The 3D genomics of lampbrush chromosomes highlights the role of active transcription in chromatin organization.

Skip to main content An official website of the United States government Here's how you know Here's how you know Official websites use .gov A .gov website belongs to an official government organization in the United States. Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites. Search Log in Dashboard Publications Account settings Log out Search… Search NCBI Primary site navigation Search Logged in as: Dashboard Publications Account settings Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide PERMALINK Copy As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Nucleic Acids Res . 2026 Apr 13;54(7):gkag316. doi: 10.1093/nar/gkag316 Search in PMC Search in PubMed View in NLM Catalog Add to search The 3D genomics of lampbrush chromosomes highlights the role of active transcription in chromatin organization Timofey Lagunov Timofey Lagunov 1 Institute of Cytology and Genetics, Novosibirsk, Russia 2 Novosibirsk State University, Novosibirsk, Russia Find articles by Timofey Lagunov 1, 2, c , Maria Gridina Maria Gridina 3 Institute of Cytology and Genetics, Novosibirsk, Russia 4 Novosibirsk State University, Novosibirsk, Russia 5 Sirius University of Science and Technology, Sirius, Russia Find articles by Maria Gridina 3, 4, 5, c , Artem Nurislamov Artem Nurislamov 6 Institute of Cytology and Genetics, Novosibirsk, Russia 7 Novosibirsk State University, Novosibirsk, Russia 8 Sirius University of Science and Technology, Sirius, Russia Find articles by Artem Nurislamov 6, 7, 8 , Tatiana Kulikova Tatiana Kulikova 9 Laboratory of Cell Nucleus Structure and Dynamics, Saint-Petersburg State University, Saint-Petersburg, Russia Find articles by Tatiana Kulikova 9 , Antonina Maslova Antonina Maslova 10 Laboratory of Cell Nucleus Structure and Dynamics, Saint-Petersburg State University, Saint-Petersburg, Russia Find articles by Antonina Maslova 10 , Viktor Konstantinov Viktor Konstantinov 11 Institute of Cytology and Genetics, Novosibirsk, Russia 12 Novosibirsk State University, Novosibirsk, Russia Find articles by Viktor Konstantinov 11, 12 , Andrey Popov Andrey Popov 13 Institute of Cytology and Genetics, Novosibirsk, Russia 14 Sirius University of Science and Technology, Sirius, Russia 15 Artificial Intelligence Research Institute, Moscow, Russia Find articles by Andrey Popov 13, 14, 15 , Alla Krasikova Alla Krasikova 16 Laboratory of Cell Nucleus Structure and Dynamics, Saint-Petersburg State University, Saint-Petersburg, Russia Find articles by Alla Krasikova 16, ✉ , Veniamin Fishman Veniamin Fishman 17 Institute of Cytology and Genetics, Novosibirsk, Russia 18 Novosibirsk State University, Novosibirsk, Russia 19 Sirius University of Science and Technology, Sirius, Russia 20 Artificial Intelligence Research Institute, Moscow, Russia Find articles by Veniamin Fishman 17, 18, 19, 20, ✉ Author information Article notes Copyright and License information 1 Institute of Cytology and Genetics, Novosibirsk, Russia 2 Novosibirsk State University, Novosibirsk, Russia 3 Institute of Cytology and Genetics, Novosibirsk, Russia 4 Novosibirsk State University, Novosibirsk, Russia 5 Sirius University of Science and Technology, Sirius, Russia 6 Institute of Cytology and Genetics, Novosibirsk, Russia 7 Novosibirsk State University, Novosibirsk, Russia 8 Sirius University of Science and Technology, Sirius, Russia 9 Laboratory of Cell Nucleus Structure and Dynamics, Saint-Petersburg State University, Saint-Petersburg, Russia 10 Laboratory of Cell Nucleus Structure and Dynamics, Saint-Petersburg State University, Saint-Petersburg, Russia 11 Institute of Cytology and Genetics, Novosibirsk, Russia 12 Novosibirsk State University, Novosibirsk, Russia 13 Institute of Cytology and Genetics, Novosibirsk, Russia 14 Sirius University of Science and Technology, Sirius, Russia 15 Artificial Intelligence Research Institute, Moscow, Russia 16 Laboratory of Cell Nucleus Structure and Dynamics, Saint-Petersburg State University, Saint-Petersburg, Russia 17 Institute of Cytology and Genetics, Novosibirsk, Russia 18 Novosibirsk State University, Novosibirsk, Russia 19 Sirius University of Science and Technology, Sirius, Russia 20 Artificial Intelligence Research Institute, Moscow, Russia ✉ To whom correspondence should be addressed. Email: [email protected] ✉ Correspondence may also be addressed to Alla Krasikova. Email: [email protected] c Timofey Lagunov and Maria Gridina Equal contribution Received 2025 Jul 28; Revised 2026 Jan 29; Accepted 2026 Mar 10; Collection date 2026 Apr 24. © The Author(s) 2026. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution License ( https://creativecommons.org/licenses/by/4.0/ ), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. PMC Copyright notice PMCID: PMC13076225  PMID: 41978268 Abstract Lampbrush chromosomes (LBCs) are giant meiotic bivalents that have served as a classic model system for studying chromatin organization and RNA synthesis for over a century. Despite their importance, the molecular mechanisms underlying distinctive LBC chromomere–loop architecture have remained poorly understood. Moreover, the influence of hypertranscription on chromatin organization during oogenesis remains enigmatic. Here, we provide comprehensive analysis of LBC organization by integrating single-cell Hi-C, RNA-seq, NOMe-seq, FISH mapping, and chromatin simulations. Single-nucleus Hi-C revealed CTCF-independent contact domains with stable boundaries defined by convergently oriented transcription units (TUs). Contact domains identified through Hi-C analysis correspond to insulated chromomeres in LBCs. Small transcriptionally inactive contact domains surrounded by divergently oriented TUs form “chromatin knots,” which are often detached from the chromosome axis. Transcription loops frequently manifest as a “cross” pattern with reduced contacts within chromatin domains. Integrative analysis of the whole-genome data uncovers the mechanisms underlying LBC structure, revealing how hypertranscription modulates chromatin stiffness and repositions SMC complexes to establish the distinctive chromomere–loop organization. Biophysical modeling through polymer simulation reproduces key features of LBCs, including transcription loop formation, chromomere compaction, and insulation patterns. These findings offer a unifying framework for understanding remarkable transcription-dependent organization of LBCs. Graphical Abstract Graphical Abstract. Open in a new tab Introduction Chromosomes are folded nonrandomly in the nuclear space. Their spatial organization is essential for several genomic processes [ 1 , 2 , 3 ], such as regulation of gene expression [ 4 – 7 ], DNA replication [ 8 – 10 ] and repair [ 11 ], and mechanical properties of the nucleus [ 12 , 13 ]. Consistent with this pivotal role in genome regulation, spatial organization of chromatin was shown to be conserved across large evolutionary distances of hundreds of million years [ 14 – 17 ]. There are several essential mechanisms responsible for chromatin organization in animal cells. First, DNA is a polymer organized in the nuclear space according to the principles of polymer physics. This implies that frequency of contacts between loci decreases with increase of genomic distance, following power law. For the majority of the studied cell types, scaling of contacts with genomic distance is concordant with the “crumpled” states of the chromatin [ 18 ], characterized by largely unknotted conformations with a fractal dimension of ∼3 [ 19 ]. Second, multivalent homotypic interactions between proteins cause segregation of the chromatin into compartments with distinct epigenetic states [ 20 – 23 , 24 ]. Moreover, active and inactive chromatin is directed toward different nuclear landmarks [ 2 ]. Inactive chromatin is often associated with nuclear lamina and nucleolar periphery, while active chromatin is in proximity to the interchromatin space comprising nuclear speckles [ 25 , 26 , 27 ]. Third, ATP-dependent extrusion of DNA by SMC-complexes cohesin and condensin results in chromosome compaction and increased probability of long-range interactions [ 1 , 28 , 29 ]. The extrusion process can be modulated by several factors. In the majority of vertebrate cell types, CTCF protein blocks cohesin-mediated extrusion; therefore, CTCF-binding sites demarcate compact self-interacting genomic regions, often referred to as topologically associating domains (TADs) [ 1 , 30 – 32 , 33 ]. Interactions of cohesin with other proteins, including replication-specific minichromosome maintenance (MCM) [ 34 ] or transcription-associated RNA–polymerase II complexes [ 35 – 37 , 38 ] also may modulate extrusion, although details of these mechanisms are less studied. Moreover, interaction of cohesin with MCM and RNA–polymerase have less impact on genomic contacts compared to cohesin–CTCF interactions, and their effect becomes pronounced only when artificially modulated [ 37 ] or in analysis of specific cell cycle stages [ 34 ]. Although chromatin contacts in the majority of vertebrate cells can be explained by above mentioned mechanisms with relatively high accuracy [ 39 , 40 ], specific cell types or cell cycle stages show distinct arrangement of chromatin [ 2 ]. This includes early stages of development, prior to or immediately after zygotic genome activation, and germ cells [ 41 ]. Importantly, chromatin architecture is uniquely reorganized during the oocyte-to-zygote transition. In mammals, features of genomic organization at diplotene stage include compartments, TADs, and loops [ 41 , 42 ]. Although these chromatin domains can be detected when averaged over the genome, the presence of each type of domain at a locus varies between cells [ 42 ]. TAD boundaries in murine oocytes are located concordant with somatic cells, demarcated by CTCF, and associated with cohesin-mediated loop extrusion [ 42 ]. Compartments in murine diplotene oocytes manifest as plaid-pattern of Hi-C contacts and demonstrate strong association with distribution of H3K27me3 mark [ 41 ]. Compared to somatic cells, in oocytes interactions within compartments are restricted to the regions located <10 Mb away from each other, presenting a “local compartmentalization” pattern [ 41 ]. Similar depletion of long-range interactions and local compartmentalization was also observed during meiotic division of sperm cells [ 43 , 44 ]. Strength of both TADs and compartments gradually decreases with oocyte transition from early to late diplotene stage, and these structures disappear at metaphase-II (MII) stage [ 41 , 42 ]. Previous whole-genomic studies focused on chromatin organization in oocytes did not include representatives of vertebrate taxa in which chromatin undergoes lampbrush chromosome (LBC) state. LBCs represent a specific form of nuclear chromatin that emerges at the diplotene stage of oogenesis in the majority of vertebrate species, including amphibians, fish, reptiles, birds, monotremes, and certain insects. In the 3D space of the giant oocyte nucleus, LBCs do not contact each other, are not tethered to the nuclear periphery, and do not have a preferred radial position [ 45 , 46 ]. LBCs consist of separate chromatin globules connected by a thin thread and protruding lateral loops, together forming chromomere–loop complexes [ 47 , 48 , 49 ]. The unusually enormous size of LBCs allows for chromomere–loop structure visualization using conventional light microscopy, making diplotene oocytes a preferred model for microscopy-based studies of chromatin architecture. The phenomenon of LBC organization is attributed to hypertranscriptional activity of the oocyte nucleus [ 50 , 51 ]. Hypertranscription results in formation of lateral loops tightly covered by elongating RNA polymerases and nascent transcripts [ 49 , 52 , 53 ] and containing single or multiple genes essential for maternal products biosynthesis [ 54 – 56 , 57 ]. Another structural unit of LBC are lampbrush chromomeres—compact insulated chromatin globules present along LBC axes [ 48 ]. Chromomeres do not clearly correspond to any of the known spatial chromatin domains of the interphase nucleus. Genomic loci of neighboring somatic TADs can be detected as a part of a single chromomere, while genomic regions belonging to one somatic TAD can be detected as a part of two adjacent chromomeres [ 55 , 58 ]. There is, however, partial correspondence between gene-poor heterochromatic regions of constitutive B-compartments identified in somatic cells and clusters of large DAPI-positive chromomeres, which are also enriched for inactive chromatin marks [ 58 – 61 , 62 ]. Overall, chromatin organization in LBCs drastically differs from somatic cells. Hypertranscription being a hallmark of LBCs is also a widespread phenomenon common in the other cell types, including developmental hypertranscription and hypertranscription in adult stem/progenitor cells [ 63 , 64 ]. Our work aims to investigate the influence of hypertranscription on chromatin organization using LBCs as a model system. Following this aim, we present the first Hi-C analysis of LBCs, supplemented by transcriptomic and epigenomic profiling. We show that chromatin contacts in LBCs display unique features, including restriction of long-range and inter-chromosomal interactions and formation of CTCF-independent chromatin domains and loops. We show strong correlation between location and orientation of transcription units (TUs) and formation of chromatin domain boundaries. Based on our observations, validated by FISH and other complementary analysis, we develop a physical model of LBCs explaining LBC architecture via interplay between SMC and RNA–polymerase complexes. We show that the polymer model can qualitatively explain observed Hi-C patterns genome-wide and accurately reproduce spatial configuration of individual loci. Finally, the developed model highlights the importance of the transcription-mediated ribonucleoprotein matrix accumulation, nucleosome removal, and cohesive cohesin repositioning as factors determining spatial structure of LBC. Materials and methods Oocytes isolation Chicken oocytes, with diameters ranging from 1 to 2 mm as LBC and from 4 to 7 mm as post-LBC stages, were dissected from the ovary of healthy chicken and placed in individual drops of cooled “5:1” medium (83 mM KCl, 17 mM NaCl, 6.5 mM Na 2 HPO 4 , 3.5 mM KH 2 PO 4 , 1 mM MgCl 2 , and 1 mM Ditiotreitol (DTT), pH 7.2). All animal experiments were approved by the bioethics committee of the Institute of Cytology and Genetics SB RAS (Protocol no.: 66, 9 October 2020). International guidelines were followed during experimental procedures (Guide for the Care and Use of Laboratory Animals [ 65 ]). Nuclei were isolated as described earlier [ 45 ], washed with “5:1” medium, and transferred to a dish containing 2% formaldehyde in PBS for fixation at room temperature for 15 min. The formaldehyde was quenched by adding glycine to a final concentration of 0.125 M for 15 min at room temperature. The integrity of the nuclei was assessed using an Olympus stereomicroscope, and the Hi-C protocol was initiated immediately. Hi-C libraries preparation Individual fixed nuclei were transferred to microplate wells containing 9 μl of ice-cold lysis buffer (10 mM Tris–HCl pH 8.0, 10 mM NaCl, and 0.2% Igepal) for 20 min. The nuclei were then washed with NEB3.1 + 0.5% SDS and transferred to 0.2 ml of PCR tubes with 3 μl of NEB3.1 + 0.5% SDS. Ten microliters of freshly prepared 1.2% low melting point agarose were added to each tube under a stereomicroscope. After solidification, 10 μl of NEB3.1 + 0.6% SDS was added, and the samples were incubated at 37°C for 1 h. SDS was quenched using 10 μl of 6% Triton X-100, and chromatin fragmentation was performed with 25 U of Dpn II at 37°C overnight. The digested chromatin ends were labeled using 5 U of the Klenow fragment in the presence of biotin-15-dCTP at 22°C for 4 h, followed by overnight ligation at 16°C. Crosslinks were reversed by incubation at 65°C overnight, and the Low Melting Point Agarose was digested with 0.4 U Agarase for 1 h at 42°C. The purified DNA was fragmented with Alu I for 1 h at 37°C, and next-generation sequencing (NGS) libraries were prepared using the KAPA HyperPrep kit with modifications. The volumes for the end repair, adapter ligation, and PCR reactions were reduced by 5-fold, 5-fold, and 2-fold, respectively, and adapters were diluted to 300 nM. A biotin pull-down was performed with Dynabeads MyOne Streptavidin C1 beads after the adapter ligation step, followed by 25 cycles of PCR. Genomic libraries were sequenced