Naumova et al. (Science, 2013) showed that metaphase chromosomes fold as linearly organized arrays of consecutive ~80-120 kb chromatin loops and asked what mechanism builds these arrays in vivo, whether a central protein scaffold exists, whether loop bases are specific sequence elements, what pathway connects interphase and mitotic folding, and how much cell-to-cell variability population Hi-C masks. Twelve years of work has answered most of this cluster: condensin-mediated loop extrusion is now established in vitro and in vivo, mitotic loop bases are variably positioned rather than sequence-determined, single-cell DNA tracing has exposed the cell-to-cell variability directly, and the folding pathway through mitotic entry and exit is mapped in outline. What survives is a precise, explicitly documented controversy between the two most direct 2025 studies: Hi-C-based perturbation analysis in chicken DT40 cells supports consecutive, non-overlapping ~400 kb condensin II loops whose extruding condensins stall upon meeting each other, organized as a disordered discontinuous helical scaffold, whereas nanoscale DNA tracing in single human cells supports strongly overlapping megabase-scale loops whose condensins pass one another, with rod-shaped architecture emerging from extrusion plus chromatin self-repulsion and no helical order. The open problem is to determine which loop architecture the vertebrate mitotic chromosome actually adopts in vivo, and the encounter and extrusion rules of condensins on chromatin in vivo that generate it.

Background

During cell division (mitosis), each eukaryotic chromosome must be refolded from a transcriptionally active, loosely compacted interphase state into a compact, rod-shaped pair of sister chromatids that microtubule-driven forces can segregate without entangling the DNA. How this rod is built at the molecular scale has been debated for over a century, between 'loops-on-a-scaffold' models (DNA loops emanating from a central proteinaceous axis), hierarchical coiling models, and network or melt models. Chromosome conformation capture methods made the question quantitative: Hi-C (genome-wide chromosome conformation capture) is a sequencing assay that estimates the frequency of physical contact between every pair of genomic loci in a cell population, from which the average 3D folding of chromosomes can be inferred. Naumova, Imakaev, Fudenberg, Zhan, Lajoie, Mirny and Dekker (2013) applied Hi-C and 5C (a targeted variant) across the human cell cycle and found that metaphase chromosomes adopt a homogeneous, locus- and cell-type-independent folding state that is inconsistent with classic hierarchical models and best described by polymer models of consecutive, non-overlapping chromatin loops of roughly 80-120 kilobases (kb), linearly ordered along the chromosome axis and axially compressed. They proposed a two-stage process - loop formation followed by axial compression - and noted that the loops could be generated by SMC (structural maintenance of chromosomes) protein complexes extruding loops, an idea proposed theoretically by Alipour and Marko (2012): a loop-extruding motor anchors at one DNA site and reels in DNA from one side, progressively enlarging a loop. The authors explicitly flagged as unresolved: (1) the in vivo mechanism of the loop arrays; (2) the existence, composition and structure of any central scaffold, since models with a compact, diffuse, or no scaffold all fit their Hi-C data equally well; (3) whether specific sequence elements serve as loop bases, since population-averaged data cannot exclude different cells using different anchor subsets; (4) the folding pathway through prophase, prometaphase and telophase; and (5) how much single-cell variability the population averages conceal. The prime mechanistic suspects are the two condensin complexes, ring-shaped SMC-family ATPases (condensin II, which acts in the nucleus from prophase onward, and condensin I, which accesses chromosomes only after nuclear envelope breakdown); the related cohesin complex extrudes loops in interphase, where loops are anchored at CTCF (a sequence-specific DNA-binding protein) boundary sites and collectively produce topologically associating domains (TADs), the contact domains visible in interphase Hi-C maps. Major subsequent results: Ganji et al. (2018) directly visualized single yeast condensin complexes extruding DNA loops in vitro at up to 1500 base pairs per second, and Kim et al. (2020) showed in vitro that two condensins can traverse one another, forming Z-loop structures (two extruded loops that share a segment of DNA, so that the combined path is Z-shaped); Walther et al. (2018) quantified condensin abundance and positioning in human cells, finding ~35,000 condensin II complexes stably confined to the chromatid axis and ~195,000 condensin I complexes dynamically bound more peripherally, organized as ~450 kb outer loops refined to ~90 kb inner loops; Gibcus et al. (2018) used time-resolved Hi-C in synchronized chicken DT40 cells to map mitotic entry, showing condensin II forming arrays of ~60 kb loops in prophase and condensin I nesting ~80 kb loops within ~400 kb outer loops in prometaphase, arranged along a helical 'spiral staircase' condensin scaffold whose turns grow to ~12 Mb; Hildebrand et al. (2024) showed by Hi-C and polymer modeling that mitotic chromosomes are self-entangled and disentangle through a topoisomerase II (an enzyme that passes one DNA duplex through another to resolve tangles)-dependent two-stage exit during anaphase, telophase and early G1; and Schooley et al. (2026) mapped the folding programmes that re-establish interphase organization during mitotic exit. Reviews by Dekker and Mirny (2024) and Hirano (2025) summarize the consensus that loop extrusion by condensins, together with topo II and histone-driven compaction, is the core reaction of mitotic genome folding. In 2025 the field's two most direct studies converged on condensin-driven loop extrusion but inferred incompatible architectures, and a commentary by Vitoria Gomes and Haering (2025) explicitly identified the disagreement as the open question, as detailed in Problem Statement.

