1. Separate a domain pattern from a functional unit
A topologically associating domain is identified as a region of preferential internal contacts in a contact map. Calling such a region a functional regulatory unit is an additional claim: its boundary should change the access or response of relevant enhancers and promoters when perturbed. A reproducible contact pattern alone cannot establish that every boundary is an absolute insulator or that every gene within a domain shares one regulatory program. [1] [2]
2. A tested boundary-dependent regulatory neighborhood
At the WNT6/IHH/EPHA4/PAX3 limb-development locus, Lupiáñez and colleagues linked patient structural variants with engineered mouse rearrangements. In mouse embryonic limbs, altered contact profiles accompanied ectopic expression of neighboring genes. Boundary-sparing deletions lacked the ectopic expression produced by corresponding boundary-disrupting deletions. These comparisons support regulatory insulation by the boundary region in this particular developmental setting. [1]
This selected locus is not a survey of all mammalian boundaries. Rearrangements affect sequence and enhancer neighborhood, so the comparison is not a geometry-only experiment. Human fibroblast contact measurements support altered architecture but are not direct measurements of expression in a developing human limb. This audit checked the primary abstract and figure legends, with full methods and supplements still unresolved. [1]
3. Architecture and short-term function can be partly separable
In mouse embryonic stem cells, Hsieh and colleagues combined three-hour architectural-protein depletion, Micro-C and RNA profiling. CTCF or RAD21 loss disrupted structural loops, yet most enhancer–promoter contacts and gene expression persisted over this interval. A weak overall enhancer–promoter contact reduction and sensitive subsets remained. This challenges a requirement for intact structural-loop architecture to continuously maintain every measured regulatory interaction. [2]
It does not show that domains never matter. Maintenance of an established state differs from initial establishment, developmental responses or chronic perturbation. The authors also distinguish a contact-map dot from a proven loop-shaped conformation. Functional necessity should therefore be stated with the cell state, time window and readout used, rather than attached automatically to the TAD label. [2]
4. A domain need not be a permanently closed loop
Live tracking of one 505-kb Fbn2 domain in mouse embryonic stem cells found changing anchor distances and infrequent, transient fully boundary-bridged states inferred by a Bayesian model. That result does not make domain organization a population-averaging artifact: factor perturbations and Micro-C controls supported a real architectural contribution. It instead cautions against equating an ensemble domain pattern with one permanent loop in every cell. Fbn2 was silent, so this study supplies structural dynamics rather than direct regulatory necessity. [3]
5. Assessment: conditional regulatory insulation
The defensible conclusion is that some boundary regions constrain regulation in tested contexts, while contact-domain structure is not a universal, indivisible unit of gene control. The studies here do not measure what fraction of all boundaries are required, prove that active transcription creates technical boundary artifacts, or establish that all inter-domain regulation is blocked. Those stronger claims require their own primary evidence and are excluded from this edition. [1] [2] [3]
Contact domains and enhancer-driven organization remain distinct — 12 October 2026
In zebrafish embryos, Galitsyna and colleagues reported enhancer-associated Hi-C “fountains” during zygotic genome activation. Removing pioneer transcription factors reduced these patterns, linking an enhancer-activation perturbation to architectural change in that setting. A proposed mechanism involving facilitated cohesin loading and desynchronized extrusion came from polymer simulations. This adds evidence for regulatory activity influencing folding, while preserving the distinction between the measured contact pattern and its mechanistic model. Galitsyna et al. (2026).
Enhancer-associated fountains and conventional TAD boundaries are different contact patterns. The new developmental result therefore adds a context in which regulatory activity affects organization, rather than demonstrating that every TAD is an indivisible regulatory unit. Galitsyna et al. (2026).
This addition uses primary studies. Access depth and claim limits are recorded with each added reference; it is a dated literature appraisal, not independent experimental replication.
Primary studies and review scope
- Lupiáñez et al. (2015), Cell 161:1012–1025. Disruptions of topological chromatin domains cause pathogenic rewiring of gene-enhancer interactions. DOI: 10.1016/j.cell.2015.04.004. Primary research.
- Hsieh et al. (2022), Nature Genetics 54:1919–1932. Enhancer–promoter interactions and transcription are largely maintained upon acute loss of CTCF, cohesin, WAPL or YY1. DOI: 10.1038/s41588-022-01223-8. Primary research.
- Gabriele et al. (2022), Science 376:496–501. Dynamics of CTCF- and cohesin-mediated chromatin looping revealed by live-cell imaging. DOI: 10.1126/science.abn6583. Primary research.
- Galitsyna AA, Ulianov SV, et al. (2026). Extrusion fountains are hallmarks of chromosome organization emerging upon zygotic genome activation. Nature Communications 17. DOI: 10.1038/s41467-026-69105-9. Primary research.
Independent update access: Consensus fetched primary abstract and Scite primary abstract inspected; full Results and supplements not independently appraised. Primary source.
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.