Nuclear Condensates: Quality Control or Functional Regulators?

Review scope — partial audit

Targeted scientific correction checked 3 October 2026. This concise overview checks selected LLPS mechanisms, assay interpretation, and the primary studies linked below. It is not a systematic review or validation of every associated document, figure label, or historical citation. Downloadable source documents remain outside this correction.

Function and quality control

Enrichment of transcriptional machinery and quality control are distinct questions that may coexist. Functional evidence requires showing how an assembly changes a biological output; demonstrating LLPS requires physical evidence for demixing. One does not automatically establish the other. Evidence and interpretation: Boija et al. (2018); Cho et al. (2018); Itoh et al. (2021); Sahin et al. (2014).

The original attribution is corrected: Boija et al. (2018) was published in Cell, whereas Cho et al. (2018) appeared in Science. Neither study establishes that all super-enhancer transcription depends on a universal liquid compartment.

Transcription-linked evidence

Boija et al. (2018) reported that mutations of OCT4 and GCN4 activation-domain residues affected both in vitro co-condensation with Mediator and gene activation. This supports a connection; changing interaction residues can also change ordinary coactivator binding, limiting a phase-specific causal interpretation.

Cho et al. (2018) imaged endogenous Mediator and Pol II clusters in mouse embryonic stem cells, including their association with active genes and responses to JQ1 and transcription inhibition. The experiment supports regulated clustering in that context.

Perturbation limits

JQ1 disrupts BET bromodomain-dependent chromatin recruitment, so loss of clusters and transcription is not a selective test of LLPS. The relevant issue is the drug’s known biological target and its downstream effects, rather than a blanket claim of unspecified off-target activity. FRAP and morphology also need complementary mechanism tests. Evidence and interpretation: Cho et al. (2018).

Itoh et al. (2021) found that hexanediol treatment affects chromatin motion and condensation. Transcription changes after this treatment cannot be attributed exclusively to droplet dissolution.

Quality-control evidence has its own scope

PML-associated protein processing provides a concrete example: Sahin et al. (2014) linked stress-induced body assembly to partner SUMOylation and, for some partners, RNF4-associated ubiquitination and degradation. This does not establish that Mediator/BRD4 hubs exist mainly to sequester misfolded proteins.

An experiment separating these hypotheses would monitor nascent transcription and protein-quality-control outputs while controlling scaffold abundance, binding, localization, and cell stress. The proposed “condensates mainly as quality control” explanation remains a hypothesis for transcription hubs. Evidence and interpretation: Sahin et al. (2014).

Evidence assessment

Regulated assemblies are observed; general functional necessity and material mechanism remain separate tests. The evidence cited here supports links between selected interaction domains, clustering, and transcription. It does not justify either a universal LLPS requirement or a universal quality-control explanation. Evidence and interpretation: Boija et al. (2018); Cho et al. (2018); Itoh et al. (2021); Sahin et al. (2014).

Primary sources for this correction

1. Boija et al. (2018). Transcription Factors Activate Genes through the Phase-Separation Capacity of Their Activation Domains. Cell 175, 1842–1855.e16. DOI: 10.1016/j.cell.2018.10.042.
2. Cho et al. (2018). Mediator and RNA polymerase II clusters associate in transcription-dependent condensates. Science 361, 412–415. DOI: 10.1126/science.aar4199.
3. Itoh et al. (2021). 1,6-hexanediol rapidly immobilizes and condenses chromatin in living human cells. Life Science Alliance 4, e202001005. DOI: 10.26508/lsa.202001005.
4. Sahin et al. (2014). Oxidative stress-induced assembly of PML nuclear bodies controls sumoylation of partner proteins. Journal of Cell Biology 204, 931–945. DOI: 10.1083/jcb.201305148.