using BGI DNBseq G400 machine in 2 × 150 paired-end mode. Hi-C data analysis and quality control The illumina adapters from raw sequences were cut with cutadapt ver-4.1 [ 66 ]. Data were aligned with using juicer ver-1.5.6 [ 67 ] script against chicken telomere-to-telomere genome assembly (ASM2420605v2 [ 68 ]). The read pairs without duplicates were filtered by mapping quality ≥ 30. Quality statistics were computed using juicer output with slight modifications in cis/trans ratio estimation described previously [ 69 ]. We performed deep sequencing only for samples with duplicates percentage below 30% (estimated in test run with less than 200k read pairs sequenced) and fraction of intra-fragment reads percentage 20% (see Supplementary Note S1 and Supplementary Figs S14 – S16 for additional details). We used juicer output to generate mcool files with cooler ver-0.10.2 [ 70 ]. Due to the low number of contacts in the dataset (∼500 000 per cell), we didn’t use the cooler balance function. Instead, we set weights for bins with zero coverage as “nan” and weights of bins with non-zero coverage as 1. Aggregated plots were made with coolpuppy ver-1.1.0 [ 71 ] with “maxshift” of 2 MB and “local” flag. The P (s) curves were calculated using cooltools [ 72 ] Gaussian smooth sigma value equal to 0.1. The raw reads from [ 42 ] were aligned against GRCm39 genome assembly and processed the same way (except cooler balance was used for mcool maps). The raw reads from [ 15 ] were aligned against ASM2420605v2 genome assembly and processed the same way (except cooler balance was used for mcool maps). Domain boundaries annotation To annotate domain boundaries for both chicken oocytes, mouse oocytes and chicken somatic cells the lavaburst ver-0.2.0 was used ( https://github.com/nezar-compbio/lavaburst ). This algorithm divides the genome into domains in such a way as to maximize a global domain scoring function. The algorithm depends on gamma hyperparameter which should be found empirically. We found the value of gamma separately for each single oocyte the same way as in the [ 42 ]: (i) 40 values of gamma in range from 6 to 24 were scanned; (ii) the value of gamma was multiplied by chromosome size and divided on (3250 * resolution); (iii) the average domain size was calculated for each gamma (taking into account only domains >200 kb); (iv) the gamma value that makes the average domain size closest to 500 kb for mouse NSN oocytes was used. For chicken oocytes the preferable average domain size was 1 Mb (motivated by the visual assessment of Hi-C maps) and the gamma range was from 0.5 to 10.5. Both analyses (for mouse and chicken oocytes) were performed on resolution 20 kb. To assess the stability of domain-boundary positions at single-cell resolution, we computed an F1 score for boundary concordance. Boundaries on the same chromosome were counted as true positives (TP) if their coordinates differed by ≤80 kb; boundaries without a counterpart within this window contributed to false positives (FP) and false negatives (FN). The F1 score was calculated as F1 = 2*TP/(2*TP + FP + FN). For cross-species comparison (chicken vs mouse), we normalized the F1 by a permutation baseline: for each single oocyte, domain intervals were randomly re-positioned along each chromosome while preserving the total length of inter-domain gaps, and 10 such permutations were generated per oocyte. For each oocyte pair, the permutation baseline was defined as the mean F1 across the corresponding permutation replicates. The normalized pairwise F1 was then obtained by dividing the observed pairwise F1 by this baseline. The domains were divided into LUDs and SSDs by N50 threshold: all domains were sorted by their size and the domain size that covers the half of the summed domain length was set as threshold. Hi-C map insulation score track Insulation function from cooltools package ver-0.5.1 [ 72 ] has been used for calculating the Insulation scores on merged oocytes. Resolution size is 16 kb and window size is 2 Mb. Cross pattern score track “Cross pattern score” (CPS) for genome position n was calculated using the equation: Where s = 35 kb, W = 4 Mb, and Cij is the number of contacts between loci i and j . The resolution of the calculations was 1 kb. A schematic of the score computation is provided in the Supplementary Fig. S1 . Aggregated Hi-C maps analysis The aggregated Hi-C maps were generated using the coolpuppy package ( https://zenodo.org/badge/latestdoi/147190130 ). For normalization, the coolpuppy program utilized random shifting, a built-in feature designed to adjust positional biases in the data. Insulation scores for the aggregated maps were computed with this package by calculating the relative mean signal in the first and third quadrants compared to the second and fourth quadrants. Higher insulation scores indicate higher levels of spatial separation in this context. Saddle strength analysis To quantify A/B compartment interaction strength, we performed a saddle plot analysis based on the first principal eigenvector of the Hi-C contact matrix derived from chicken embryonic fibroblasts (CEFs). This eigenvector was computed at 100 kb resolution using the cooltools package (version 0.7.1) and served as a reference compartment annotation for all subsequent analyses. Eigenvector values were divided into 40 percentile-based bins. To minimize the influence of extreme regions with ambiguous compartment identity, bins corresponding to the highest and lowest mean eigenvector values were excluded from further analysis. Saddle plots were generated by calculating the observed-over-expected contact frequencies between each pair of bins. Compartment strength was defined as the ratio of mean intra-compartment interactions (AA + BB) to inter-compartment interactions (AB + BA). All computations were performed using functions implemented in cooltools (ver-0.7.1). Chicken liver ChIP-seq data preparation The CTCF ChIP-seq data were taken from [ 73 ] as sequenced reads. The reads were aligned at ASM2420605v2 chicken genome assembly [ 68 ] and bamCoverage ver-3.5.4 were used to get bigwig files. The ChIP signal bigwig was divided on input signal bigwig for each bin and logarithm from results were taken. NOMe-seq experiments and data analysis Nuclei were isolated and washed as described above. Individual nuclei were transferred immediately to 3 μl of GC Reaction Buffer (NEB) with 0.2% Igepal for 20 min at RT. Reaction of methylation was performed in 1× GC Reaction Buffer supplemented with 160 µM S -adenosylmethionine with 6U GpC methyltransferase at 37°C for 1 h. Proteines were digested by proteinase K at 65°C for 2 h. Libraries for whole genome bisulfite sequencing were prepared using Pico Methyl-Seq Library Prep Kit (Zymo Research) according to the user manual. For control experiments we also supplemented chicken oocytes nuclei with human K562 cells. Genomic libraries were sequenced using the BGI DNBseq G400 machine in 2 × 150 paired-end mode. The illumina adapters from raw sequences were cut with cutadapt ver-4.1 [ 66 ]. To process NOMe-seq reads, we used Bismark version 0.23.0 [ 74 , 75 ] in conjunction with Bowtie2 version 2.4.4 [ 76 – 78 , 79 ], setting the -t parameter to 10. Reads were aligned to the ASM2420605v2 genome assembly [ 68 ], and any reads with MAPQ < 10 were excluded. Methylation data from the aligned reads were then converted to the .bedgraph format using the bismark-methylation-extractor. The GpC methylation profile was obtained from the NOMe-seq experiment described above. The CpG methylation profile was obtained from the NOMe-seq experiment and merged with the profile from the Methyl-seq experiment from [ 80 ]. RNA-seq experiments and data analysis Nuclei were isolated and washed as described above. Individual nuclei were frozen immediately. Total RNA was isolated from the nuclei using TRIzol reagent (Thermo Fisher Scientific) according to previously described protocol [ 57 ]. The RNA samples were sent to BGI company to generate transcriptomic libraries using experimental workflow including the following steps: (i) oligoT-enrichment, (ii) RNA fragmentation, (iii) first-strand synthesis with random primers, (iv) second strand synthesis with dUTP instead of dTTP, (v) end-repair, 3′-adenylation and adaptors ligation, (vi) uracil–DNA–glycosylase treatment followed by PCR amplification and circularization, and (vii) sequencing on DNBSEQ in PE100 paired-end mode. Together with poly(A)-enriched stranded RNA-seq data generated in this study, the total RNA stranded RNA-seq data from [ 57 ] were used. The Illumina adapters from raw sequences were cut with cutadapt ver-4.1 [ 66 ]. All data were aligned with hisat2 ver-2.2.1 [ 81 ] on ASM2420605v2 genome assembly [ 68 ]. Bam files were prepared using samtools ver-1.15.1 [ 82 ], and bigwig files were made using bamCoverage ver-3.5.4 ( https://github.com/BGI-shenzhen/BamCoverage ). Each bam file was split on subfiles by strand (the direction of transcription) and duplicates were marked with “samtools markdup.” The unique reads were merged from all experiments (still grouped by strand) and de novo LBC transcript structure recovery was made with stringtie ver-2.2.2 [ 83 ]. The forward and reverse transcription files were separately used with stringtie and called as “FORW” and “REV,” respectively. Maximum gap allowed between read mappings was set as 300 bp; minimum assembled transcript length = 1000 bp; minimum junction coverage = 5; minimum anchor length for junctions = 10 bp; minimum isoform fraction = 0.05; minimum reads per bp coverage to consider for single-exon transcript = 3.5; minimum reads per bp coverage to consider for multi-exon transcript = 1.2. For the expression analysis, stranded RNA-seq data of the oocyte nuclei from this work and [ 57 ] was compared with the RNA-seq data of the CEFs from [ 84 ]. All raw reads were aligned on GRCg6a genome assembly. The TPMCalculator ver-0.0.3 was used to calculate TPM values for genes of interest for ncbiRefSeq.2020-04-01, ensGene.v101 and refGene annotations (see Supplementary Table S4 ). Transcriptional activity analysis The FPKM value was calculated for each gene and used as a measure of transcriptional activity in analysis. where i denotes the transcript index, : number of reads mapped to transcript i , : total number of mapped reads, : length of transcript i . To estimate the amount of nascent RNA connected to the elongating RNA polymerases, the values of intron FPKM and exon FPKM were calculated separately for each TU annotated. The minimum transcript FPKM value was used for the further analysis. For a block of consecutive co-directional genes (a tandem, t ), transcriptional activity was summarized by a length-weighted mean: Genes and their tandems were then stratified into four classes by median splits on their length and transcriptional activity: long strong, short strong, long weak, and short weak. Mitotic metaphase chromosomes preparation Mitotic metaphase chromosomes were prepared from CEF culture. Cell cycle was arrested by 0.1 μg/ml colcemid (Biological Industries) for 5 h, then the cells were fixed in cold 3:1 ethanol:acetic acid mixture. The CEF suspension was dropped onto the slides which had been heated in a water bath to 60°C and then air dried. Lampbrush chromosomes preparation LBC spreads were prepared according to a well-described procedure [ 85 ]. The experimental procedure was approved by the Ethics Committee for Animal Research of St. Petersburg State University (protocol #131-04-6 dated 25 March 2019). Nuclei were manually isolated from oocytes (1–2 mm), one by one, using tungsten needles under a stereomicroscope Leica S9D or M165C (Leica Microsystems) in the isolation medium “5:1” (83 mM KCl, 17 mM NaCl, 6.5 mM Na 2 HPO 4 , 3.5 mM KH 2 PO 4 , 1 mM MgCl 2 , and 1 mM dithiothreitol, pH 7.2). The isolated nuclei were then washed in the hypotonic solution (“5:1” medium diluted four times with the addition of 0.1% formaldehyde) and transferred to the isolation chamber mounted on a microscope slide and filled with the hypotonic solution. Nuclear envelope was broken manually with a pair of thin tungsten needles to release the nuclear content and slides were left to disperse for 15–20 min at +4°C. Slides were then centrifuged at 1600 × g for 30 min at +4°C in a cytology centrifuge (Hettich Universal 320R), fixed in 2% formaldehyde in PBS for 30 min, and dehydrated in a series of ethanol solutions of increasing concentrations. Specimens were mounted in an antifade solution containing 1 mg/ml (4′,6-diamidino-2-phenylindole) DAPI (Sigma), 55% glycerol, 10 mM Tris–HCl, 2% DABCO (Sigma), and were imaged prior to FISH procedure. Preparation of DNA probes To obtain DNA probes for FISH mapping of the selected scHi-C chromatin domains, 29 BAC clones were selected: 14 BAC clones in a region (55–50 Mb) on chromosome 1; 2 and 7 BAC clones in two regions (50–53 and 70–73 Mb) on chromosome 4, respectively; and 6 BAC clones in the region 3–7 Mb of chromosome 13 from the chicken BAC-library CHORI-261 ( https://bacpacresources.org/chicken261.htm ) ( Supplementary Table S3 ). BAC DNA was isolated from night culture by standard alkaline lysis. DNA was labeled by nick-translation [ 86 ] with dUTP-bio (DNA-Synthesis), dUTP-dig (Jena Bioscience), and dUTP-Atto637 (Jena Bioscience) using DNA polymerase I (ABclonal) and DNAse I (New England BioLabs). Labeled DNA probes in various combinations (up to 10 probes in a single mixture) were precipitated with salmon sperm DNA and, in case of DNA probes to the region of chromosome 1, with the C0t-5 fraction of chicken genomic DNA, and dissolved to a final concentration of 20 ng/μl for each DNA probe in the hybridization buffer containing 50% formamide, 10% dextran sulfate and 2× SSC. Fluorescence in situ hybridization Chromosomal position of each BAC clone-based DNA probe was first verified by DNA/DNA FISH on CEF metaphase chromosomes: preparations were pretreated with RNAse A (Thermo Scientific) and pepsin (Sigma–Aldrich). For LBCs, the DNA/DNA + RNA FISH and RNA FISH procedures were used, omitting the RNAse treatment [ 56 , 87 ]. Chromosome preparations were denatured in 70% formamide in 2× SSC at 70°C for 10 min and dehydrated in a series of cold (−20°C) ethanol solutions of increasing concentration. Hybridization mixtures were denatured at 90°C and cooled on ice and applied to the slides with the denatured (in case of DNA + RNA FISH) or undenatured (in case of RNA FISH) chromosomes. Hybridization was carried out overnight in a humidity chamber at 37°C, followed by a post-hybridization wash in 0.2× SSC at 60°C. Biotin was detected with Alexa488-conjugated streptavidin (Jackson ImmunoResearch Laboratories) and biotinylated anti-streptavidin antibody (Vectorlabs), and digoxigenin was detected with Cy3-conjugated mouse anti-digoxigenin antibody (Invitrogen) and Cy3-conjugated anti-mouse IgG (Jackson ImmunoResearch Laboratories). Immunostaining LBC preparations were immunostained with mouse monoclonal antibodies against phosphorylated serine 5 of CTD repeat YSPTSPS of RNA polymerase II [H14] (Abcam, ab24759), followed by the detection with Alexa-488-conjugated goat anti-mouse IgG (Molecular Probes). Immunofluorescence staining procedure was described previously [ 88 , 89 ]. Microscopy and image processing Chromosome preparations were analyzed with the fluorescence microscope Leica DM4000 (Leica Microsystems, GmbH) equipped with a monochrome CCD camera (1.3 Mp) and corresponding fluorescence filter cubes. LAS AF software (Leica Microsystems, GmbH) was used for image acquisition, and Adobe Photoshop CS2 (Adobe) and ImageJ ( http://rsbweb.nih.gov/ij/ ) were used for image processing and layout. Schematic drawings were created in CorelDRAW X5 and represent an approximation of the FISH mapping of DNA probes to lampbrush chromatin domains seen in at least four micrographs for each DNA probe. Polymer modeling We used polymer modeling based on the Polychrom tool [ 90 ] to gain insights into the roles of individual RNA polymerases in the folding of LBCs. The simulation was performed in the coarse-grained model approximation, where clusters of nucleosomes and individual polymerases acted as beads. 