Confined migration induces non-lethal DNA damage in developing neurons

Lists of the antibodies and reagents used are provided in Supplementary Tables 1 and 2.

Plasmids

The pCAG-mScarlet plasmid was generated using the mScarlet sequence obtained from Addgene (85042; a gift from D. Gadella). To generate the pCAG-53BP1-mNeonGreen construct, mouse 53BP1 cDNA was amplified from a mouse brain cDNA library and inserted into the pENTR1A vector. In parallel, the Gateway cassette from pDest-eGFP-N1 was inserted into the EcoRI site of pmNeonGreen using In-Fusion Cloning to generate pDest-mNeonGreen. The 53BP1 cDNA in pENTR1A was recombined into pCAG-Dest-mNeonGreen using LR Clonase II Enzyme Mix (Invitrogen, 11791020) to produce the final pCAG-53BP1-mNeonGreen construct. The pTRIP-CMV-GFP-Flag-cGAS plasmid was obtained from Addgene (86674; a gift from N. Manel). mCherry-KASH1 was constructed as previously decribed¹³. The pAAV-Neurod1-GFP plasmid was constructed by replacing the CAG promoter in pAAV-CAG-EGFP⁵¹ with the Neurod1 promoter, which was cloned from the cDNA of CGNs isolated from P6 ICR mice.

Animals

All animal experiments were approved by the Animal Experiment Committee of Kyoto University (KUIAS 1-9) and conducted in accordance with the guidelines of the National Centre for the Replacement, Refinement and Reduction of Animals in Research. For some in vitro experiments, timed-pregnant C57BL/6J and ICR mice were purchased from Japan SLC. We generated conditional deletion mice by breeding Lig4^flox/flox mice³⁰ with Neurod1-cre (Tg (Neurod1-cre) RZ24Gsat/Mmucd, MMRRC_036320-UCD) or Gabra6-cre (B6.129P2-Gabra6^tm2(cre)Wwis/Mmucd, MMRRC_015968-UCD) mice from MMRRC at UC Davis. The Neurod1-GFP (Tg (Neurod1-eGFP) CR99Gsat/Mmucd, MMRRC_000329-UCD) mouse strain was also obtained from MMRRC. These lines were maintained on a predominantly C57BL/6J mixed background, with mice of both sexes used for experiments. Animals were housed under a 12 h–12 h light–dark cycle at 23 ± 3 °C and 50% humidity.

Genotyping

Tail or nail clippings were digested in 50 mM NaOH at 95 °C for 10 min. After neutralizing by adding 1 M Tris-HCl (pH 8.0), samples were centrifuged at 4 °C for 15 min. The supernatant was subjected to PCR using the following primers: Lig4^flox allele, 5′-GAGCTGCAACAGTTTGTGAAGTTTGTGAGGA-3′ and 5′-GTGTTGGTCAGGACCAGAAGGAAAGCA-3′; Neurod1-cre allele, 5′-TAGGATTAGGGAGAGGGAGCTGAA-3′ and 5′-CGGCAAACGGACAGAAGC-3′; Neurod1-GFP allele, 5′-TAGGATTAGGGAGAGGGAGCTGAA-3′ and 5′-Gabra6-cre allele, 5′-GATCTCCGGTATTGAAACTCCAGC-3′ and 5′-GCTAAACAT GCTTCATCGTCGG-3′.

