-GFP droplets were formed and incubated until hardened (top row). After 50 min, a fresh solution of Klf4-mCherry was added (middle row). After only 10 min (bottom row), the green Klf4-GFP droplets enriched red signal of Klf4-mCherry. Left, as a control, to test whether both Klf4-GFP and Klf4-mCherry individually (top and middle row) or in combination (bottom row) are able to form droplets under these conditions, the standard incubation time for droplet assays was used (short incubation: 20 min). See Supplementary Methods for details. All confocal microscopy images of the same colour in the same panel have the same contrast settings. b, The droplet radius (left) and number of droplets in the field of view (right) of a solution with 20 µM Klf4-GFP was measured every minute after induction of phase separation for one hour. Many fusion events could be observed initially (white and yellow areas). However, after 43 min the droplets stopped fusing (brown area). This is one order of magnitude larger than the observation time in the optical tweezers assay (200 s). The onset of hardening was determined following a similar procedure than in Extended Data Fig. 2a (see Methods). c, Individual examples of droplet fusion events are depicted at different times after phase separation was induced, as indicated at the top. Note how the fusion time increases with time until fusion stops and droplets only stick to each other when hardened (“ seconds, ‘ minutes). In order to compare fusion times of droplets that are roughly of the same size, the examples are at different Klf4-GFP concentrations ranging from 5 to 20 µM.