Molecular Sink vs. LLPS Models for Nuclear Compartments

Review scope — bounded HTML audit

Earlier text/reference appraisal: 3 October 2026. The scientific assertions in this current HTML overview, its metadata, and its comparison text were checked against the primary studies linked below. This is a bounded assessment of this page, with no systematic-search or literature-completeness claim. Associated downloadable documents, image provenance and licenses, and the operational literature-watch service remain outside this assessment. The overall review remains partial.

Binding and phase coexistence are separate questions

Here, a binding-site or “molecular sink” model means enrichment through reversible association with locally abundant binding sites. Liquid–liquid phase separation (LLPS) means demixing into coexisting liquid phases. Banani et al. (2016) measured selective binding of clients within phase-separated scaffolds, while Li et al. (2012) linked demixing with multivalent assembly. Binding and phase behavior are separate questions that can be coupled.

Experiments demonstrating selective phase-separated assembly

Banani et al. (2016) reconstituted three engineered scaffold pairs: SUMO/SIM, SH3/proline-rich motifs, and PTB/RNA. Fluorescent clients partitioned according to scaffold stoichiometry and client valency. In the SUMO/SIM system, binding-site mutants and GFP controls supported recruitment through cognate interactions. This directly supports selective client recruitment in the tested phase-separated systems.

Welles et al. (2024) examined model disordered proteins in cellular and purified-protein systems. The studied model disordered proteins formed spatially distinct condensates; sequence mutagenesis altered co-partitioning. The constructs and expression conditions matter: these experiments establish a possibility for selective assembly, with endogenous nuclear-body mechanisms requiring their own tests.

An endogenous compartment with a defined context

Erdel et al. (2020) assessed pericentric heterochromatin in mouse fibroblasts. Their tested chromocenters lacked a separated liquid HP1 pool and preferential internal mixing. The authors favored a collapsed-polymer interpretation for those foci.

This example shows why enrichment and nuclear compartmentalization need a component-specific physical explanation. It establishes neither a universal binding-site mechanism nor a universal absence of LLPS. The current check uses the primary article’s descriptive Summary, internal-mixing Results, and Discussion; detailed experimental procedures and datasets were not reassessed. Evidence and interpretation: Erdel et al. (2020).

Measurements that can distinguish models

Li et al. (2012) measured concentration-dependent demixing of reconstituted multivalent proteins and found that changing valency shifted the phase boundary. Their scattering and photobleaching results supported large dynamic assemblies and a likely coupled sol–gel transition. Network connectivity and demixing can thus be coupled; neither word alone identifies the complete mechanism.

An experimental comparison should state its predicted observables: client association for a binding model; phase compositions and a concentration series for a demixing model; relaxation and connectivity for a network model. These are proposed questions inferred from Banani et al. (2016) and Li et al. (2012). A sharp recruitment switch is insufficient by itself: Banani observed one within an already phase-separated mixture.

Evidence assessment

The cited studies support selective phase-separated model assemblies and a different physical interpretation of the tested mouse chromocenters. They do not assign the mechanism of every endogenous transcription hub. A conclusion for a particular body should identify the measured component, mixture, perturbation, and alternative model. Evidence and interpretation: Banani et al. (2016); Li et al. (2012).

Primary sources for this overview

1. Banani et al. (2016). Compositional Control of Phase-Separated Cellular Bodies. Cell 166, 651–663. DOI: 10.1016/j.cell.2016.06.010.
2. Welles et al. (2024). Determinants that enable disordered protein assembly into discrete condensed phases. Nature Chemistry 16, 1062–1072. DOI: 10.1038/s41557-023-01423-7.
3. Erdel et al. (2020). Mouse Heterochromatin Adopts Digital Compaction States without Showing Hallmarks of HP1-Driven Liquid-Liquid Phase Separation. Molecular Cell 78, 236–249.e7. DOI: 10.1016/j.molcel.2020.02.005.
4. Li et al. (2012). Phase transitions in the assembly of multivalent signalling proteins. Nature 483, 336–340. DOI: 10.1038/nature10879.