In vivo and in utero electroporation and brain slice imaging

Organotypic cerebellar slice culture was performed as previously described⁵². Plasmid electroporation into P5 cerebella was conducted using an electroporator (CUY21, Nepagene). A total of six square-wave pulses (70 mV, 50 ms duration, 150 ms intervals) were applied. Cerebella were collected at P7 and embedded in 3.5% low-melting-point agarose (Nacalai, 01651-76), and coronally sliced into 300-μm-thick sections using a vibratome (NLS-AT, Dosaka EM). The slices were placed onto Millicell-CM membrane inserts (Millipore, PICM0RG50) and embedded in a collagen gel matrix (FUJIFILM Wako, 631-00651). The slices were maintained in medium composed of 60% basal medium Eagle (BME, Sigma-Aldrich, B9638), 25% Earle’s balanced salt solution (Sigma-Aldrich, E7510), 15% heat-inactivated horse serum (HS, Invitrogen, 26050070), 3 mM l-glutamine (Gibco, 35050-061), 1 mM sodium pyruvate (Sigma-Aldrich, S8636), 5.6 g l⁻¹ glucose (Sigma-Aldrich, G7021), 1.8 g l⁻¹ sodium bicarbonate (Sigma-Aldrich, S5761), and N-2 supplement (Invitrogen, 17502001). In utero electroporation into cerebral cortex was performed at E12. Brains were dissected at E14 and coronally sliced using a surgical knife⁵³. Slices were mounted in a collagen gel matrix (FUJIFILM Wako, 631-00651) in DMEM/F12 (Sigma-Aldrich, D2906) supplemented with 5% HS, 5% FBS (FBS, BioWest, S1400-500), penicillin–streptomycin (Invitrogen, 15140122), and N-2 supplement. The slices were placed in a stage-top incubator at 37 °C with humidified 85% O₂/5% CO₂ gas flow, mounted onto an upright microscope (BX61WI, Olympus)⁵⁴. Time-lapse imaging was performed using a laser-scanning confocal microscope (FV1000, Olympus) equipped with a GaAsP detector and a ×40 water-immersion objective lens (NA 0.8). Images were acquired every 5 or 10 min over 6 h.

Live-cell imaging on microfabricated constriction substrates

The microfabricated PDMS substrate were made as previously described¹⁸. CGNs from P6 mouse cerebella were dissociated using the Neuron Dissociation Kit (Wako Pure Chemical Industries, 291-78001). For transfection, Nepa21-S (Nepagene) was used according to the manufacture’s instruction with a total of 6 μg plasmid DNA. After electroporation, cells were recovered in non-coated dishes for 1 h in 10% HS/BME, and then seeded onto microfabricated PDMS substrates coated with laminin (Sigma-Aldrich, L2020) and incubated in culture medium BME with 26.4 mM glucose, 25 mM sodium bicarbonate, 1% BSA (Sigma-Aldrich, A3156), 1× N-2 supplement}. HeLa cells were maintained at around 90% confluence in Dulbecco’s modified Eagle’s medium (DMEM, Invitrogen, 11965092) supplemented with 10% FBS and penicillin–streptomycin on surface-treated culture dishes (Corning, 300-035). HeLa cells were transfected using Lipofectamine 2000 (Thermo Fisher Scientific, 11668019) 1 day after passaging and transferred to microfabricated substrates. Time-lapse imaging was performed using the CV1000 system (Yokogawa) with a ×40 (NA1.3) or a ×100 (NA1.4) oil-immersion objective lens at 37 °C in a humidified chamber with 5% CO₂. Images were captured at 5-min intervals.

Transwell assay

Dissociated CGNs were seeded at 1.7 × 10⁵ cells per cm² onto polycarbonate membranes with pore sizes of 0.4 μm, 3 μm or 8 μm (Corning, 3401, 3402, 3414, 3412, 3422) coated with poly-d-lysine (Sigma-Aldrich, P6407) and laminin. Drugs or DMSO (Nacalai, 13445-74) were added to both sides of the membrane 2 h after seeding, except for nocodazole, which was added to the bottom side 2 h after seeding. Cells were allowed to transmigrate for the indicated durations and then either fixed or collected for further analyses. For Transwell assays with transfected CGNs, cells were first plated onto poly-d-lysine- and laminin-coated 2.5 cm culture dishes (Corning) in culture medium. At 2.5 h (mCherry-KASH1) or 24 h (all other constructs) after plating, cells were washed with prewarmed PBS and dissociated using Accutase (Nacalai, 12679-54) at 37 °C for 5–10 min. Cells were collected by centrifugation, resuspended in a fresh culture medium and then reseeded onto Transwell inserts coated with poly-d-lysine and laminin and incubated for the indicated durations. For HeLa cells, transfected cells were suspended in DMEM and transferred to 1.12 cm² Transwell inserts that were preincubated with the medium at 1 × 10⁵ cells per well. DMEM supplemented with 10% FBS was added into the bottom chamber to promote migration.