1D The modeled object is a one-dimensional array of N beads of sizes ranging from 45 to 225 bp, with dynamically loaded-unloaded RNA polymerases and active extruders (loop extrusion factors, LEFs) and nonremovable cohesins without the loop extruding activity (loaded with the density of 45 kb). The number of base pairs in a bead differs for TUs where RNA polymerases are located and for transcriptionally inactive regions with a dense occupancy by nucleosomes. The size of 45 bp per bead was used to represent one RNA polymerase (∼15 nm, [ 91 ]). The regions between the TUs are condensed by the nucleosomes ∼5 times (because ∼140 bp entangle it and ∼70 bp are between nucleosomes). Our estimation is also close to the estimation from [ 92 ]. TUs are placed according to the annotation obtained from RNA-seq data. RNA polymerases are represented as objects with the following properties: they are loaded at the beginning of TUs with a probability proportional to the activity of the TU (estimated from FPKM metric), from where they begin movement at a speed of vp = 100 bp/s [ 93 ] (the speed of LEFs is vL = 625 bp/s for the naked DNA [ 94 ]) along the transcript. Nascent RNA fiber growth from each RNA polymerase at length vp*dt for the time step of dt seconds. The introns are excised from the pre-mRNA attached to the RNA polymerases. If the bead in the direction of movement is occupied by a LEF or cohesive cohesin (i.e. SMC), then RNA polymerase pushes the SMC [ 37 ] and all subsequent SMCs in the direction of movement until they collide with another RNA polymerase. The mechanism of LEF action is as follows. It is randomly loaded onto a pair of adjacent beads that are not occupied by another LEF or RNA polymerase. Each time one of the two LEF retaining components can reattach to the next bead in the corresponding direction releasing the occupied one. The direction of movement changes every timestep (effectively two-way movement). Each part of the LEF can move independently of the other. If the next bead is occupied by another LEF then no movement occurs. If the next bead is occupied by RNA polymerase then the LEF can bypass the RNA polymerase with the stable probability Pbypass . The cohesive cohesin can be pushed by LEF if the next bead is not occupied. Extruding continues until the LEF is removed from the chromatin. The probability of removal is (1/lifetime ) for each timestep. Some parameters for these LEFs were determined by using P (s) data from the experiment Hi-C data on lampbrush-stage oocyte nuclei (see Supplementary Methods S1 : Modeling): lifetime = 1 Mb (∼60% of LEFs will pass this distance on naked DNA), average_loop_size = 90 kb (determines the density of loaded condensins on chromatin), and Pbypass = 0.01 (as in [ 37 ] article). 3D The next step after forming 1D snapshots of the SMCs and RNA polymerase locations on the chromatin is to transfer the interactions into 3D. For this purpose all beads are grouped into bead clusters of 10 which include nucleosomes and TUs. The sizes of the clusters vary from 450 to 2250 bp. New beads are placed into the starting conformation form of two parallel 2D fractals, corresponding to the two sister chromatids. The distances between beads in 3D varied according to the fractal dimension of the site ( Supplementary Methods S1 : Modeling). With the launch of a three-dimensional simulation, at each moment in time, a spring of a fixed size (∼40 nm, [ 95 ]) is stretched between each pair of beads connected by a given LEF, and the beads move according to a random Brownian motion. RNA tails from 1D simulation are converted into increased bead sizes by the Marco and Siggia model [ 96 ]. The repulsive potential between beads is realized as 0.5*k*(r1 + r2)^2 . A more detailed analysis of the model and parameters used is presented in the Supplementary Methods S1 : Modeling. In order to reproduce the patterns observed in LBC under microscopy, the original three-dimensional regions of the chromosome were flattened to 1 µm (see Supplementary Methods S1 : Modeling). LBC chromatin simulations were additionally validated using obtained FISH-mapping data in individual chromosomal regions (see Supplementary Note S3 ). The 3D simulated structures were visualized using plotly [ 97 ]. Online Supplementary files can be found via link ( https://genedev.bionet.nsc.ru/ftp/by_User/tlagunov/LBCs/htmls/ ) or be downloaded from zenodo ( https://doi.org/10.5281/zenodo.18998039 ). Simulated Hi-C The resulting 3D polymers were used to construct contact maps similar to Hi-C maps. The beads of different sizes are correctly converted to kb resolution that allow us to compare simulated and experimental contact maps of the specific chromatin slices. Since the contact map collected from one polymer conformation is essentially one conformation of an oocyte, the sum of contacts from one “ Dpn II site” in a bead was normalized to 8 as in the experiment (see Supplementary Methods S1 : Modeling for the details). Simulated Hi-C maps were then stored in mcool format and the analyses that are the same with experiment Hi-C maps were performed. Results Isolation and quality control of chicken diplotene oocytes To investigate chromatin organization in avian LBCs, we selected chicken oocytes at the lampbrush stage based on follicle size. Individual nuclei were isolated and processed using a previously established single-cell Hi-C protocol [ 98 ] (Fig. 1A ). Multiple quality control measures were conducted to validate the integrity and reliability of the resulting data. Figure 1. Open in a new tab Experimental workflow preserves structural integrity of chicken LBCs. ( A ) Experimental workflow, including oocyte dissection, oocyte separation by size (LBC stage oocyte: 1–2 mm diameter; post-LBC stage oocyte: 4–7 mm diameter), microsurgical isolation of nuclei, and preparation of single-nucleus Hi-C libraries. Selected oocytes and nuclei were randomly chosen for quality control. ( B ) LBCs preserve their 3D-structure after formaldehyde fixation. The entire nucleus is visualized after fixation and staining with SYTO 61. ( C ) Isolated DAPI-stained chicken oocyte nucleus (no evidence of contamination with other cell nuclei). To confirm that formaldehyde fixation preserves the LBC structure, we stained fixed nuclei with SYTO 61 dye and visualized them using confocal microscopy. The analysis demonstrated the characteristic lampbrush morphology of the captured chromosomes (Fig. 1B and Supplementary Fig. S2A ), consistent with previous findings [ 45 ]. Next, we ensured that oocyte genomic material was not contaminated by somatic cells. Chicken follicles include various cell types, such as erythrocytes, which are significantly smaller than oocyte nuclei. To exclude contamination, nine randomly selected oocyte nuclei were stained with DAPI post-isolation and visually inspected. Microscopy analysis revealed no DAPI signals outside oocyte chromosomes (Fig. 1C ). As a control, erythrocytes were manually added to oocyte samples, confirming that erythrocyte nuclei are detectable during visual inspection ( Supplementary Fig. S2B ). To further validate the absence of contamination, we leveraged the fact that a single oocyte nucleus contains no more than four copies of each genomic fragment. Consequently, no more than eight unique interactions should be detected per Dpn II restriction fragment (two ends per fragment, [ 99 ]). Genomic fragments exhibiting more than eight interactions would indicate contamination or Hi-C artifacts. Our analysis showed that the vast majority of fragments exhibit fewer than nine interactions. Comparison with the bulk Hi-C data downsampled to the same sequencing depth shows that the portion of fragments with fewer than nine interactions in chicken oocytes is lower than expected based on total Hi-C contacts count ( Supplementary Fig. S2C ). Finally, the Hi-C contact profiles of chicken oocyte nuclei display a characteristic pattern distinct from that of typical somatic cells (described below). Taken together, these results confirm that the isolated oocyte nuclei were free of contamination from other chicken cell types. Single-nucleus Hi-C experiments were performed on a total of 96 oocyte nuclei. Initial library quality was evaluated through shallow sequencing (∼200 000 reads per sample), using metrics such as the fraction of unmapped reads, duplicate reads, intra-chromosomal reads, and intra-fragment reads (see Materials and methods). Based on these criteria, 25 single-cell oocyte libraries were selected for deeper sequencing (∼3–40 million read pairs per nucleus) and downstream analysis ( Supplementary Table S1 ). The resulting Hi-C maps cover >80% of bins with at least one interaction at 4 kb resolution, with a few exceptions where one or more chromosomes were completely absent but the remaining genome showed consistent coverage ( Supplementary Fig. S2D ). Deeply sequenced Hi-C maps displayed a high degree of similarity across individual oocyte nuclei, confirming the high quality of the obtained data. Lampbrush chromosomes form elongated structures with minimal inter-chromosomal interactions LBCs at the diplotene stage of meiosis are known to form elongated structures that are spatially separated within the giant oocyte nucleus [ 45 ]. This spatial arrangement is also preserved after fixation, as shown in Fig. 1B and Supplementary Fig. S2A . Hi-C data analysis further confirms these unique features of LBCs. Our results show a strikingly low frequency of inter-chromosomal (trans) contacts, with only 0.2%–1% of spatial interactions occurring between nonhomologous chromosomes (Fig. 2A and Supplementary Table S1 ). This is markedly lower than the ∼10% reported for chicken fibroblasts ([ 15 ] and Fig. 2A ) and the 10%–30% typical of somatic cells in various species ([ 14 , 100 ] and Fig. 2A ). These data underscore the exceptional rarity of inter-chromosomal interactions in lampbrush-stage oocytes (Fig. 2A and Supplementary Table S1 ). Figure 2. Open in a new tab Chromatin organization of LBCs at different genomic distances. ( A ) Fraction of inter-chromosomal contacts across different cell types and species (see Materials and methods). Chicken oocytes at the LBC stage exhibit a markedly reduced proportion of inter-chromosomal contacts, consistent with cytological observations of spatial chromatin organisation. ( B ) Contact probability versus the genomic distance for single and merged oocyte data (lampbrush chromosome stage, LBC) and CEFs (see Materials and methods). The genomic distances are cut to display only representative contacts of oocytes. See extended version in Supplementary Fig. S3A . ( C ) Hi-C contact maps comparing pseudo-bulk chicken oocytes and bulk fibroblast cells for chromosome 11. The fibroblast map was downsampled to match oocyte data (both maps sum to 19 946 contacts within the depicted region). Each bin represents 32 kb. See extended version in Supplementary Fig. S3B . ( D ). Chromomere–loop structure of isolated chicken lampbrush chromosome 1. Active RNA polymerase II is labeled with antibodies (green), and chromatin is counterstained with DAPI (blue). Intra-chromosomal ( cis) interactions in LBCs exhibit an abrupt drop in contact frequency beyond ∼1 Mb (Fig. 2B and C , and Supplementary Fig. S3A ), consistent with the elongated, rope-like morphology of these chromosomes (Fig. 2D ). The depletion of both long-range and trans -chromosomal interactions in LBC Hi-C maps results in the absence of the plaid-pattern of contacts typically observed in interphase nucleus Hi-C data, which is associated with chromatin compartmentalization ([ 101 ], Supplementary Figs S3B and S4 ; see Materials and methods). Chromatin of lampbrush chromosomes forms domains with stable boundaries shared across individual oocytes In contrast to the absence of a plaid-pattern, chromatin domains are clearly visible on LBC Hi-C maps at scales of 0.5–2 Mb (Fig. 2C ). To understand whether the domain organization observed in chicken LBCs represents a general feature of meiotic chromatin or a unique, potentially LBC-specific mechanism, we compared our results with mouse diplotene oocytes. Available Hi-C maps for individual nuclei of mouse diplotene oocytes, obtained by a similar approach [ 42 ], provide an opportunity to compare contact chromatin maps in organisms with different types of oogenesis. In mouse diplotene oocytes at the nonsurrounded nucleolus (NSN) stage, chromatin interaction analysis reveals TADs with highly variable boundary locations across individual cells [ 42 ]. In contrast, single-nucleus Hi-C maps from chicken lampbrush-stage oocytes demonstrate striking consistency in domain boundary positions (Fig. 3A – C ), with a mean normilized F1 score (see Materials and methods) of 2.3 compared to 1.3 for mouse NSN oocytes (Fig. 3A ). This stability in domain boundaries across chicken LBCs suggests fundamental differences in the mechanisms underlying chromatin domain formation between these two systems. Figure 3. Open in a new tab Chromatin domains in chicken LBCs. ( A ) Pairwise normalized F1 scores for domain boundary concordance across individual oocytes. Left panel: chicken lampbrush-stage oocytes; right panel: top 14 mouse NSN oocytes by coverage [ 42 ]. Oocytes are sorted by the number of Hi-C contacts per bp. Higher normalized F1 scores indicate greater concordance in boundary positions. P -value between chicken and mouse normalized F1 score distributions <10 −30 . ( B ) Domain annotations for individual cells and merged Hi-C data for chicken (left) and mouse (right) oocytes. Note the higher concordance of domain boundaries in chicken oocytes compared to mouse oocytes. Resolution: 20 kb. ( C ) Aggregated Hi-C contacts and domain annotations for 14 chicken oocytes (left) and top 14 mouse oocytes by coverage (right). ( D ) Merged Hi-C data from chicken oocytes with domain annotations overlaid. Domains were manually refined after automatic annotation (see Materials and methods). The dashed line indicates the N50 domain size (∼1.3 Mb). The dash-and-dot line indicates genomic distance where a sharp drop in the frequency of contacts was observed (distance with intra-domain fractal dimension 1, stick-like polymer, see Supplementary Fig. S7A ). ( E ) Each horizontal segment represents a single chromatin domain, with the Y -position of the line indicating the domain’s average characteristic (CpG methylation, GpC methylation, or AT fraction, see Materials and methods). The dashed horizontal line represents the genome-wide average for each characteristic. This visualization highlights the distinct properties of LUDs and SSDs. The genome-wide CpG and GpC methylation distributions can be found in Supplementary Fig. S5C and D . The genome-wide domain AT fraction distribution can be found in Supplementary Fig. S7B and C . ( F ) Distribution of LUDs and SSDs lengths on chicken chromosomes. ( G ) TUs density and activity distribution show that the SSDs contain significantly more TUs, and TUs with higher activity, than LUDs. ( H ) CpG and GpC methylation levels statistically compared between SSDs and LUDs (see Materials and methods). As a baseline, domains were randomly redistributed across the genome while preserving the number of intervals and their length distributions (indicated as “random”). SSDs exhibit significantly higher GpC methylation levels, reflecting increased chromatin accessibility. The genome-wide CpG and GpC methylation distributions can be found in Supplementary Fig. S5C and D . Panels F–H collectively demonstrate that LUDs are enriched on macrochromosomes and the Z chromosome and are characterized by low gene density and transcriptional quiescence, whereas SSDs are gene-dense, transcriptionally active, and broadly distributed across the genome. Contact domains in LBCs can be classified into two types based on their length (Fig. 3D and E ): (i) l arge u nstructured d omains (LUDs), defined as domains longer than 1.3 Mb (N50 of domains length, see Materials and methods) and (ii) s mall s quared d omains (SSDs), shorter than 1.3 Mb. Visually, LUDs have more dispersed contacts that are connecting loci up to ∼2 Mb from each other ( Supplementary Fig. S5A ). Almost all LUDs are located on macro- (chr1-chr5) and Z chromosomes (Fig. 3F ). To characterize chromatin states in LUDs and SSDs, we performed a single cell NOMe-seq experiment. NOMe-seq measures open chromatin via deposition of exogenous GpC methylation, enabling simultaneous analysis of chromatin accessibility and CpG methylation levels in individual nuclei. We augmented NOME-seq data with CpG methylation from [ 80 ] (see Materials and methods). We also quantified transcription levels, reanalyzing RNA-seq data from the previous chicken oocyte nucleus transcriptome study [ 57 ] and supplementing it with additional replicas generated in this work. Analysis of these data shows that LUDs have lower median gene density (Fig. 3G and Supplementary Fig. S5B ) and lower normalized expression levels than SSDs (Fig. 3G ). We found that SSDs have significantly higher levels of chromatin openness (measured in NOME-seq experiment via GpC