Problem Statement

Determine the actual in vivo loop architecture of the vertebrate mitotic chromosome and the condensin encounter and extrusion rules on chromatin in vivo that generate it, resolving the conflict between the two incompatible 2025 models. Specifically: (a) when two condensin II complexes extruding loops along the same chromatid meet on chromatin in vivo, do they stall, forming consecutive non-overlapping loops of a few hundred kilobases as inferred from Hi-C perturbation data and polymer modeling (Samejima et al., 2025), or do they bypass or traverse one another, forming strongly overlapping loops of several megabases as inferred from single-cell nanoscale DNA tracing (Beckwith et al., 2025), noting that in vitro single-molecule assays on bare DNA show condensins capable of traversing one another? (b) Are the loop bases arranged in a (disordered) helical order along the chromatid axis, or is the ordering aperiodic, with the rod-shaped chromosome and the axial enrichment of condensin II emerging from loop extrusion combined with chromatin self-repulsion without any helical scaffold? (c) What in vivo extrusion rates, processivities, densities, and encounter outcomes correspond to each of the competing architectures, and do the answers generalize beyond the two systems studied (chicken DT40 and human HeLa cells)? Direct observation of individual condensin-condensin encounters on chromatinized substrate or in intact chromosomes, and structural tracing that resolves loop-base ordering across full loop lengths at near-native preservation, are among the routes the field's commentary suggests could help discriminate between the competing architectures. A complete answer addresses all three parts (a)-(c): each part may be resolved either by an evidence-based adjudication between the two competing architectures or by an experimentally decisive test that would settle the conflict, and, where direct in vivo measurement is unavailable, established ranges, parameter estimates or model predictions from the cited literature are acceptable in place of measured values.