Immunofluorescence analysis of Transwell cultures

Cells seeded on Transwell inserts were fixed with 4% paraformaldehyde (PFA, Nacalai, 02890-45) and permeabilized with 0.1% Triton X-100 (Nacalai, 35501-02) in PBS at room temperature. After blocking in 5% skimmed milk (Becton Dickinson, 232100), cells were incubated with primary antibodies diluted in blocking buffer overnight at 4 °C. After thorough washes, cells were incubated with Alexa-Fluor-conjugated secondary antibodies for 8–10 h at 4 °C, followed by nuclear counterstaining with 4′−6-diamidino-2-phenylindole (DAPI; FUJIFILM Wako, 340-07971) (5 μg ml⁻¹). For Ku70 staining, cells were washed with PBS and incubated with CSK + Rbuffer (3 mM MgCl₂, 10 mM PIPES, pH 7.0, 100 mM NaCl, 300 mM sucrose, 0.7% Triton X-100 and 0.3 mg ml⁻¹ RNase) at room temperature for 3 min for pre-extraction. Cells were then washed with PBS and fixed as described above⁵⁵. The TUNEL assay was performed according to the manufacturer’s instructions for the Click-iT TUNEL Alexa Fluor Imaging Assay kit (Thermo Fisher Scientific, C10245). The membranes were then mounted using ProLong Gold Antifade Mountant (Invitrogen, P36961). Imaging was performed using a laser-scanning confocal microscope (FV4000-BX61, Olympus) equipped with CellSens FV software (v.3.2.1.85) through ×20 (0.5 NA), ×40 (0.95 NA) objectives or a ×100 (1.4 NA) oil-immersion objective, or a lattice structured illumination (SIM) microscope (Elyra 7, Zeiss) through a ×63 (1.40 NA) oil-immersion objective.

Immunofluorescence and H&E staining of brain tissues

Mice were anaesthetized with isoflurane (FUJIFILM Wako, 099-06571) and perfused with 4% PFA in phosphate buffer (PB). Brains were dissected immediately and post-fixed in 4% PFA/PB for 4–6 h (for P15 or younger mice) or overnight (for P30 or older mice) at 4 °C. Brains were washed in PBS then dehydrated in 30% sucrose (Nacalai, 30404-45) in PBS overnight at 4 °C and were embedded in OCT compound (Sakura Finetek, 4583). Sagittal sections at a thickness of 15 μm were prepared using a cryostat (Leica, CM1950). The sections were incubated with HistoVT One (Nacalai, 06380-05) at 70 °C for 20 min for antigen retrieval. After permeabilization with 0.5% Triton X-100/PBS and blocking, immunofluorescence was done as described above. For γH2AX immunostaining, all solutions, including blocking buffer and antibody diluents, were prepared using Tris-buffered saline (TBS) instead of PBS. After staining with DAPI (10 μg ml⁻¹), the sections were mounted using Fluoromount-G (Cosmo Bio, SBA-0100-01-25). Imaging was performed using the confocal and SIM microscope systems described above. For haematoxylin and eosin (H&E) staining, anaesthetized mice were perfused with PBS followed by 10% formaldehyde neutral buffer solution (Nacalai, 37152-51). The brains were post-fixed in the same fixative for 1 week at 4 °C. Fixed tissues were processed for paraffin embedding using standard dehydration and clearing protocols, then sectioned at a thickness of 5 μm and subjected to H&E staining. Images were acquired using an upright optical microscope (DM5000B, Leica) equipped with 2.5× (NA 0.07) and 10× (NA 0.3) objective lenses, an AdvanCam-U3X camera (Advan Vision) and AdvanView imaging software.