methylation), indicating that shorter domains contain more accessible chromatin ( Supplementary Fig. S5C ), while there is almost no difference in CpG methylation levels between SSDs and LUDs (Fig. 3H and Supplementary Fig. S5D ), as well as only a weak correlation between promoter methylation and gene expression levels ( Supplementary Fig. S6 ). We note that the length threshold separating LUDs and SSDs is somewhat arbitrary, as the properties of both gradually change with the domain length ( Supplementary Fig. 5E ). Our findings indicate that LUDs correspond to gene-poor, transcriptionally inactive chromatin regions predominantly found on macrochromosomes and the Z chromosome. In contrast, Small Squared Domains are associated with gene-rich, transcriptionally active chromatin regions distributed across all chromosomes. To further validate the correlation between domain types and DAPI staining in chromomeres, we analyzed sequence markers with known positions in cytological maps of chicken LBCs, primarily derived from DNA- and RNA-FISH experiments using specific BAC clones (Fig. 4A and B ). The distribution of LUDs and SSDs correlated with the arrangement of DAPI-positive and DAPI-negative chromomeres at certain regions of chromosomes. For instance, we observed multiple LUDs between markers W030P07 and W030B21, coinciding with a region enriched in DAPI-positive chromomeres (Fig. 4A ), as well as in the region at the end of the long arm of chromosome 1 (Fig. 4B ). However, the number of cytologically visible chromomeres and contact domains on Hi-C maps does not always match exactly, probably due to the limited resolution of microscopy analysis and/or errors in domain calls. Together with sparse availability of sequence markers distinguishing chromomeres and resolution of cytological chromomere–loop maps this limited a genome-wide analysis of the relationship between SSDs, LUDs, and chromomeres. Figure 4. Open in a new tab Concordance between microscopically visible chromatin structures and Hi-C map patterns along chicken LBCs. ( A ) Comparison of Hi-C contact domains with chromatin domains identified through microscopy visualization of LBCs. From top to bottom: Hi-C map for chicken LBC 1, showing domain boundaries and genomic positions of DNA markers; Contact domain segmentation, with each triangle representing a single domain. LUDs and SSDs are distinguished by triangle colors, and black lines indicate DNA marker positions; Cytological map of chromomere distribution along chicken LBC 1 [ 102 ], with DAPI-positive chromomeres shown as black circles and DAPI-negative chromomeres as white circles. The size of the circles reflects the size of the chromomere on the phase contrast images. The average lateral loop length in a region is shown by rectangles. Two “marker loops” are also indicated. Red circle indicates the position of centromere; DAPI-stained image of chicken LBC 1 (a fire look-up table corresponding to the intensity of fluorescence is used) (scale bar: 20 μm). The chromosome is oriented with its left end toward the telomere of the short arm. Note that the chromomere pattern is almost the same in two homologous half-bivalents joined at chiasmata. ( B ) Zoomed region of a Hi-C map for a fragment of LBC 1 presented on panel (A). Designations as on panel (A). ( C ) Hi-C map displaying an example of a “cross” pattern. The “cross” region overlaps with a high RNA-seq signal and CPS (see Materials and methods) and shows no reduction in overall Hi-C coverage (denoted as “bin coverage”). ( D ) Aggregated Hi-C contact maps centered on the local minima of CPS; resolution: 10 kb. ( E ) Rescaled observed-over-expected Hi-C contact map averaged around the longest (top 50% by length) and most transcriptionally active (top 50% by FPKM) genes. A characteristic cross-shaped contact pattern is observed, extending approximately ± 6 gene lengths from the gene end. Data represent aggregated signals from multiple genes at 2 kb resolution. Based on this characteristic, we propose that SSDs and LUDs in single-nucleus Hi-C maps correspond to chromomeres in LBCs. Additionally, given their properties, we speculate that LUDs represent more compact, DAPI-positive chromomeres, while SSDs correspond to less compact, DAPI-negative chromomeres. Supporting this hypothesis, statistical analysis of the AT fraction in domains—reflecting the genomic proportion susceptible to DAPI staining—shows a weak but significant correlation between AT fraction and domain length (Spearman’s R = 0.55, compared to R = 0.28 for randomly shuffled domains, Supplementary Fig. S7B and C ). Hi-C patterns of transcription units and compact chromatin domains align with transcription loops and chromomeres identified by microscopy analysis Transcription loops are a hallmark of LBCs. These loops represent stretched fibers of the actively transcribed chromatin emanating from the LBC axis (Fig. 2D ) . Given their spatial separation from each other and the LBC axis, we hypothesize that loci corresponding to transcription loops lack interactions with the rest of the genome. Visual inspection of Hi-C maps reveals loci with a distinct “cross” pattern, where interactions are limited to nearest neighbors (Fig. 4C ). These loci compensate reduced long-range interactions by increasing frequency of proximal interactions, thus maintaining consistent coverage levels (Fig. 4C ). To systematically identify and visualize these features, we introduce a quantitative CPS that captures the relative enrichment of short- versus long-range contacts at each locus (see Materials and methods). Although we observe some variability on “cross” pattern manifestation across loci, aggregated Hi-C maps centered on the regions with low CPS (predominant short-range contacts) display characteristic “cross” pattern (Fig. 4D ), demonstrating that “cross” features are reproducible genome-wide. “Cross” patterns frequently occur within chromatin domains and almost always coincide with actively transcribed genes. Supporting this observation, aggregated Hi-C maps centered on long, highly expressed genes (top 50% by length and TPM) show a pronounced “cross” pattern (Fig. 4E ). These findings suggest that “cross” patterns in Hi-C maps represent the contact signatures of transcription loops in LBCs. To align lampbrush chromatin contact patterns identified by Hi-C analysis with chromatin structures observable via microscopy, we conducted FISH mapping using probes specific to chromatin domain boundaries, internal domain regions, and loci forming “cross” patterns. A total of 29 BAC clones were hybridized in various combinations on LBCs 1, 4, and 13 ( Supplementary Table S3 ). The large size of LBCs enables detailed microscopy analysis, allowing us to distinguish chromomeres and transcription loops, as well as determine the orientation of TUs. The locus analyzed on chromosome 1 contains six Hi-C domains and was covered by 14 BAC clone based probes. Regions 4, 6, 7, 9, and 12 coincide with “cross” patterns with reduced long-range interactions observed in Hi-C maps and correspond to actively transcribed genes (Fig. 5A ). Note that “cross” patterns demonstrate locus-specific heterogeneity in their structure depending on the length and orientation of TUs. Based on the annotated TUs' location and orientation, we anticipated transcription loops with forward-oriented transcription for regions 4, 9, and 12, and reverse-oriented transcription for regions 6 and 7. FISH mapping of LBC1 using BAC clone-based probes with a DNA + RNA hybridization protocol confirmed these predictions. Long lateral loops with a prominent RNP matrix were observed for BACs 4, 9, and 12, with transcription co-directional to the chromosome (Fig. 5B , subpanels 2a, and 3a and b; Supplementary Fig. S8 , subpanel c). Similarly, genomic regions 6, 7, and 9, covered by three different BACs, correspond to TUs forming one or two pairs of lateral loops (Fig. 5B , subpanels 2c–e; Supplementary Fig. S8 , subpanels b and c). BAC 5 doesn’t correspond to a region with an obvious “cross” pattern according to the Hi-C data; however, it covers the transcriptionally active region (in reverse direction), thus it can form a transcription loop (Fig. 5B , subpanels 2a and b). In our FISH experiments, BAC 5 marks a reverse oriented transcription loop anchored in the chromomere (Fig. 5A and C ). Another combinations of DNA-probes mapped on the chromosome 1 are represented on Supplementary Fig. S8 . Additionally, RNA-FISH experiments without chromosomal DNA denaturation for several transcribed regions confirmed that BAC clone-based probes hybridize with nascent transcripts on the lateral loops ( Supplementary Fig. S8 , subpanel c). These findings confirm that “cross” patterns observed on Hi-C maps of avian LBCs correspond to the microscopically visible lampbrush lateral loops containing actively transcribed genes. Figure 5. Open in a new tab Comparison of Hi-C map patterns with microscopically identified chromatin domains and loops in LBCs 1 and 4. ( A ) Hi-C map and associated data for a fragment of chromosome 1. From top to bottom: combined Hi-C map of the oocyte nuclei, with colored highlights indicating interactions in regions covered by BAC clone-based probes; CPS (see Materials and methods); Total stranded RNA-seq data for lampbrush-stage oocyte nuclei [ 57 ]; Annotation of TUs based on de novo transcriptome analysis (see Materials and methods); BAC clone numbers and names (see Supplementary Table S3 ); resolution: 16 kb. ( B ) DNA + RNA FISH mapping images for different combinations of BAC clone-based probes on LBC 1. Chromosome fragments are oriented with their left end toward the left telomere. All panels except a1 include DAPI signals (blue or grayscale). Panels a1, e2, a3, and d3 merge fluorescent images with phase contrast images. White arrows in panels 1a–e, 2c–e, and 3a–c indicate a compact chromatin domain without transcriptional activity (covered by BAC 8) that may either anchor to the chromosome axis (as seen in one sister chromatid in the lower half-bivalent on panel 2c–e) or remain insulated from the axis forming an isolated knot. Arrowheads in panels 3d and e indicate partial colocalization of genomic regions covered by BACs 10 and 11 within a compact chromomere. Another combinations of mapped DNA-probes can be found in Supplementary Fig. S8 . ( C ) Schematic representation of chromomere–loop organization for a fragment of LBC 1, derived from FISH mapping data (half-bivalent is shown). Numbers correspond to the locations of BAC clone-based FISH probes aligned on chromosome 1. ( D ) Hi-C map for region A on chromosome 4. The data are presented similarly as in panel (A) above; resolution: 16 kb. ( E ) Schematic representation of chromomere–loop organization for region A on LBC 4, based on FISH mapping data (half-bivalent is shown). Numbers correspond to the locations of BAC clone-based FISH probes aligned on region A. ( F ) DNA + RNA FISH mapping images for two BAC clone-based probes in region A on LBC 4 (half-bivalent is shown). Fluorescent images are merged with DAPI-stained chromatin (grayscale) or phase contrast images; scale bars: 10 μm . The intra-domain regions 1, 2, 3, 8, 10, and 14 correspond to five of the six contact domains identified on LBC Hi-C maps, with BACs 2 and 3 covering the same domain (Fig. 5A ). Consequently, we anticipated observing five insulated chromomeres along the chromosome axis in microscopy images, one of which would be associated with two BACs. FISH experiments confirmed this, with BAC clone-based probes mapping to five distinct chromomeres (or chromatin knots in case of BAC8, see below) (Fig. 5B , subpanels 1a–e, and 3d and e; Fig. 5C ; Supplementary Fig. S8 , subpanels e–g). The chromomeres covered by BACs 2 and 3 are closely spaced but remain distinguishable (Fig. 5B , subpanels 1c and d; Supplementary Fig. S8 , subpanel a). Interestingly, the compact chromatin domain covered by BAC 8, which lacks TUs, exhibits variable positioning. This chromatin domain can either anchor directly to the chromosome axis (Fig. 5B , panels 2c–e; Supplementary Fig. S8 , subpanel d) or remain detached from it (indicated by white arrows in Fig. 5B , panels 1a–e, 2c–e, and 3a–c), forming a compact “knot” on a lateral loop emanating from the chromosomal axis. Four signals for BAC8 correspond to four chromatin knots on each of the sister chromatids (two sister chromatids on each half-bivalent). These chromatin knots, located within transcriptionally active lateral loops, have been previously described as small genomic regions devoid of transcribed genes (Fig. 6 in [ 57 ] ). Together, these results demonstrate that (i) chromatin contact domains identified through Hi-C analysis correspond to insulated chromomeres in LBCs, and (ii) small regions between actively transcribed genes can form microdomains (knots) that are optionally tethered to the chromosomal axis. Region 11 exhibits partial transcriptional activity and is positioned near the boundaries of chromatin domains (Fig. 5A ). Based on this, we anticipated that BAC 11 would cover a portion of chromatin spanning two adjacent chromomeres, one of which is covered by BAC 10. FISH mapping confirmed this prediction, showing partial colocalization of BACs 10 and 11 within a prominent DAPI-positive chromomere (Fig. 5B , subpanels 1a–d, and 3d and e; Supplementary Fig. S8 , subpanels a and f). The DAPI-positive chromomere partially covered by BACs 10 and 11 corresponds to the most prominent contact domain within analysed genomic locus. Similarly, BAC 13, located near the boundary of contact domain, produced a signal spatially close to that of BAC 14, which maps to the same domain. While the signals from BACs 13 and 14 are closely positioned, they remain distinct and do not intermix ( Supplementary Fig. S8 , subpanel e). These results confirm that boundaries between domains identified by Hi-C are concordant with LBC chromomere boundaries (see additional evidence supporting this statement in the Supplementary Note S2 ). On LBC 4, we mapped nine BAC clone-based probes across two regions of interest. In the region A (51–52 Mb), Hi-C data revealed the formation of a contact domain with interactions at the top of the domain triangle and a wide “cross” pattern within domain, indicative of chromatin loop formation (Fig. 5D ). Nuclear RNA-seq data showed that several genes within region A are transcribed in varying orientations. FISH mapping confirmed that this genomic region corresponds to a long transcription loop with multiple RNP matrix gradients (Fig. 5E and F ), consistent with the transcriptional activity detected. Detailed results for region B are provided in Supplementary Note S2 and Supplementary Fig. S9 A-C. We integrated the FISH data to reconstruct the chromatin organization of the analyzed region on chromosome 1 (Fig. 5C ), accounting for two alternative states of chromatin knots discussed above. Schematic drawing represents an approximation of the FISH mapping data for different combinations of DNA probes to lampbrush chromatin domains seen in at least 4 micrographs (half-bivalent is shown). This reconstruction demonstrates a strong concordance between microscopically visible chromatin structures—such as chromomeres and transcription loops—and the contact domains and “cross” patterns observed in Hi-C maps. The reconstruction of the region A on chromosome 4 also confirmed the correspondence between the transcription loops and the “cross” patterns (Fig. 5D and E ). Additionally, similar analyses of two other regions on chromosomes 4 and 13, which included BACs mapped to Hi-C domain boundaries, intra-domain fragments, and “cross” patterns, yielded consistent results ( Supplementary Fig. S9 and Supplementary Note S2 ). These findings validate the Hi-C maps of chicken oocyte nuclei and confirm the interpretation of Hi-C patterns as representations of LBC chromomeres and lateral loops. Chromatin domains in lampbrush chromosomes are formed by a CTCF-independent mechanism In the majority of vertebrate cell types, including chicken fibroblasts [ 15 ], TAD boundaries are demarcated by CTCF-binding sites [ 1 , 101 ]. The moderate conservation of CTCF-binding site distribution across cell types, combined with a strong correlation between CTCF binding and TAD boundaries, allows CTCF enrichment to be detected even when comparing Hi-C maps to CTCF ChIP-seq data from different cell types [ 15 ] and cell-type agnostic motif enrichment. We show that LBC domain boundaries display almost no enrichment of CTCF motifs (Fig. 6A ). Concordantly, almost no insulation was detected when LBC Hi-C contacts were aggregated across CTCF