References

  1. Naumova N, Imakaev M, Fudenberg G, Zhan Y, Lajoie BR, Mirny LA, Dekker J. Organization of the mitotic chromosome. Science 342:948-953 (2013). DOI 10.1126/science.1236083. https://doi.org/10.1126/science.1236083
  2. Alipour E, Marko JF. Self-organization of domain structures by DNA-loop-extruding enzymes. Nucleic Acids Research 40:11202-11212 (2012). DOI 10.1093/nar/gks925. https://doi.org/10.1093/nar/gks925
  3. Ganji M, Shaltiel IA, Bisht S, Kim E, Kalichava A, Haering CH, Dekker C. Real-time imaging of DNA loop extrusion by condensin. Science 360:102-105 (2018). DOI 10.1126/science.aar7831. https://doi.org/10.1126/science.aar7831
  4. Kim E, Kerssemakers J, Shaltiel IA, Haering CH, Dekker C. DNA-loop extruding condensin complexes can traverse one another. Nature 579:438-442 (2020). DOI 10.1038/s41586-020-2067-5. https://doi.org/10.1038/s41586-020-2067-5
  5. Walther N, Hossain MJ, Politi AZ, Koch B, Kueblbeck M, Odegaard-Fougner O, Lampe M, Ellenberg J. A quantitative map of human Condensins provides new insights into mitotic chromosome architecture. Journal of Cell Biology 217:2309-2328 (2018). DOI 10.1083/jcb.201801048. https://doi.org/10.1083/jcb.201801048
  6. Gibcus JH, Samejima K, Goloborodko A, Samejima I, Naumova N, Nuebler J, Kanemaki MT, Xie L, Paulson JR, Earnshaw WC, Mirny LA, Dekker J. A pathway for mitotic chromosome formation. Science 359:eaao6135 (2018). DOI 10.1126/science.aao6135. https://doi.org/10.1126/science.aao6135
  7. Samejima K, Gibcus JH, Abraham S, Cisneros-Soberanis F, Samejima I, Beckett AJ, Pucekova N, Abad MA, Spanos C, Medina-Pritchard B, Paulson JR, Xie L, Jeyaprakash AA, Prior IA, Mirny LA, Dekker J, Goloborodko A, Earnshaw WC. Rules of engagement for condensins and cohesins guide mitotic chromosome formation. Science 388:eadq1709 (2025). DOI 10.1126/science.adq1709. https://doi.org/10.1126/science.adq1709
  8. Beckwith KS, Brunner A, Morero NR, Jungmann R, Ellenberg J. Nanoscale DNA tracing reveals the self-organization mechanism of mitotic chromosomes. Cell 188:2656-2669.e17 (2025). DOI 10.1016/j.cell.2025.02.028. https://doi.org/10.1016/j.cell.2025.02.028
  9. Vitoria Gomes M, Haering CH. The fine art of chromatin folding: Revealing the path of DNA inside mitotic chromosomes. Cell Genomics 5:100921 (2025). DOI 10.1016/j.xgen.2025.100921. https://doi.org/10.1016/j.xgen.2025.100921
  10. Hildebrand EM, Polovnikov K, Dekker B, Liu Y, Lafontaine DLJ, Fox AN, Li Y, Venev SV, Mirny LA, Dekker J. Mitotic chromosomes are self-entangled and disentangle through a topoisomerase-II-dependent two-stage exit from mitosis. Molecular Cell 84:1422-1441.e14 (2024). DOI 10.1016/j.molcel.2024.02.025. https://doi.org/10.1016/j.molcel.2024.02.025
  11. Schooley A, Venev SV, Aksenova V, Lehman JW, Navarrete E, Pai AA, Dasso M, Dekker J. Interphase chromosome conformation is specified by distinct folding programmes inherited through mitotic chromosomes or the cytoplasm. Nature Cell Biology 28:82-97 (2026). DOI 10.1038/s41556-025-01828-1. https://doi.org/10.1038/s41556-025-01828-1
  12. Zou F, Li Y, Foldes T, Pinholt HD, Smith C, Mirny L, Bai L. Condensin accelerates long-range intra-chromosomal interactions. Nature Communications 17:4020 (2026). DOI 10.1038/s41467-026-70538-5. https://doi.org/10.1038/s41467-026-70538-5
  13. Piveteau V, Dupont C, Salari H, Dumont A, Savocco J, Jost D, Piazza A. Condensin loop extrusion properties, roadblocks, and role in homology search during recombination in S. cerevisiae. The EMBO Journal 45:3124-3155 (2026). DOI 10.1038/s44318-026-00748-6. https://doi.org/10.1038/s44318-026-00748-6
  14. Hirano T. Mitotic genome folding. Journal of Cell Biology 224:e202504075 (2025). DOI 10.1083/jcb.202504075. https://doi.org/10.1083/jcb.202504075
  15. Dekker J, Mirny LA. The chromosome folding problem and how cells solve it. Cell 187:6424-6450 (2024). DOI 10.1016/j.cell.2024.10.026. https://doi.org/10.1016/j.cell.2024.10.026