Isolation and quantification of the TOP2cc

Dissociated CGNs were seeded onto culture dishes or Transwell inserts of the indicated pore sizes. After incubation for the indicated times, cells either on the bottom of 3 µm or top of 0.4 µm membranes were washed with PBS and collected by incubating with Accutase at 37 °C for 15 min. TOP2cc isolation was performed as previously described with a few modifications⁵⁶. In brief, cells were lysed sequentially in sucrose-based buffers to isolate chromatin. The chromatin precipitate was washed ten times to remove non-covalently bound TOP2β, subsequently dissolved in 6 M guanidinium chloride (Fuji Film, 074-05005). After DNA quantification, 1 µg of DNA was loaded onto a density-gradient cushion of caesium chloride (Nacalai, 07807-11) and subjected to ultracentrifugation at 100,000g for 16 h at room temperature. Then, 1 ml aliquots were collected from the top to the bottom of the gradient. For slot blot analysis, 100 µl of each aliquot was loaded to a PVDF membrane pretreated with methanol using the Bio-Dot apparatus (Bio-Rad, 31460). After washing with 0.2 M PB, the membrane was probed with an anti-TOP2β antibody, followed by a HRP-conjugated anti-rabbit secondary antibody. Image Gauge software (Fujifilm) was used to detect TOP2β signals. The raw mean intensity within a defined region of the TOP2cc fractions (third and subsequent dots from the left) was quantified. The background signal was subtracted, and the resulting values were plotted as the base-2 logarithm of the intensity plus one (log₂[intensity + 1]) to quantify TOP2cc dimers.

TOP1cc detection by DUST assay

CGNs were seeded on culture dishes and incubated for 30 min in the presence or absence of 300 nM topotecan. TOP1ccs were detected using the detection of ubiquitylated and SUMOylated TOP-DNA complexes (DUST) assay as previously described⁵⁷. After treatment, cells were washed with PBS and lysed directly in 800 µl DNAzol reagent (Invitrogen) supplemented with 1× Complete EDTA-free protease inhibitor cocktail (Roche) and 1 mM dithiothreitol. DNA was precipitated by the addition of 400 µl of 200-proof ethanol, collected by centrifugation, washed with 75% ethanol and resuspended in TE buffer. The samples were incubated at 65 °C for 15 min and sonicated twice for 15 s at 20% amplitude using an Ultrasonic Processor (Cole-Parmer). After centrifugation at 15,000 rpm for 5 min at 4 °C, the supernatants were collected and the DNA concentration was determined. Next, 5 µg of DNA from each sample were digested with 2,000 U of micrococcal nuclease (2,000 U µl⁻¹; New England Biolabs) in micrococcal nuclease reaction buffer for 30 min at 37 °C. The digested samples were run on 4–12% Bis-Tris gels (NuPAGE, Invitrogen) and wet-transferred onto nitrocellulose membranes. TOP1cc was detected using an anti-TOP1 antibody, followed by HRP-conjugated anti-rabbit secondary antibody and chemiluminescence detection. As a migration control, 3 µg of whole-cell extract from untreated CGNs was loaded.