motifs (Fig. 6B ). To provide a ground for comparison, we performed the same analysis for chicken fibroblasts data downsampled to the same sequencing depth. In fibroblasts, we detect strong insulation across CTCF motifs and enrichment of motifs at TAD boundaries (Fig. 6A and B ). We next replicated this analysis using CTCF ChIP-seq signal from chicken liver, which confirms the absence of CTCF-mediated insulation in LBCs (Fig. 6C ). In contrast, strong enrichment of chicken liver CTCF ChIP-seq signal was observed at fibroblast TAD boundaries (Fig. 6C ), validating the approach and reinforcing the conclusion that chromatin contact domains in LBCs are formed independently of CTCF binding. Figure 6. Open in a new tab The mechanism of chromatin contact domain formation in LBCs is CTCF independent, despite transcriptional activity of the CTCF gene. ( A ) Frequency distribution of CTCF-motifs at domain boundaries in LBC and CEF maps. Both LBC and CEF boundaries show enrichment relative to the genome average, with peak enrichment factors of ∼1.4 and ∼2.3, respectively. ( B ) Aggregated Hi-C contact maps centered on CTCF-motifs with uniform orientation. While strong insulation is evident around CTCF-motifs in CEF, only weak insulation is observed in LBC contact maps; resolution: 10 kb, fibroblast data subsampled to match number of Hi-C contacts in merged LBC. ( С ) Input-normalized CTCF ChIP-seq signal from chicken liver (SRR4068194, see Materials and methods) at oocyte contact domain boundaries and fibroblast contact domain boundaries. The absence of CTCF enrichment across LBC domain boundaries contrasts with the strong signal observed in fibroblast boundaries. ( D ) Comparison of transcript levels for cohesin subunits, condensin subunits, and CTCF (TPM fold-change relative to the somatic cell median). Oocytes exhibit reduced transcript levels for SMC3, SMC1A, SMC2, and SMC4 genes, higher transcript levels for SMC1B and REC8 genes, and similar CTCF transcript levels compared to fibroblasts. Gene annotations were sourced from ncbiRefSeq, except for STAG3, which used ensGene annotations (see Supplementary Table S4 ). Green dots represent transcript levels from three chicken oocyte RNA-seq datasets, and black dots represent five replicates of CEFs [ 84 ]. Oocytes 1: Total RNA stranded RNA-seq data [ 57 ]. Oocytes 2 and 3: Poly(A)-enriched stranded RNA-seq replicates generated in this study . To understand which chromatin architecture proteins can contribute to the formation of contact domain boundaries in oocytes, we explored the chicken oocyte nucleus transcriptome data (see Materials and methods). Despite the absence of CTCF binding enrichment in the domain boundaries, in oocytes CTCF gene has the comparable transcript levels with fibroblasts. Genes for the other architectural proteins, such as RAD21, STAG1, STAG2 (cohesin) and NCAP-H, NCAP-D2, NCAP-G, NCAP-D3, NCAP-H2 , and NCAP-G2 (condensin) subunits also show comparable transcript levels in fibroblasts and oocytes (Fig. 6D and Supplementary Table S4 ). Consistent with this, the presence of SMC1A, SMC3, Rad21, STAG1, and STAG2 on avian LBC axes in a punctate pattern has previously been shown by immunostaining, with an enrichment of cohesin subunits in the interchromomere regions and in chromomere cores [ 88 ]. Among the condensin subunits, XCAPD2 was found to accumulate in chromomeres of Xenopus LBCs [ 103 ]. However, genes for the main cohesin subunits (SMC3 and SMC1A) and the main condensin subunits (SMC2 and SMC4) have about ten times weaker transcript levels in the LBC stage oocyte (see Fig. 6D and Supplementary Table S4 ). In contrast, gene for the meiosis specific cohesin subunit SMC1B in oocytes has an almost thousand times higher level of transcripts than in fibroblasts (Fig. 6D and Supplementary Table S4 ). Thus, although components of extrusion complexes are present at LBC-stage, their activity might be modulated by quantity of proteins and usage of cell type specific subunits, potentially explaining why extrusion is not blocked by CTCF binding sites as in typical somatic cells. Moreover, the synaptonemal complex protein SYCP3 has been found at the axes of avian LBCs [ 88 ], which may affect the formation of structural chromatin loops. Transcriptional activity and orientation of transcription units define lampbrush chromosome chromatin domains To investigate the genomic features underlying the formation of LBC domains, we compared domain borders with position-dependent genomic elements. CpG methylation levels and chromatin accessibility at domain boundaries were similar to the genomic average ( Supplementary Fig. S10A , see Materials and methods). However, boundaries showed enriched GC-content relative to the background, suggesting a correlation with the presence of genes. Aggregation of contacts around exons, introns, and repetitive elements revealed no insulatory potential for these features, arguing against their role in forming domain boundaries at the lampbrush stage ( Supplementary Fig. S10B ). In contrast, we observed a significant depletion of contacts across 3′-UTRs and, to a lesser extent, across 5′-UTRs (Fig. 7A ). This suggests that the distribution and orientation of TUs may play a key role in LBC domain formation. Figure 7. Open in a new tab Aggregated 3D chromatin interaction profiles reveal genomic features underlying LBC chromatin domains. ( A ) Aggregated contact maps around 5′- and 3′-UTRs show depletion of interactions across these regions. Insulation scores are displayed in the upper-left corner of each aggregated map triangle (see Materials and methods); resolution: 10 kb. ( B ) Pie charts displaying TU) orientations for all TU pairs and those forming domain boundaries. A significant enrichment of forward-reverse (FR, convergent) orientation is observed at domain boundaries. ( C ) Aggregated contact maps centered between TUs in different orientations: forward-forward (FF, co-directional), reverse-reverse (RR, co-directional), forward-reverse (FR, convergent), and reverse-forward (RF, divergent). Insulation score, shown in the upper-left corner of each map, is highest for convergent (FR) TU pairs. RF pairs typically form a “cross” pattern, while FR, FF, and RR pairs display a stripe-like enrichment of contacts between TUs and upstream loci, with FF and RR stripes widening with distance from the diagonal. Insulation scores are displayed in the upper-left corner of each aggregated map triangle (see Materials and methods); resolution: 10 kb. ( D ) Statistics of FR-oriented TU pairs. Despite their prevalence at domain boundaries, only ∼15%–20% of FR-oriented TU pairs form boundaries. ( E ) Comparison of TU features between border-forming and non-border FR-oriented TUs. Border-forming TUs are, on average, longer, more highly expressed, and separated by greater intergenic distances. However, the length of the shorter TU and the activity of the weaker TU in the pair show higher difference between border-forming and non-border FR-oriented TUs than the features of the longer or more strongly expressed TU. P -values obtained using Mann–Whitney U test. To explore the potential link between transcription and LBC domain boundaries, we aligned RNA-seq data from poly(A)-enriched and total stranded RNA libraries ([ 57 ], see Materials and methods) with Hi-C maps and visually inspected these profiles. This analysis highlighted transcriptional activity as a critical determinant of LBC chromatin domain organization. Visual comparison of transcriptomic data with chromatin domain boundaries revealed multiple TUs absent from the current gene annotations in [ 68 ] ( Supplementary Fig. S11 ). This discrepancy reflects the dependence of chicken genome annotation on cell type and database, particularly for noncoding RNAs [ 104 ]. To address this, we performed de novo transcriptome assembly using oocyte RNA-seq data, generating a refined set of TUs active during the LBC stage (see Materials and methods). Analysis of this updated TU set confirmed a strong correlation between chromatin insulation and TU orientation (Fig. 7B and C ). Genomic regions flanked by TUs in a convergent (forward-reverse, FR) orientation were significantly enriched at chromatin domain boundaries (Fig. 7B ). Furthermore, FR-oriented TU pairs exhibited the highest insulation scores compared to other TU pair orientations (Fig. 7C ). Manual validation of domain boundaries revealed that nearly half of non-FR TU pairs initially identified at domain boundaries were artifacts of automatic domain annotation or transcriptome assembly (26 of 47 randomly selected for inspection boundaries; Supplementary Fig. S17 ). Overall, convergent transcription accounts for at least 79% of domain boundaries, with 62% supported by automatic annotation and an additional 17% identified through manual correction. Given the central role of transcription in defining LBC chromatin domains, we designate the domains detected in Hi-C maps as T ranscription- D ependent D omains ( TDDs ). Although the majority of TDD boundaries are formed by TU pairs in a convergent (FR) orientation, only ∼15% (∼20% after manual curation) of FR-oriented TU pairs create boundaries, while the remainder are located within domains (Fig. 7D ). To investigate the properties of TU pairs associated with boundary formation, we compared transcript features of “border” versus “non-border” TU pairs in a convergent orientation (Fig. 7E ). No single feature was sufficient to distinguish “border” TU pairs from “non-border” pairs. However, we observed that TU pairs forming boundaries tend to have several distinguishing characteristics: they are, on average, longer, separated by a larger intergenic interval, and exhibit higher activity of the weaker TU of the pair. The activity of the stronger TU in the pair shows minimal difference between boundary-forming and non-boundary-forming pairs. These observations suggest that TDD boundary formation is governed by a complex interaction of TU features rather than a single determinant. Fusion of transcription-dependent chromatin domains during oocyte growth and decreased transcriptional activity Growth of chicken oocyte coincides with reduced transcriptional activity and gradual contraction of both chromosomes and lateral loops [ 105 ]. To study the dynamics of this process, we isolated nuclei from the late-diplotene stage oocytes (hereinafter referred to as post-LBC) and performed a single-nucleus Hi-C experiment (Fig. 1A , and Supplementary Tables S1 and 2 ). Overall, Hi-C patterns detected in the mid-diplotene (LBC) stage persist at the post-LBC stage. However, there were quantitative differences. Concordant with cytological evidence of LBC contraction, we observe an increase of long-range interactions at the post-LBC stage (Fig. 8A and B ). Moreover, the significant decrease of the loop sizes of the chicken post-LBC chromosomes [ 105 ] agrees well with the almost complete absence of the “cross” patterns (Fig. 8C ). This observation confirms our suggestions about the critical role of the transcription loop repulsion in the “cross” pattern formation. Figure 8. Open in a new tab Comparison of oocyte chromatin at the lampbrush and post-lampbrush stages. ( A ) Contact probability as a function of genomic distance for lampbrush (LBC) and post-LBC (post-LBC) stages. Post-LBC chromatin exhibits an increase in distant contacts, consistent with microscopy observations [ 105 ]. ( B ) Pseudo-bulk Hi-C interaction maps for LBC (8719 contacts) and post-LBC (4949 contacts) chromatin. Gray lines indicate automatically adjusted domain borders for each stage. Neighboring domains fuse in the post-LBC stage, as demonstrated in the shown examples. ( C ) Aggregated observed-over-expected Hi-C contact maps comparing LBC stage (upper triangle) and post-LBC stage (lower triangle) chromatin. The characteristic cross-shaped contact enrichment associated with “strong” single genes (see Materials and methods) is markedly reduced in the post-LBC stage, whereas convergently oriented TUs continue to delineate insulating domain boundaries. The first two panels depict genes rescaled to span 1/12 of the figure width, with flanking regions extending ± 6 TU lengths from gene ends (2 kb resolution). The third panel shows aggregated maps for convergently oriented gene pairs within a ± 500 kb window around their midpoint (5 kb resolution). Insulation scores are indicated in the upper-left corner (see Materials and methods). ( D ) Distribution of domain sizes across different chromosome types. The post-LBC stage shows an increase in mean domain size alongside a reduction in the total number of domains. ( E ) Transcriptional activity of TUs located at domain borders at LBC and post-LBC stages. ( F ) Distribution of insulation scores (see Materials and methods) for domain borders at LBC and post-LBC stages. Lower scores indicate stronger insulation. Despite the absence of “cross” patterns, domain patterns can be identified on the Hi-C maps of late-diplotene oocytes, and the majority of the domain boundaries coincide with TDD boundaries of LBC oocytes. However, post-LBC Hi-C maps contain fewer boundaries, resulting in increase of domain lengths (Fig. 8D ). Visual inspection shows that about half of the neighboring LBC TDDs were fused at the post-LBC stage (44 borders out of 97 examined on chromosome 3, Fig. 8B ). Genome-wide quantification shows that interaction frequency between neighboring LBC domains in post-LBC oocytes are also increased (2% of contacts between neighboring domains in mid-diplotene oocytes versus 6% of contacts for late-diplotene oocytes). Additionally, analysis of the border of TU pairs shows that the activity of weakly expressed TUs decreases, as well as does the insulation (Fig. 8E and F ). Interestingly, despite the absence of the “cross” pattern of actively transcribed TUs, the insulation by the convergently oriented TUs is still present (Fig. 8C ). Cohesive cohesin barrier function is required for transcription-mediated chromatin domain insulation Two mechanistic models can explain why strong reduction in global transcription did not substantially diminish insulation attributed to convergent transcription. (i) Even low transcriptional activity in LBCs suffices to produce insulation via interactions between loop-extruding SMC complexes and RNA polymerase. (ii) Active transcription redistributes cohesive cohesin toward gene termini, where it acts as a barrier to extrusion. Following transcription shutdown, cohesive cohesin presumably remains in the previously occupied locations, providing insulation even in the absence of hypertranscription. To discriminate between these models, we grouped genes with the same orientation (making gene “tandems”) by length and transcriptional activity (Fig. 9A , Materials and methods) and constructed averaged contact maps of chicken oocytes at LBC and post-LBC stages. Model 1 predicts a marked stage-dependent loss of insulation for “weak” tandem pairs but preserved insulation for “strong” tandem pairs. Model 2, in contrast, predicts stronger insulation for long tandem pairs and weak insulation for short tandem gene pairs, with at most a modest overall decrease when transcription is reduced. This pattern arises because longer tandems accumulate more cohesive cohesin at gene termini, thereby forming a more effective barrier to loop extrusion. Under model 2, a global reduction in transcription is expected to attenuate insulation only modestly within each group, since even infrequent transcriptional passages that advance RNA polymerase to gene ends are sufficient to maintain cohesin in a barrier-competent (“locked”) state. Figure 9. Open in a new tab Comparison of the averaged contact maps for convergent gene tandem pairs and computer models at LBC and post-LBC stages. ( A ) Schematic of tandem construction: adjacent co-directional genes are merged into a single tandem unit. Tandem length and “strength” are defined in Materials and methods; transparency encodes transcriptional activity (lower opacity = lower activity). Analyses in this study focus on pairs of neighboring tandems in a convergent orientation. ( B ) Aggregate observed-over-expected (O/E) Hi-C maps for four classes of convergent gene tandem pairs (“long strong,” “long weak,” “short strong,” “short weak”; see Materials and methods) for (upper triangle) and post-LBC (lower triangle) stages. Insulation scores are shown at the upper left of each panel. All panels show aggregated maps for convergently oriented tandem pairs within a ± 500 kb window around their midpoint; resolution: 10 kb. ( C ) Schematic of the one-dimensional simulation framework. ( D–G ) Aggregate O/E maps comparing simulation outputs (upper triangle) versus Hi-C (lower triangle) for LBC (first row) and post-LBC (second row) chromatin; insulation scores as in (A); all panels show aggregated maps for convergently oriented gene pairs within a ± 500 kb window around their midpoint; 10-kb resolution. Parameter regimes (schematics above each panel set): (D) low loop-extruding SMC bypass probability for cohesive cohesin and near-certain bypass for RNA polymerase; (E) low bypass probability for both; (F) near-certain bypass for cohesive cohesin and low for RNA polymerase; (G) near-certain bypass for both . Consistent with Model 2, Fig. 9B shows robust insulation at long convergent tandem gene pairs with only a slight decrease from LBC to post-LBC, whereas short “strong” convergent tandem pairs exhibit weak insulation and short “weak” convergent tandem pairs show little to none. To further test Model 2, we implemented a polymer simulation of a lampbrush-like chromosomal segment (Fig. 9C and Supplementary Fig. S12 ; see Materials and methods) with four interaction regimes between loop-extruding SMC complexes and moving barriers (RNA polymerase and cohesive cohesin): Low loop-extruding SMC bypass probability for cohesive cohesin, but near-certain bypass for RNA polymerase (Fig. 9D ) Low bypass probability for both (Fig. 9E ) Near-certain bypass probability for cohesive cohesin, low for RNA polymerase (Fig. 9F ) Near-certain bypass probability for both (Fig. 9G ) In all regimes, transcribing RNA polymerases push ahead both loop-extruding SMC complexes and cohesive cohesin; initial conditions therefore place cohesive cohesin near tandem ends. The simulations support Model 2: only regimes in which cohesive cohesin functions as a barrier reproduce the post-LBC insulation pattern (Fig. 9D and E ), and accurate recapitulation of the LBC maps further requires that loop-extruding SMC complexes can efficiently bypass RNA polymerase (Fig. 9D ). When loop-extruding SMC rarely bypasses RNA polymerase, the model produces artifactual contact enrichment between convergent gene ends (Fig. 9E and F ). These integrative analyses and simulations demonstrate that cohesive cohesin functions as a transcription-anchored barrier essential for domain insulation in LBCs. The data strongly support Model 2, where active transcription redistributes cohesive cohesin to gene termini (thus accumulating cohesive cohesin at convergent gene pairs), establishing persistent insulation even after transcriptional reduction. Additionally, the consistency of computer Model 2 with lampbrush-chromosome architecture is supported by the close visual agreement between the simulated structures and FISH-mapping data (see Supplementary Note S3 and Supplementary Fig. S13 ). Discussion A unified model of LBCs formation . In many vertebrates, oogenesis is accompanied by hypertranscription aimed to accumulate maternal RNA and leading to significant chromosome decondensation into a lampbrush conformation. Our study offers the first comprehensive, genome-wide analysis of chromatin architecture in chicken oocytes during the LBC stage. Despite LBCs being discovered over a century ago [ 106 , 107 ], our findings provide a new understanding of their architecture and function at unprecedented resolution and across the entire genome. By integrating these results with established knowledge and applying polymer physics, we propose a unified model elucidating the mechanisms underlying lampbrush chromatin organization (Fig. 10 ). Figure 10. Open in a new tab Mechanisms underlying the formation of chromomere–loop complexes in LBCs. Active transcription increases chromatin stiffness and generates pressure that pushes TUs outward from the chromatin axis. Elongating RNA-polymerases remove histones and translocate cohesive cohesin toward the ends of TUs. In the case of convergent transcribed gene pairs, this results in accumulation of cohesive cohesin complexes between TUs, anchoring two transcription loops to the chromosomal axes. In the case of divergent transcribed gene pairs, the depletion of cohesive cohesin rings between TUs results in the detachment of the condensed chromatin knot from the chromatin axes. Loop-extruding SMC complexes active within transcriptionally silent regions form compact chromatin domains located on chromosomal axis. Transcriptionally silent regions separated by divergently oriented TUs can be bridged by extruders moving co-directionally with transcription and/or by intrachromatid entrapment by nonextruding cohesin. However, separation by convergently oriented TUs forms impenetrable barrier for extrusion due to cohesive cohesin accumulation, precluding interactions between flanking regions. These mechanisms underlie the segmentation of LBC into distinct chromomere–loop complexes. Hypertranscription is the primary driver of LBC formation, which determines their chromomere–loop organization [ 57 ]. Here, we highlight several mechanisms linking transcriptional activity to their distinctive organization. First, the accumulation of ribonucleoprotein (RNP) complexes along the transcription loop axis increases chromatin stiffness [ 51 , 92 , 96 ] and generates pressure that pushes TUs outward from the chromosome axis. Second, RNA-polymerases travel along a DNA template as a tightly spaced convoy, displacing DNA-bound proteins from TUs [ 108 , 109 ]. The removal of histones, combined with increased chromatin stiffness, extends the length of transcription loops. Not all proteins, however, can be displaced by RNA polymerase. Cohesive cohesin, which topologically entraps DNA to hold sister chromatids together, most probably resists disassembly [ 110 ]. Instead, RNA polymerase translocates SMC complexes along the DNA [ 110 , 111 , 112 ], functioning as a motor that slides cohesive cohesin toward the ends of TUs [ 110 ]. Similarly, cohesin with extrusion activity is translocated or disassembled during transcription [ 36 , 37 , 38 ]. This creates directional barriers for the extrusion machinery at TU ends, either through direct interactions with the transcription complex or indirectly via accumulated cohesive cohesin, which blocks reloading of dynamic cohesin extruders [ 113 ]. In LBCs, active TUs remain free of cohesive cohesins, leading to local decoupling of sister chromatids. In contrast, downstream regions of TUs enriched in cohesive cohesin remain tethered to the chromosomal axis (Fig. 10 ). A defining feature of LBCs is the presence of transcription loops, both monogenic and multigenic, varying in length [ 57 ]. Integration of Hi-C maps, oocyte nuclear RNA-seq data, and FISH mapping reveals that transcription loops often appear as Hi-C “cross” patterns, characterized by reduced contacts with adjacent regions. This pattern is consistent with loops being histone-free TUs extruded from the chromatin axis due to ribonucleoprotein pressure, increased stiffness, and cohesive cohesin depletion (Fig. 10 ). Transcriptionally inactive regions are enriched with extrusion complexes and cohesive cohesins, leading to their compaction and appearance as contact domains with long-range interactions (Fig. 10 ). FISH analysis shows that transcriptionally inactive regions correspond to intrachromomeric loci (this study and [ 57 , 61 ]). Our model of inactive chromatin organisation within chromomeres is based on previous models of lampbrush chromomere structure that involve cohesive cohesin and condensin complexes in maintaining LBC axes [ 50 , 114 , 115 , 116 ]. In particular, cohesin complex subunits have been revealed within chromomere cores and in interchromomeric regions [ 88 ], while condensin subunit has been found to be enriched within chromomeres [ 103 ]. Moreover, we have taken into account microscopic observations of the rosette structure of discrete lampbrush chromomeres separated by interchromomeric fibre [ 48 ]. Our data reveal how transcription-dependent repositioning of SMC complexes establishes boundaries between chromomere–loop complexes (Fig. 10 ). A single transcribed gene creates a unidirectional barrier at its termination site but does not fully insulate flanking regions, as extrusion complexes can move co-directionally with transcription. Similarly, clusters of closely spaced co-directional genes result in SMC complex accumulation at the end of the last TU, allowing residual interactions between adjacent regions. In contrast, convergent gene pairs concentrate cohesive cohesin complexes between TUs, forming an impenetrable barrier and anchoring two transcription loops to the chromosomal axis (Fig. 10 ). Consistent with our Hi-C and cytological data, convergent gene pairs define boundaries between chromomere–loop complexes. Longer genes in a convergent pair increase the likelihood of at least one cohesive cohesin molecule being translocated to the intergenic region, reinforcing the boundary. Thus, the probability of boundary formation scales with gene length. Unlike intergenic regions between convergent gene ends, those separating divergent gene starts are depleted of cohesive cohesin (Fig. 10 ). Extrusion complexes and histones loaded within these regions mediate their compaction. These regions are tethered to the chromosomal axis if residual cohesive cohesin is present or remain untethered and manifest as a part of lateral loop if it is absent (Fig. 10 ). Consistent with this, we observed small contact domains corresponding to cytologically visible chromatin knots between adjacent TUs. These knots frequently occur when divergent genes are transcribed in opposite directions from a compact chromatin domain [ 57 ], aligning with the described mechanisms. Extrusion complexes loaded between divergent gene pairs can traverse the genes co-directionally with transcription, bridging flanking regions. Thus, a divergent gene pair can form a loop without strong insulation of neighboring loci. In contrast, as discussed above, convergent gene pairs block extrusion complexes due to cohesive cohesin accumulation between them (Fig. 10 ). This mechanism explains the observed enrichment of convergent gene pairs at contact domain boundaries. Simulations based on these mechanisms using a physical chromatin model accurately reproduce both the genome-wide statistical properties of chromatin contact distribution and the specific packaging observed at individual loci. Opened questions . Although the obtained data unequivocally shows that interaction between RNA polymerases and SMC complexes is the major mechanism explaining LBC formation, nuances of this mechanism remain to be elucidated. It is not clear how many extrusion complexes are involved in LBC organization and what is their subunit composition. It is established that the XCAP-D2 (condensin I subunit) and SMC1A, SMC3, Rad21, STAG1, and STAG2 (cohesin subunits) are present on the LBCs [ 88 , 103 ]. However, it remains unclear whether RAD21 functions solely within cohesive cohesin or also as a part of extrusive cohesin. This leaves the possibility of one to four types of extruders contributing, including cohesin-STAG1, cohesin-STAG2, condensin I, and condensin II, each potentially varying in lifetime, density, and velocity. From the observed peak positions of the P (s) slope ( Supplementary Fig. S7A ), we estimate an extruder density of approximately 1 per 100–200 kb in our models. This is comparable to the density of condensin II in interphase or condensin I in prometaphase [ 117 ]. Another open question concerns the probability of an extruder traversing chromatin in the direction opposite to RNA polymerase. Banigan et al. [ 37 ] assumed the permeability of RNA polymerase to SMC with probability ∼0.01 s −1 , but our analysis suggests that this value is too low to reproduce LBC Hi-C data. In LBCs, RNA polymerase complexes are tightly spaced [ 109 ]. Given a polymerase velocity of ∼100 bp/s [ 93 ] and the extruder velocity of 675 bp/s [ 94 ], the overstep probability would be ∼0.15 s −1 . This overstep probability indicates that the extruder would frequently become blocked within the gene body. Therefore, the permeability must exceed 0.15 s −1 for extruders moving against the direction of transcription to pass through the gene body effectively. This increased permeability would allow extruders to traverse TUs with tightly spaced RNA polymerases, aligning with the observed chromatin organization in LBCs. Our model highlights the critical role of cohesive cohesin relocation in response to transcriptional activity, alongside interactions between RNA polymerase and extrusion complexes, in shaping LBC architecture. Interestingly, in yeast, transcription has been shown to promote chromatin loop formation even in the absence of loop extrusion ability in the cohesin variants [ 112 ]. The authors propose that cohesin can facilitate intrachromatid loop formation by topologically entrapping chromatin near transcription sites. Similarly, the hypertranscription characteristic of the LBC stage could contribute to chromatin loop formation through a transcription-dependent mechanism involving DNA loop capture by cohesin, independent of extrusion activity. This suggests that cohesive cohesins may partially supplement the role of extrusion complexes in establishing LBC architecture, emphasizing a combined mechanism of loop extrusion and transcription-dependent loop capture in shaping these remarkable chromatin structures (Fig. 10 ). Another unresolved question is why CTCF-mediated insulation, a hallmark of chromatin organization in typical somatic cells, does not shape LBC contact domains in oocytes of birds in contrast to mouse diplotene oocytes at the NSN stage [ 42 ]. One hypothesis is that alterations in cohesin subunits during LBC formation, such as unknown post-translational modifications, could result in complexes that lose their CTCF-interacting ability. An alternative explanation is that CTCF is unable to bind active chromatin in LBCs due to the high density of RNA polymerase complexes occupying these regions. This transcription-driven exclusion could prevent CTCF from establishing its characteristic insulatory boundaries. Further investigation is required to determine the precise mechanisms underlying this phenomenon. Principles underlying LBC organization are universal across cell types . Although LBCs exhibit a unique type of chromatin organization, the mechanisms underlying their structure are universal and can be observed across various cell types. For instance, transcription loops formed by long, highly expressed genes in somatic interphase nuclei show extensive externalization and reduced cis- contacts, resembling LBC transcription loops [ 92 ]. Similarly, loss of contacts or chromatin “melting” was observed in the case of highly expressed long neuronal genes [ 118 ]. LBC transcription loops also resemble actively transcribed genes in polytene chromosomes, which appear as puffs [ 53 ]. Hi-C maps of Drosophila polytene chromosomes revealed absence of TADs or chromatin contact domains in the puffed regions [ 119 ]. In addition, interbands in polytene chromosomes correspond to inter-TAD regions, which contain clusters of actively transcribed genes with a broad expression pattern [ 120 ] or the promoters and 5′ ends of such genes [ 121 , 122 ]. Thus, loss of chromatin contacts within LBC loops is a common feature of highly transcribed genes in different model species and cell types. Looping of actively transcribed genes, which do not interact with neighboring transcriptionally inactive chromatin but also do not form interactions with active loops from other domains, resembles the local compartments reported in oocytes [ 41 ] and meiotic sperm cells [ 44 ]. Similar attenuation of long-range interactions within compartments has also been observed in Anopheles mosquitoes [ 14 ]. Lack of compartments was also detected in the Drosophila polytene chromosomes [ 119 ]. In all four cases, the chromosomes adopt a rod-like structure, which likely explains the absence of long-range homotopic interactions in these models. Previous studies have demonstrated that RNA polymerase can act as a “moving barrier” to extruding complexes in bacteria [ 35 ], yeast [ 36 ], slime mold Dictyostelium discoideum [ 38 ] and even for vertebrate cohesin [ 37 ]. However, in vertebrate cells, interactions between cohesin and RNA polymerase generally have a minor impact on chromatin architecture, becoming noticeable only in conditions that extend cohesin’s extrusion activity, such as Wapl + CTCF double knockout, and through extensive data aggregation across the genome [ 37 ]. In contrast, our findings suggest that extruder–RNA polymerase interactions as well as cohesive