Two measured cluster populations in mouse stem cells
Cho and colleagues imaged endogenously tagged Mediator and RNA polymerase II in mouse embryonic stem cells. Super-resolution imaging resolved small transient clusters and larger clusters that persisted through the acquisition window. The latter often colocalized across Mediator and Pol II channels. During differentiation, large-cluster size and number decreased. “Stable” here means stable over the observation period, rather than a permanent transcription factory present in every cell type. Cho et al., 2018.
Transcriptional perturbations affect the components differently
In the same study, JQ1 treatment disrupted both Mediator and Pol II clusters. With the transcription-associated inhibitor DRB, stable Pol II clusters dissolved while stable Mediator clusters remained, and their colocalization recovered after washout. This supports distinct dependencies for the two assemblies. Because these interventions alter recruitment and transcriptional activity, the resulting cluster changes do not independently establish that a liquid phase is required for transcription. Cho et al., 2018.
Exchange, merging and selected-locus proximity
Light-sheet imaging documented cluster merging, and photobleaching experiments showed molecular exchange. Esrrb RNA labeling and Mediator tracking in six cells showed intermittent close proximity rather than continuous overlap. The authors proposed a dynamic interaction model. The optical proximity threshold does not demonstrate direct molecular contact, and FRAP recovery measures exchange rather than uniquely identifying a liquid-liquid phase transition. These observations support dynamic organization while leaving the microscopic mechanism open. Cho et al., 2018.
Activation domains provide complementary interaction evidence
Boija and colleagues found that purified MED1 disordered-region droplets enriched OCT4 relative to a GFP control. OCT4 activation-domain acidic-residue mutations reduced incorporation into droplets and reduced activation of an engineered GAL4 reporter in mouse stem cells. This links the same interaction chemistry to partitioning and activation. It does not separate phase behavior from ordinary coactivator interactions, and it does not demonstrate that every Pol II transcription event requires a condensate. Boija et al., 2018.
A current counterexample to equating clusters with condensates
Evidence update · 12 October 2026
In developing fly embryos, Mukherjee and colleagues find that accumulation of transcriptionally engaged polymerases explains the examined gene-associated clusters. Imaging and simulations support this particular account; it does not erase the different coactivator assemblies observed in mouse stem cells. A single cluster of RNA Polymerase II molecules is stably associated with active genes (2026).
A 2025 NDF–FACT study adds biochemical and stem-cell evidence for a specific elongation-associated condensate mechanism. Named components and transcriptional stages should remain explicit when the two systems are compared. Phase-separated NDF−FACT condensates facilitate transcription elongation on chromatin (2025).
Primary sources for this update
- Mukherjee et al. (2026). A single cluster of RNA Polymerase II molecules is stably associated with active genes. Primary study. DOI: 10.1038/s41467-026-70775-8.
- Phase-separated NDF−FACT condensates facilitate transcription elongation on chromatin (2025). Primary study. DOI: 10.1038/s41556-025-01778-8.
Source access: official Pol II main Results/Discussion; selected NDF–FACT primary Results. The study-specific findings above do not certify associated Word drafts, raw data or every proposed nuclear-body mechanism.
Primary studies and review scope
- Cho et al. (2018), Science 361:412–415. Mediator and RNA polymerase II clusters associate in transcription-dependent condensates. DOI: 10.1126/science.aar4199. Primary research.
- Boija et al. (2018), Cell 175:1842–1855.e16. Transcription factors activate genes through the phase separation capacity of their activation domains. DOI: 10.1016/j.cell.2018.10.042. Primary research.
The retained scientific text and the identity and fit of these references were checked on 3 October 2026. The discussion covers these experiments rather than an exhaustive literature search. Source-access limitations and outstanding work remain.