γH2AX western blot analysis

CGNs were subjected to Transwell migration through inserts with 3-µm or 8-µm pore size for 12 or 24 h. After migration, inserts were washed with PBS and whole-cell extracts (WCEs) were prepared by excising the insert membranes and resuspending them directly in Laemmli lysis buffer containing 4% SDS, 20% glycerol, 125 mM Tris-HCl (pH 7.4), 50 mM β-glycerophosphate disodium salt, 2 mM PMSF and 1× Complete Mini EDTA-free protease inhibitor cocktail (Roche). The samples were boiled for 5 min, cooled to room temperature on ice, centrifuged at 13,000 rpm for 10 min and the supernatant was then sonicated twice for 15 s at 20% amplitude using an Ultrasonic Processor (Cole-Parmer), and then centrifuged at 13,000 rpm for 10 min. Supernatants were collected and protein concentration was determined using the Pierce BCA Protein Assay Kit (Thermo Fisher Scientific). Then, 4 µg of total protein per sample was run on 12% Bis-Tris gels (NuPAGE, Invitrogen) and wet-transferred onto PVDF membranes. Membranes were blocked for 1 h at room temperature in TBST containing 5% ECL Prime Blocking Reagent (Cytiva), then incubated for 2 h at room temperature with primary antibodies against γH2AX and α-tubulin diluted in blocking buffer. The membranes were washed three times with TBS containing 1% Tween-20 and incubated for 1 h at room temperature with HRP-conjugated secondary antibodies. Signals were detected by chemiluminescence and imaged using an ImageQuant LAS 4000 system. Band intensities were quantified by densitometry using ImageJ software (NIH). For each blot, background signal was subtracted and the γH2AX signal intensity was normalized to the corresponding α-tubulin band from the same lane.

AAV injection

AAVs (10⁹–10¹⁰ plaque-forming units) were produced using AVB Sepharose High Performance (GE Healthcare, 28-4112-01) according to the manufacturer’s instructions. In brief, transfected HEK293T cells were collected, then subjected to freeze–thaw cycles (three times using liquid nitrogen and a 37 °C water bath). The cell lysates were treated with 5–10 µl benzonase, incubated at room temperature for 5 min and centrifuged. The supernatant was filtered before bead binding. 400 µl AVB beads were washed three times with wash buffer (20 mM Tris-Cl pH 8.0, 250 mM NaCl, 10 mM MgCl₂) and incubated with the viral lysates at room temperature for 15 min. The beads were then washed three times and transferred to the Mobicol Classic (MoBiTec, M1002) column. Bound AAVs were eluted acidic elution buffer (250 mM NaCl, 10 mM MgCl₂, pH 3.0 adjusted with HCl) and immediately neutralized with 1 M Tris-HCl (pH 8.0). Eluted virus was concentrated to 100–200 μl using Amicon Ultra-4 Ultracel 50 K centrifugal filters (Millipore, UFC805024).

Mice aged 12 months were anaesthetized with an intraperitoneal injection (0.1 ml per 10 g body weight) of a mixed anaesthetic solution containing 0.03 mg ml⁻¹ medetomidine (Kyoritsu Seiyaku), 0.4 mg ml⁻¹ midazolam (Astellas Pharma) and 0.5 mg ml⁻¹ butorphanol (Meiji Seika Pharma). Then, 1 µl of virus was injected over 1 min at a stereotaxic location 6.84 mm posterior to bregma, along the midline and 1.5 mm below the dural surface. After injection, the incision was sutured and the mice received an intraperitoneal injection of 0.3 mg ml⁻¹ atipamezole (Kyoritsu Seiyaku) for anaesthesia reversal. Animals were kept in a warm (37 °C) recovery cage until fully awake. Then, 1 week after injection, mice were perfusion-fixed. Brains were post-fixed, washed in PBS, and embedded in 4% low-melting-point agarose. Coronal sections of 100 µm thickness were prepared using a vibratome. Immunostaining for GFP and VGLUT1/2 was performed as described. Images were acquired using a laser-scanning microscope (Andor Dragonfly 500) equipped with a ×100 oil-immersion objective (NA 1.49) and Fusion software (v.2.3.0.50).