cohesin barrier function play a significantly greater role in the organization of LBCs, explaining the majority of short-range interactions in diplotene oocytes. These results suggest that transcription-mediated repositioning of SMC complexes represents an ancient, conserved mechanism shaping chromatin architecture across the tree of life. Supplementary Material gkag316_Supplemental_Files gkag316_supplemental_files.zip (19.7MB, zip) Acknowledgements This manuscript is dedicated to the memory of Joseph G. Gall. This research was supported in part through computational resources of HPC facilities at the collaborative center “Bioinformatics” ICG SB RAS. We thank the Collective Center of ICG SB RAS, “Collection of Pluripotent Human and Mammalian Cell Cultures for Biological and Biomedical Research” ( https://ckp.icgen.ru/cells/ ), for providing the equipment and facilities used for oocyte isolation. Microscopy analysis was performed using the equipment of the Resource Center “Molecular and Cell Technologies” (Saint-Petersburg State University). We acknowledge the collective centers of the Ministry of Science and Higher Education of the Russian Federation (state project FWNR-2022-0019, continued as FWNR-2026-0032) for providing facilities for genomic libraries preparation. The original manuscript text was composed by the authors; proofreading was conducted with the assistance of ChatGPT version 4, after that the text was corrected and edited by the authors. Author contributions : V.F. and A.K. conceptualized the study. M.G. conducted the genomics experiments with assistance from A.K., T.K., and A.M., who contributed to oocyte isolation, and A.N., who provided support for staining and NOMe-seq experiments. Cytological analyses, including FISH-mapping experiments, were performed by T.K. and A.K. with contribution from A.M., who performed probe labeling. Bioinformatic analyses were carried out by T.L., with contributions from V.F., A.P., A.N., and V.K. T.L. developed the methodology for polymer simulations and conducted the modeling experiments with input from V.K. T.L. and V.F. drafted the manuscript with contributions by A.K.; authors jointly developed mechanism explaining LBC organization, and A.K. suggested the final scheme of chromomere–loop complex formation. All authors reviewed, revised, and approved the final manuscript. Contributor Information Timofey Lagunov, Institute of Cytology and Genetics, Novosibirsk, Russia; Novosibirsk State University, Novosibirsk, Russia. Maria Gridina, Institute of Cytology and Genetics, Novosibirsk, Russia; Novosibirsk State University, Novosibirsk, Russia; Sirius University of Science and Technology, Sirius, Russia. Artem Nurislamov, Institute of Cytology and Genetics, Novosibirsk, Russia; Novosibirsk State University, Novosibirsk, Russia; Sirius University of Science and Technology, Sirius, Russia. Tatiana Kulikova, Laboratory of Cell Nucleus Structure and Dynamics, Saint-Petersburg State University, Saint-Petersburg, Russia. Antonina Maslova, Laboratory of Cell Nucleus Structure and Dynamics, Saint-Petersburg State University, Saint-Petersburg, Russia. Viktor Konstantinov, Institute of Cytology and Genetics, Novosibirsk, Russia; Novosibirsk State University, Novosibirsk, Russia. Andrey Popov, Institute of Cytology and Genetics, Novosibirsk, Russia; Sirius University of Science and Technology, Sirius, Russia; Artificial Intelligence Research Institute, Moscow, Russia. Alla Krasikova, Laboratory of Cell Nucleus Structure and Dynamics, Saint-Petersburg State University, Saint-Petersburg, Russia. Veniamin Fishman, Institute of Cytology and Genetics, Novosibirsk, Russia; Novosibirsk State University, Novosibirsk, Russia; Sirius University of Science and Technology, Sirius, Russia; Artificial Intelligence Research Institute, Moscow, Russia. Supplementary data Supplementary data is available at NAR online. Conflict of interest None declared. Funding This study was initiated with support of RSF Project # 20-64-46021 in 2020-2022 (most of the genomic and cytological experiments). Access to computational resources was provided by Novosibirsk State University, supported by the Ministry of Education and Science of the Russian Federation (FSUS-2024-0018). Part of the bioinformatic analysis (including LBC 3D-modelling and genome annotation) and manuscript preparation was supported by grant of the state program of the “Sirius” Federal Territory “Scientific and technological development of the “Sirius” Federal Territory” (agreement # 26-03, 27/09/2024). Data availability All raw sequencing data and proceed data are available under GEO accession GSE288415 . The program code used for LBC model simulation is freely and publicly available at https://github.com/genomech/LBC_model and https://doi.org/10.5281/zenodo.18998039 . Any additional data in support of the findings of this study can be obtained from the corresponding authors upon reasonable request. References 1. Kabirova  E, Nurislamov  A, Shadskiy  A  et al.  Function and evolution of the loop extrusion machinery in animals. Int J Mol Sci. 2023;24:5017. 10.3390/ijms24055017. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 2. Misteli  T. The self-organizing genome: principles of genome architecture and function. Cell. 2020;183:28–45. 10.1016/j.cell.2020.09.014. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 3. 4D Nucleome Consortium, Dekker  J, Oksuz  BA, Zhang  Y  et al.  An integrated view of the structure and function of the human 4D nucleome. bioRxiv, 10.1101/2024.09.17.613111, 27 October 2024, preprint: not peer reviewed. [ DOI ] 4. Kabirova  E, Ryzhkova  A, Lukyanchikova  V  et al.  TAD border deletion at the kit locus causes tissue-specific ectopic activation of a neighboring gene. Nat Commun. 2024;15:4521. 10.1038/s41467-024-48523-7. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 5. Salnikov  P, Korablev  A, Serova  I  et al.  Structural variants in the Epb41l4a locus: TAD disruption and Nrep gene misregulation as hypothetical drivers of neurodevelopmental outcomes. Sci Rep. 2024;14:5288. 10.1038/s41598-024-52545-y. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 6. Rajderkar  S, Barozzi  I, Zhu  Y  et al.  Topologically associating domain boundaries are required for normal genome function. Commun Biol. 2023;6:435. 10.1038/s42003-023-04819-w. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 7. Lupiáñez  DG, Kraft  K, Heinrich  V  et al.  Disruptions of topological chromatin domains cause pathogenic rewiring of gene–enhancer interactions. Cell. 2015;161:1012–25. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 8. Pope  BD, Ryba  T, Dileep  V  et al.  Topologically associating domains are stable units of replication-timing regulation. Nature. 2014;515:402–5. 10.1038/nature13986. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 9. Klein  KN, Zhao  PA, Lyu  X  et al.  Replication timing maintains the global epigenetic state in human cells. Science. 2021;372:371–8. 10.1126/science.aba5545. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 10. Moindrot  B, Audit  B, Klous  P  et al.  3D chromatin conformation correlates with replication timing and is conserved in resting cells. Nucleic Acids Res. 2012;40:9470–81. 10.1093/nar/gks736. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 11. Piazza  A, Bordelet  H, Dumont  A  et al.  Cohesin regulates homology search during recombinational DNA repair. Nat Cell Biol. 2021;23:1176–86. 10.1038/s41556-021-00783-x. [ DOI ] [ PubMed ] [ Google Scholar ] 12. Strom  AR, Biggs  RJ, Banigan  EJ  et al.  HP1α is a chromatin crosslinker that controls nuclear and mitotic chromosome mechanics. eLife. 2021;10:e63972. 10.7554/eLife.63972. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 13. Uhler  C, Shivashankar  GV. Regulation of genome organization and gene expression by nuclear mechanotransduction. Nat Rev Mol Cell Biol. 2017;18:717–27. 10.1038/nrm.2017.101. [ DOI ] [ PubMed ] [ Google Scholar ] 14. Lukyanchikova  V, Nuriddinov  M, Belokopytova  P  et al.  Anopheles mosquitoes reveal new principles of 3D genome organization in insects. Nat Commun. 2022;13:1960. 10.1038/s41467-022-29599-5. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 15. Fishman  V, Battulin  N, Nuriddinov  M  et al.  3D organization of chicken genome demonstrates evolutionary conservation of topologically associated domains and highlights unique architecture of erythrocytes’ chromatin. Nucleic Acids Res. 2019;47:648–65. 10.1093/nar/gky1103. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 16. Krefting  J, Andrade-Navarro  MA, Ibn-Salem  J. Evolutionary stability of topologically associating domains is associated with conserved gene regulation. BMC Biol. 2018;16:87. 10.1186/s12915-018-0556-x. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 17. Hoencamp  C, Dudchenko  O, Elbatsh  AMO  et al.  3D genomics across the tree of life reveals condensin II as a determinant of architecture type. Science. 2021;372:984–9. 10.1126/science.abe2218. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 18. Polovnikov  KE, Brandão  HB, Belan  S  et al.  Crumpled polymer with loops recapitulates key features of chromosome organization. Phys Rev X. 2023;13:041029. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 19. Mirny  LA. The fractal globule as a model of chromatin architecture in the cell. Chromosome Res. 2011;19:37–51. 10.1007/s10577-010-9177-0. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 20. Razin  SV, Gavrilov  AA. The role of liquid–liquid phase separation in the compartmentalization of cell nucleus and spatial genome organization. Biochem Moscow. 2020;85:643–50. 10.1134/S0006297920060012. [ DOI ] [ Google Scholar ] 21. Rada-Iglesias  A, Grosveld  FG, Papantonis  A. Forces driving the three-dimensional folding of eukaryotic genomes. Mol Syst Biol. 2018;14:e8214. 10.15252/msb.20188214. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 22. Strom  AR, Brangwynne  CP. The liquid nucleome—phase transitions in the nucleus at a glance. J Cell Sci. 2019;132:jcs235093. 10.1242/jcs.235093. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 23. Cremer  T, Cremer  M, Hübner  B  et al.  The interchromatin compartment participates in the structural and functional organization of the cell nucleus. Bioessays. 2020;42:1900132. 10.1002/bies.201900132. [ DOI ] [ Google Scholar ] 24. Li  H, Playter  C, Das  P  et al.  Chromosome compartmentalization: causes, changes, consequences, and conundrums. Trends Cell Biol. 2024;34:707–27. 10.1016/j.tcb.2024.01.009. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 25. Padeken  J, Heun  P. Nucleolus and nuclear periphery: velcro for heterochromatin. Curr Opin Cell Biol. 2014;28:54–60. 10.1016/j.ceb.2014.03.001. [ DOI ] [ PubMed ] [ Google Scholar ] 26. Quinodoz  SA, Ollikainen  N, Tabak  B  et al.  Higher-order inter-chromosomal hubs shape 3D genome organization in the nucleus. Cell. 2018;174:744–57. 10.1016/j.cell.2018.05.024. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 27. Solovei  I, Thanisch  K, Feodorova  Y. How to rule the nucleus: divide et impera. Curr Opin Cell Biol. 2016;40:47–59. 10.1016/j.ceb.2016.02.014. [ DOI ] [ PubMed ] [ Google Scholar ] 28. Dekker  J, Mirny  LA. The chromosome folding problem and how cells solve it. Cell. 2024;187:6424–50. 10.1016/j.cell.2024.10.026. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 29. Davidson  IF, Peters  J-M. Genome folding through loop extrusion by SMC complexes. Nat Rev Mol Cell Biol. 2021;22:445–64. 10.1038/s41580-021-00349-7. [ DOI ] [ PubMed ] [ Google Scholar ] 30. Szabo  Q, Bantignies  F, Cavalli  G. Principles of genome folding into topologically associating domains. Sci Adv. 2019;5:eaaw1668. 10.1126/sciadv.aaw1668. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 31. Wang  J, Ma  Q, Fang  P  et al.  Manipulation of TAD reorganization by chemical-dependent genome linking. STAR Protoc. 2021;2:100799. 10.1016/j.xpro.2021.100799. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 32. Dixon  JR, Selvaraj  S, Yue  F  et al.  Topological domains in mammalian genomes identified by analysis of chromatin interactions. Nature. 2012;485:376–80. 10.1038/nature11082. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 33. Nora  EP, Lajoie  BR, Schulz  EG  et al.  Spatial partitioning of the regulatory landscape of the X-inactivation centre. Nature. 2012;485:381–5. 10.1038/nature11049. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 34. Dequeker  BJH, Scherr  MJ, Brandão  HB  et al.  MCM complexes are barriers that restrict cohesin-mediated loop extrusion. Nature. 2022;606:197–203. 10.1038/s41586-022-04730-0. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 35. Brandão  HB, Paul  P, Van Den Berg  AA  et al.  RNA polymerases as moving barriers to condensin loop extrusion. Proc Natl Acad Sci USA. 2019;116:20489–99. 10.1073/pnas.1907009116. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 36. Rivosecchi  J, Jost  D, Vachez  L  et al.  RNA polymerase backtracking results in the accumulation of fission yeast condensin at active genes. Life Sci Alliance. 2021;4:e202101046. 10.26508/lsa.202101046. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 37. Banigan  EJ, Tang  W, Van Den Berg  AA  et al.  Transcription shapes 3D chromatin organization by interacting with loop extrusion. Proc Natl Acad Sci USA. 2023;120:e2210480120. 10.1073/pnas.2210480120. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 38. Zhegalova  IV, Ulianov  SV, Galitsyna  AA  et al.  Convergent pairs of highly transcribed genes restrict chromatin looping in Dictyostelium discoideum . Nucleic Acids Res. 2024;53:gkaf006. 10.1093/nar/gkaf006. [ DOI ] [ Google Scholar ] 39. Belokopytova  P, Fishman  V. Predicting genome architecture: challenges and solutions. Front Genet. 2021;11:617202. 10.3389/fgene.2020.617202. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 40. Rowley  MJ, Corces  VG. Organizational principles of 3D genome architecture. Nat Rev Genet. 2018;19:789–800. 10.1038/s41576-018-0060-8. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 41. Du  Z, Zheng  H, Kawamura  YK  et al.  Polycomb group proteins regulate chromatin architecture in mouse oocytes and early embryos. Mol Cell. 2020;77:825–39. 10.1016/j.molcel.2019.11.011. [ DOI ] [ PubMed ] [ Google Scholar ] 42. Flyamer  IM, Gassler  J, Imakaev  M  et al.  Single-nucleus Hi-C reveals unique chromatin reorganization at oocyte-to-zygote transition. Nature. 2017;544:110–4. 10.1038/nature21711. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 43. Wang  Y, Wang  H, Zhang  Y  et al.  Reprogramming of meiotic chromatin architecture during spermatogenesis. Mol Cell. 2019;73:547–61. 10.1016/j.molcel.2018.11.019. [ DOI ] [ PubMed ] [ Google Scholar ] 44. Cheng  G, Pratto  F, Brick  K  et al.  High resolution maps of chromatin reorganization through mouse meiosis reveal novel features of the 3D meiotic structure. bioRxiv, 10.1101/2024.03.25.586627, 28 March 2024, preprint: not peer reviewed. [ DOI ] 45. Krasikova  A, Khodyuchenko  T, Maslova  A  et al.  Three-dimensional organisation of RNA-processing machinery in avian growing oocyte nucleus. Chromosome Res. 2012;20:979–94. 10.1007/s10577-012-9327-7. [ DOI ] [ PubMed ] [ Google Scholar ] 46. Maslova  AV, Krasikova  AV. Spatial arrangement of macro-, midi-, and microchromosomes in transcriptionally active nuclei of growing oocytes in birds of the order galliformes. Cell Tiss Biol. 2011;5:281–93. 10.1134/S1990519X11030072. [ DOI ] [ Google Scholar ] 47. Vlad  M, Macgregor  HC. Chromomere number and its genetic significance in lampbrush chromosomes. Chromosoma. 1975;50:327–47. 10.1007/BF00327073. [ DOI ] [ PubMed ] [ Google Scholar ] 48. Macgregor  HC. Chromomeres revisited. Chromosome Res. 2012;20:911–24. 10.1007/s10577-012-9310-3. [ DOI ] [ PubMed ] [ Google Scholar ] 49. Krasikova  A, Fishman  V, Kulikova  T. Lampbrush chromosome studies in the post-genomic era. BioEssays. 2023;45:2200250. 10.1002/bies.202200250. [ DOI ] [ Google Scholar ] 50. Gall  JG. Transcription in the Xenopus oocyte nucleus. In: Kloc  M, Kubiak  JZ (eds.), Xenopus Development. Hoboken, New Jersey: Wiley, 2014, 3–15. 10.1002/9781118492833. [ DOI ] [ Google Scholar ] 51. Morgan  GT. Imaging the dynamics of transcription loops in living chromosomes. Chromosoma. 2018;127:361–74. 10.1007/s00412-018-0667-8. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 52. Kaufmann  R, Cremer  C, Gall  JG. Superresolution imaging of transcription units on newt lampbrush chromosomes. Chromosome Res. 2012;20:1009–15. 10.1007/s10577-012-9306-z. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 53. Mirny  LA, Solovei  I. Keeping chromatin in the loop(s). Nat Rev Mol Cell Biol. 2021;22:439–40. 