Electron microscopy

Brains fixed in 4% PFA/PB were washed in PB followed by distilled water. First post-fixation was carried out on ice using 2% osmium tetroxide (OsO₄) with 1.5% potassium ferrocyanide for 2 h. After rinsing, second post-fixation was performed at room temperature using 1% OsO₄ in PB for 2 h. Tissues were then dehydrated through graded ethanol (50–100%), cleared in propylene oxide and infiltrated with EPON resin through graded propylene oxide: EPON mixtures, followed by overnight incubation in 100% EPON. Polymerization was conducted at 45 °C for one night and 60 °C for two nights. Semi-thin sections (700 nm–1 μm) were prepared for light microscopy screening and ultrathin sections (60–80 nm) were used for electron microscopy. The sections were stained with uranyl acetate and lead citrate, and images were captured using a transmission electron microscope (Hitachi, H-7650 or JEOL, JEM-1400Flash).

RNA-seq

Total RNA was extracted from dissociated CGNs from WT P6 ICR mice cultured for 8 h on normal dishes and Transwell inserts; control and Lig4^ND1-cre cerebella at 2 months using the miRNeasy Mini Kit (Qiagen, 217004) and assessed for quality using a BioAnalyzer (Agilent Technologies). The libraries were prepared using the TruSeq Stranded mRNA Library Prep Kit (Illumina) and sequenced on the NovaSeq 6000 system (Illumina) with 100 bp single-end reads. Raw Fastq files were trimmed and aligned to the mouse reference genome (mm10) using STAR (v.2.7.11a)⁵⁸. The mapped reads were assembled with FeatureCounts (v.2.0.8)⁵⁹, and differential gene expression was analysed using DESeq2 (v.2.11.40.8)⁶⁰ based on read counts. GO analysis was performed using upregulated and downregulated DEGs (P < 0.05, determined by DESeq2) using the DAVID functional annotation tool^61,62.

For public data analysis, raw sequence files were obtained from the BioProject database (PRJNA281127)³⁵ and from the GEO database (GSE212336 (ref. ³⁶), GSE221124 (ref. ³⁷) and GSE174265 (ref. ³⁸)). The reads were mapped to mouse genome (mm10) using hisat2 (v.2.1.0)⁶³, and mapped reads were assembled with FeatureCounts (v.2.0.0).

scRNA-seq data analysis of cerebellar cell types

A publicly available Seurat object containing single-cell RNA-seq (scRNA-seq) data from cerebellar tissue (24,409 genes across 611,034 cells) was obtained from a previously published study (https://singlecell.broadinstitute.org/single_cell/study/SCP795/a-transcriptomic-atlas-of-the-mouse-cerebellum). The object was loaded into R (v.4.1.2) and was updated from v.2.X to v.3.X using the UpdateSeuratObject function to ensure compatibility with our Seurat version (v.4.1.0). No additional quality-control or filtering steps were applied beyond those performed in the original study. Raw counts were log-normalized with a scaling factor of 10,000. The dataset was subset into 17 distinct cell type clusters, including macrophages, microglia, astrocytes, Bergmann glia, fibroblasts, endothelial mural and stalk cells, ependymal cells, oligodendrocyte precursor cells, oligodendrocytes and neuronal populations (granule cells, Purkinje cells, Golgi cells, molecular layer interneurons 1 and 2, Purkinje layer interneurons and unipolar brush cells). To optimize computational efficiency for dot plot generation, cells were downsampled to a maximum of 5,000 cells per cluster using a random seed, resulting in 57,833 total cells. Gene expression patterns were visualized using custom dot plots generated with ggplot2, where average normalized expression values were calculated using the AverageExpression function from the normalized RNA assay slot, and percent expression was computed as the fraction of cells within each identity with expression greater than zero.

ChIP–seq

Dissociated CGNs cultured on poly-d-lysine and laminin-coated dishes (Corning) were collected with Accutase as described above. For tissue preparation, cerebella were dissected from P15 mice. Nuclei were isolated according to a previously reported protocol⁶⁴. The nucleus pellets were resuspended and filtered through a 30-μm cell strainer (Miltenyi Biotec, 130-098-458) into prechilled tubes. ChIP for H3K4me1, H3K27ac, H3K4me3, H3K9me3 and lamin B1 was performed according to previously published methods⁶⁵. In brief, isolated nuclei or cell pellets were resuspended in prewarmed medium and fixed with 1% formaldehyde (Sigma-Aldrich, F1635) at 37 °C for 10 min. Fixation was quenched by adding 1 M glycine/PBS to a final concentration of 125 mM. Cells were washed twice with cold PBS. For lysis, fixed cell pellets were resuspended in cold RIPA buffer (10 mM Tris-HCl pH 7.5, 1 mM EDTA, 0.1% SDS, 0.1% sodium deoxycholate, 1% Triton X-100) supplemented with protease inhibitors (Thermo Fisher Scientific, 78442). Chromatin was sheared using a Covaris S220 sonicator (20% duty cycle, 175 peak powers, 200 cycles per burst) for 30 min at 4 °C. Lysates were cleared by centrifugation and precleared with Dynabeads Protein A (Invitrogen, DB10002) for 30 min at 4 °C. Antibodies (10 μg) were conjugated to prewashed Dynabeads in PBS for 10 min at room temperature, washed and incubated with chromatin overnight at 4 °C with rotation. Beads were then sequentially washed: twice with RIPA buffer, twice with RIPA + 0.3 M NaCl, twice with LiCl buffer (0.25 M LiCl, 0.5% Igepal-630, 0.5% sodium deoxycholate), once with TE + 0.2% Triton X-100 and once with TE. Crosslinks were reversed by incubation at 65 °C for over 4 h in 0.3% SDS and 1 mg ml⁻¹ proteinase K (Nacalai, 15679-06). DNA was purified and eluted in 10 mM Tris-HCl. DNA obtained from ChIP was used to construct sequencing libraries using the KAPA HyperPrep Kit (KAPA Biosystems). Libraries were sequenced on the NextSeq 500 platform (Illumina) using 75 bp single-end reads.

ATAC–seq

Dissociated CGNs cultured on poly-d-lysine- and laminin-coated dishes (Corning) were collected with Accutase as described above. Cells were resuspended in ice-cold lysis buffer (10 mM Tris-HCl pH 7.4, 10 mM NaCl, 3 mM MgCl₂, 0.1% IGEPAL CA-630) and centrifuged again under the same conditions. The supernatant was removed. For transposition, 2× Tagmentation buffer (Diagenode, C01019043) and Tagmetases (Diagenode, C01070012-10) were diluted in water to make reaction mixture. The reaction mixture was then added to cell pellet and incubated at 37 °C for 30 min. DNA was purified using the Qiagen PCR Purification Kit (QIAGEN, 28104) and eluted in elution buffer. Purified ATAC DNA was used for library amplification using the Q5 Hot Start High-Fidelity 2× Master Mix (NEB, M0494S). The libraries were sequenced on the NextSeq 500 platform (Illumina) using 75 bp single-end reads.

END-seq

Dissociated CGNs were seeded on 3-μm or 8-µm Transwell inserts and incubated for 6 h in the presence of palbociclib (5 µM). Cells were collected from the bottom side of the Transwell inserts with Accutase. For tissue samples, nuclei were isolated from P15 control and Lig4^ND1-cre cerebella as described above. Isolated nuclei or cells were embedded in agarose plugs and END-seq was performed as previously described³⁴. Agarose plugs were treated with proteinase K for 1 h at 50 °C and then for 7 h at 37 °C. Plugs were washed in wash buffer 10 mM Tris-HCl pH 8.0, 50 mM EDTA) and in TE (10 mM Tris-HCl pH 8.0, 1 mM EDTA) followed by RNase A treatment for 1 h at 37 °C. Inside the plugs, DNA ends were blunted for 1 h at 37 °C with exonuclease VII followed by exonuclease T (NEB) for 1 h at 25 °C to detect TOP2 breaks. After blunting, A-tailing was performed, followed by biotinylated END-seq hairpin adaptor 1 ligation using NEB Quick Ligase. The agarose plugs were melted and dialysed and DNA was sonicated using Covaris S220 sonicator for 4 min at 10% duty cycle, peak incident power 175, 200 cycles per burst, at 4 °C. DNA was ethanol precipitated and dissolved in 80 μl TE buffer. Biotinylated DNA was isolated using MyOne Streptavidin C1 Beads (Thermo Fisher Scientific, 650–01), followed by end repair (dNTPs, T4 polymerase (NEB), Klenow (NEB), T4 PNK) and dA-tailing (Klenow exo- (NEB), dATP). The second end was ligated to END-seq hairpin adaptor 2 using NEB Quick Ligase. Hairpins were digested using USER (NEB), and the resulting DNA fragments were PCR amplified using Illumina double barcoded primers. PCR fragments were isolated by size selection from agarose gel and purification using NEB Monarch Gel Extraction Kit. Libraries were quantified using KAPA Library Quantification Kit and sequenced using Illumina NextSeq 550 and NovaSeq X.