10.1038/s41580-021-00337-x. [ DOI ] [ PubMed ] [ Google Scholar ] 54. Keinath  MC, Davidian  A, Timoshevskiy  V  et al.  Characterization of axolotl lampbrush chromosomes by fluorescence in situ hybridization and immunostaining. Exp Cell Res. 2021;401:112523. 10.1016/j.yexcr.2021.112523. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 55. Kulikova  T, Maslova  A, Starshova  P  et al.  Comparison of the somatic TADs and lampbrush chromomere–loop complexes in transcriptionally active prophase I oocytes. Chromosoma. 2022;131:207–23. 10.1007/s00412-022-00780-5. [ DOI ] [ PubMed ] [ Google Scholar ] 56. Kulikova  T, Krasikova  A. RNA-FISH—On lampbrush chromosomes: visualization of individual transcription units. In: Cytogenetics and Molecular Cytogenetics. Boca Raton: CRC Press, 2022, 307–17. 10.1201/9781003223658. [ DOI ] [ Google Scholar ] 57. Krasikova  A, Kulikova  T, Schelkunov  M  et al.  The first chicken oocyte nucleus whole transcriptomic profile defines the spectrum of maternal mRNA and non-coding RNA genes transcribed by the lampbrush chromosomes. Nucleic Acids Res. 2024;52:12850–77. 10.1093/nar/gkae941. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 58. Zlotina  A, Maslova  A, Pavlova  O  et al.  New insights into chromomere organization provided by lampbrush chromosome microdissection and high-throughput sequencing. Front. Genet.  2020;11:57. 10.3389/fgene.2020.00057. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 59. Zlotina  A, Kulikova  T, Kosyakova  N  et al.  Microdissection of lampbrush chromosomes as an approach for generation of locus-specific FISH-probes and samples for high-throughput sequencing. Bmc Genomics [Electronic Resource]. 2016;17:126. 10.1186/s12864-016-2437-4. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 60. Kulikova  T, Surkova  A, Zlotina  A  et al.  Mapping epigenetic modifications on chicken lampbrush chromosomes. Mol Cytogenet. 2020;13:32. 10.1186/s13039-020-00496-0. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 61. Krasikova  A, Kulikova  T, Rodriguez Ramos  JS  et al.  Assignment of the somatic A/B compartments to chromatin domains in giant transcriptionally active lampbrush chromosomes. Epigenetics Chromatin. 2023;16:24. 10.1186/s13072-023-00499-2. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 62. Krasikova  AV, Kulikova  TV. Distribution of heterochromatin markers in lampbrush chromosomes in birds. Russ J Genet. 2017;53:1022–9. 10.1134/S1022795417090071. [ DOI ] [ Google Scholar ] 63. Kim  Y-K, Ramalho-Santos  M. 20 years of stemness: from stem cells to hypertranscription and back. Stem Cell Rep. 2025;20:102406. 10.1016/j.stemcr.2025.102406. [ DOI ] [ Google Scholar ] 64. Yu  G, Xu  K, Xia  W  et al.  Establishment of chromatin architecture interplays with embryo hypertranscription. Nature. 2025;646:208–17. 10.1038/s41586-025-09400-5. [ DOI ] [ PubMed ] [ Google Scholar ] 65. Guide for the care and use of laboratory animals: eighth edition . Washington, D.C: National Academies Press, 2011. [ Google Scholar ] 66. Martin  M. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet j. 2011;17:10. 10.14806/ej.17.1.200. [ DOI ] [ Google Scholar ] 67. Durand  NC, Shamim  MS, Machol  I  et al.  Juicer provides a one-click system for analyzing loop-resolution hi-C experiments. Cell Syst. 2016;3:95–8. 10.1016/j.cels.2016.07.002. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 68. Huang  Z, Xu  Z, Bai  H  et al.  Evolutionary analysis of a complete chicken genome. Proc Natl Acad Sci USA. 2023;120:e2216641120. 10.1073/pnas.2216641120. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 69. Gridina  M, Mozheiko  E, Valeev  E  et al.  A cookbook for DNase Hi-C. Epigenetics Chromatin. 2021;14:15. 10.1186/s13072-021-00389-5. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 70. Abdennur  N, Mirny  LA. Cooler: scalable storage for Hi-C data and other genomically labeled arrays. Bioinformatics. 2020;36:311–6. 10.1093/bioinformatics/btz540. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 71. Flyamer  IM, Illingworth  RS, Bickmore  WA. Coolpup.Py : versatile pile-up analysis of hi-C data. Bioinformatics. 2020;36:2980–5. 10.1093/bioinformatics/btaa073. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 72. Open2C, Abdennur  N, Abraham  S, Fudenberg  G  et al.  Cooltools: enabling high-resolution Hi-C analysis in Python. bioRxiv, 10.1101/2022.10.31.514564, 1 November 2022, preprint: not peer reviewed. [ DOI ] 73. Jerković  I, Ibrahim  DM, Andrey  G  et al.  Genome-wide binding of posterior HOXA/D transcription factors reveals subgrouping and association with CTCF. PLoS Genet. 2017;13:e1006567. 10.1371/journal.pgen.1006567. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 74. Krueger  F, Kreck  B, Franke  A  et al.  DNA methylome analysis using short bisulfite sequencing data. Nat Methods. 2012;9:145–51. 10.1038/nmeth.1828. [ DOI ] [ PubMed ] [ Google Scholar ] 75. Krueger  F, Andrews  SR. Bismark: a flexible aligner and methylation caller for Bisulfite-Seq applications. Bioinformatics. 2011;27:1571–2. 10.1093/bioinformatics/btr167. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 76. Ferragina  P, Manzini  G. Opportunistic data structures with applications. In: Proceedings 41st Annual Symposium on Foundations of Computer Science. Redondo Beach, CA, USA: IEEE Comput. Soc, 2000, 390–8. 10.1109/SFCS.2000.892127. [ DOI ] [ Google Scholar ] 77. Langmead  B, Salzberg  SL. Fast gapped-read alignment with Bowtie 2. Nat Methods. 2012;9:357–9. 10.1038/nmeth.1923. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 78. Langmead  B, Trapnell  C, Pop  M  et al.  Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol. 2009;10:R25. 10.1186/gb-2009-10-3-r25. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 79. Langmead  B, Wilks  C, Antonescu  V  et al.  Scaling read aligners to hundreds of threads on general-purpose processors. Bioinformatics. 2019;35:421–32. 10.1093/bioinformatics/bty648. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 80. Nurislamov  A, Lagunov  T, Gridina  M  et al.  Single-cell DNA methylation analysis of chicken lampbrush chromosomes. Int J Mol Sci. 2022;23:12601. 10.3390/ijms232012601. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 81. Kim  D, Paggi  JM, Park  C  et al.  Graph-based genome alignment and genotyping with HISAT2 and HISAT-genotype. Nat Biotechnol. 2019;37:907–15. 10.1038/s41587-019-0201-4. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 82. Danecek  P, Bonfield  JK, Liddle  J  et al.  Twelve years of SAMtools and BCFtools. GigaScience. 2021;10:giab008. 10.1093/gigascience/giab008. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 83. Shumate  A, Wong  B, Pertea  G  et al.  Improved transcriptome assembly using a hybrid of long and short reads with StringTie. PLoS Comput Biol. 2022;18:e1009730. 10.1371/journal.pcbi.1009730. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 84. Li  D, He  M, Tang  Q  et al.  Comparative 3D genome architecture in vertebrates. BMC Biol. 2022;20:99. 10.1186/s12915-022-01301-7. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 85. Solovei  I, Gaginskaya  E, Allen  T  et al.  A novel structure associated with a lampbrush chromosome in the chicken, Gallus domesticus . J Cell Sci. 1992;101:759–72. 10.1242/jcs.101.4.759. [ DOI ] [ PubMed ] [ Google Scholar ] 86. Cremer  M, Grasser  F, Lanctôt  C  et al.  Multicolor 3D fluorescence in situ hybridization for imaging interphase chromosomes. In: Hancock  R (ed.), The Nucleus, Methods in Molecular Biology. Totowa, NJ: Humana Press, Vol. 463, 2012, pp. 205–39. [ Google Scholar ] 87. Zlotina  A, Krasikova  A. FISH in lampbrush chromosomes. In: Liehr  T (ed.), Fluorescence in Situ Hybridization (FISH), Springer Protocols Handbooks. Berlin, Heidelberg: Springer Berlin Heidelberg, 2017, pp. 445–57. [ Google Scholar ] 88. Krasikova  A, Barbero  JL, Gaginskaya  E. Cohesion proteins are present in centromere protein bodies associated with avian lampbrush chromosomes. Chromosome Res. 2005;13:675–85. 10.1007/s10577-005-1005-6. [ DOI ] [ PubMed ] [ Google Scholar ] 89. Krasikova  A, Kulikova  T, Saifitdinova  A  et al.  Centromeric protein bodies on avian lampbrush chromosomes contain a protein detectable with an antibody against DNA topoisomerase II. Chromosoma. 2004;113:316–23. 10.1007/s00412-004-0321-5. [ DOI ] [ PubMed ] [ Google Scholar ] 90. Imakaev  M, Goloborodko  A, Hbbrandao. mirnylab/polychrom: v0.1.0.. Zenodo. 2019. 10.5281/zenodo.3579473. [ DOI ] 91. Pradhan  B, Barth  R, Kim  E  et al.  SMC complexes can traverse physical roadblocks bigger than their ring size. Cell Rep. 2022;41:111491. 10.1016/j.celrep.2022.111491. [ DOI ] [ PubMed ] [ Google Scholar ] 92. Leidescher  S, Ribisel  J, Ullrich  S  et al.  Spatial organization of transcribed eukaryotic genes. Nat Cell Biol. 2022;24:327–39. 10.1038/s41556-022-00847-6. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 93. Wang  MD, Schnitzer  MJ, Yin  H  et al.  Force and velocity measured for single molecules of RNA polymerase. Science. 1998;282:902–7. 10.1126/science.282.5390.902. [ DOI ] [ PubMed ] [ Google Scholar ] 94. Kong  M, Cutts  EE, Pan  D  et al.  Human condensin I and II drive extensive ATP-dependent compaction of nucleosome-bound DNA. Mol Cell. 2020;79:99–114. 10.1016/j.molcel.2020.04.026. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 95. Ryu  J-K, Rah  S-H, Janissen  R  et al.  Condensin extrudes DNA loops in steps up to hundreds of base pairs that are generated by ATP binding events. Nucleic Acids Res. 2022;50:820–32. 10.1093/nar/gkab1268. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 96. Marko  JF, Siggia  ED. Polymer models of meiotic and mitotic chromosomes. MBoC. 1997;8:2217–31. 10.1091/mbc.8.11.2217. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 97. Plotly Technologies Inc . Collaborative Data Science. 2015. [ Google Scholar ] 98. Gridina  M, Taskina  A, Lagunov  T  et al.  Comparison and critical assessment of single-cell Hi-C protocols. Heliyon. 2022;8:e11023. 10.1016/j.heliyon.2022.e11023. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 99. Galitsyna  AA, Gelfand  MS. Single-cell Hi-C data analysis: safety in numbers. Brief Bioinf. 2021;22:bbab316. 10.1093/bib/bbab316. [ DOI ] [ Google Scholar ] 100. Akgol Oksuz  B, Yang  L, Abraham  S  et al.  Systematic evaluation of chromosome conformation capture assays. Nat Methods. 2021;18:1046–55. 10.1038/s41592-021-01248-7. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 101. Rao  SSP, Huntley  MH, Durand  NC  et al.  A 3D map of the human genome at kilobase resolution reveals principles of chromatin looping. Cell. 2014;159:1665–80. 10.1016/j.cell.2014.11.021. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 102. Zlotina  A, Galkina  S, Krasikova  A  et al.  Centromere positions in chicken and Japanese quail chromosomes: de novo centromere formation versus pericentric inversions. Chromosome Res. 2012;20:1017–32. 10.1007/s10577-012-9319-7. [ DOI ] [ PubMed ] [ Google Scholar ] 103. Beenders  B, Watrin  E, Legagneux  V  et al.  Distribution of XCAP-E and XCAP-D2 in the Xenopus oocyte nucleus. Chromosome Res. 2003;11:549–64. 10.1023/A:1024999316867. [ DOI ] [ PubMed ] [ Google Scholar ] 104. Degalez  F, Charles  M, Foissac  S  et al.  Enriched atlas of lncRNA and protein-coding genes for the GRCg7b chicken assembly and its functional annotation across 47 tissues. Sci Rep. 2024;14:6588. 10.1038/s41598-024-56705-y. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 105. Maslova  A, Krasikova  A. Nuclear actin depolymerization in transcriptionally active avian and amphibian oocytes leads to collapse of intranuclear structures. Nucleus. 2012;3:300–11. 10.4161/nucl.20393. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 106. Flemming  W. 1882; Zellsubstanz, kern und zelltheilung Vogel. 107. Rückert  J. 1892; Zur entwickelungsgeschichte des ovarialeies bei selachiern. 108. Morgan  GT, Jones  P, Bellini  M. Association of modified cytosines and the methylated DNA-binding protein MeCP2 with distinctive structural domains of lampbrush chromatin. Chromosome Res. 2012;20:925–42. 10.1007/s10577-012-9324-x. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 109. Franke  WW, Scheer  U, Trendelenburg  MF  et al.  Absence of nucleosomes in transcriptionally active chromatin. Cytobiologie. 1976;13:401–34. [ Google Scholar ] 110. Ocampo-Hafalla  M, Muñoz  S, Samora  CP  et al.  Evidence for cohesin sliding along budding yeast chromosomes. Open Biol. 2016;6:150178. 10.1098/rsob.150178. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 111. Davidson  IF, Goetz  D, Zaczek  MP  et al.  Rapid movement and transcriptional re-localization of human cohesin on DNA. EMBO J. 2016;35:2671–85. 10.15252/embj.201695402. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 112. Guérin  TM, Barrington  C, Pobegalov  G  et al.  An extrinsic motor directs chromatin loop formation by cohesin. EMBO J. 2024;43:4173–96. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 113. Bastié  N, Chapard  C, Cournac  A  et al.  Sister chromatid cohesion halts DNA loop expansion. Mol Cell. 2024;84:1139–48. [ DOI ] [ PubMed ] [ Google Scholar ] 114. Gaginskaya  E, Kulikova  T, Krasikova  A. Avian lampbrush chromosomes: a powerful tool for exploration of genome expression. Cytogenet Genome Res. 2009;124:251–67. 10.1159/000218130. [ DOI ] [ PubMed ] [ Google Scholar ] 115. Krasikova  A, Kulikova  T. Hromosomy tipa lampovyh shchetok: sovremennye predstavleniya i perspektivy issledovanij [Lampbrush chromosomes: current insights and research perspectives]. St Petersburg: St Petersburg University Publishing House, 2019. [ Google Scholar ] 116. Krasikova  A. Centromernye rajony hromosom i associirovannye s nimi struktury v yadrah rastushchih oocitov ptic [centromeric regions of chromosomes and associated structures in the nuclei of growing avian oocytes.]. Ph.D. Thesis. 2007. Saint-Petersburg State University. [ Google Scholar ] 117. Samejima  K, Gibcus  JH, Abraham  S  et al.  Rules of engagement for condensins and cohesins guide mitotic chromosome formation. Science. 2025. 10.1126/science.adq1709. [ DOI ] 118. Winick-Ng  W, Kukalev  A, Harabula  I  et al.  Cell-type specialization is encoded by specific chromatin topologies. Nature. 2021;599:684–91. 10.1038/s41586-021-04081-2. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 119. Eagen  KP, Hartl  TA, Kornberg  RD. Stable chromosome condensation revealed by chromosome conformation capture. Cell. 2015;163:934–46. 10.1016/j.cell.2015.10.026. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 120. Ulianov  SV, Khrameeva  EE, Gavrilov  AA  et al.  Active chromatin and transcription play a key role in chromosome partitioning into topologically associating domains. Genome Res. 2016;26:70–84. 10.1101/gr.196006.115. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 121. Stadler  MR, Haines  JE, Eisen  MB. Convergence of topological domain boundaries, insulators, and polytene interbands revealed by high-resolution mapping of chromatin contacts in the early Drosophila melanogaster embryo. eLife. 2017;6:e29550. 10.7554/eLife.29550. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 122. Kolesnikova  TD. Banding pattern of polytene chromosomes as a representation of universal principles of chromatin organization into topological domains. Biochem Moscow. 2018;83:338–49. 10.1134/S0006297918040053. [ DOI ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Data Citations Imakaev  M, Goloborodko  A, Hbbrandao. mirnylab/polychrom: v0.1.0.. Zenodo. 2019. 10.5281/zenodo.3579473. [ DOI ] Supplementary Materials gkag316_Supplemental_Files gkag316_supplemental_files.zip (19.7MB, zip) Data Availability Statement All raw sequencing data and proceed data are available under GEO accession GSE288415 . The program code used for LBC model simulation is freely and publicly available at https://github.com/genomech/LBC_model and https://doi.org/10.5281/zenodo.18998039 . Any additional data in support of the findings of this study can be obtained from the corresponding authors upon reasonable request. Articles from Nucleic Acids Research are provided here courtesy of Oxford University Press ACTIONS View on publisher site PDF (9.7 MB) Cite Collections Permalink PERMALINK Copy RESOURCES Similar articles Cited by other articles Links to NCBI Databases Cite Copy Download .nbib .nbib Format: AMA APA MLA NLM Add to Collections Create a new collection Add to an existing collection Name your collection * Choose a collection Unable to load your collection due to an error Please try again Add Cancel Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top