Genomic analysis

END-seq, ATAC–seq and ChIP–seq reads were mapped to the mouse (GRCm38p2/mm10) genomes using Bowtie2 (v.2.5.1-1)⁶⁶ using the default parameters. For END-seq, peaks were called using MACS (v.1.4.3)⁶⁷ with the parameters: -nolambda, -nomodel and -keep-dup = all (keep all redundant reads) and subsequent analysis were done using bedtools (v.2.31.1)⁶⁸ and R (v.4.3.2). For peak calling, the corresponding sample (condition/genotype) was used as control, and specific peaks were selected as not called in the other condition/genotype. Peaks were later filtered by >1.5 RPKM and >3-fold RPKM to the control for Transwell assay and >1.3 RPKM and >twofold for cerebellum.

The UCSC database⁶⁹ was used to obtain the RepeatMasker annotations and TSS, transcription end site, exons and intron positions (RefSeq annotation table). H3K9me3, H3K4me3, H3K4me1, H3K27ac ChIP–seq peaks were called using epic2⁷⁰ (v.0.0.41, with -bin 10000 for H3K9me3 and 500 for the rest). Lamin B1 domains (regarded as LADs) were called using Enriched domain detector (edd, v.1.1.18)⁷¹ with –bin-size 37 and -g 37. ATAC–seq peaks were called using the Genrich default parameters, available at GitHub (https://github.com/jsh58/Genrich). RNA-seq, ATAC–seq and histone mark ChIP–seq peaks in control samples were used to define chromatin categories (active promoters, active enhancers, active exons and introns, transcriptionally inactive and unlabelled euchromatin, heterochromatin with LADs and heterochromatin without LADs) using bedtools as follows: active promoters: H3K4me3 peaks that overlap with ATAC–seq peaks; active enhancers: ATAC–seq peaks that do not overlap with H3K4me3; active exons exons of genes in RNA-seq with DESeq2 base mean value ≥5 after subtracting H3K4me3 and ATAC–seq peaks; active introns: introns of genes in RNA-seq with DESeq2 base mean value ≥5 after subtracting H3K4me3 and ATAC–seq peaks; heterochromatin LADs: lamin B1 LADS after subtracting the union of active promoters, active enhancers, active exons, active introns; heterochromatin noLADs: H3K9me3 peaks after subtracting lamin B1 LADS, active promoters, active enhancers, active exons and active introns; Euchromatin: genome mm10 after subtracting heterochromatin; transcriptionally inactive and unlabelled euchromatin: euchromatin after subtracting active promoters, active enhancers, active exons and active introns.

Chromatin categories in cerebellar cells from P15 control and Lig4^ND1-cre mice were defined using bedtools as follows. Active promoters: H3K4me3 peaks that overlap with H3K27ac peaks; active enhancers: H3K4me1 peaks that overlap with H3K27ac peaks after subtracting active promoters; poised enhancers: H3K27ac peaks after subtracting active promoters and active enhancers; heterochromatin: H3K9me3 peaks; euchromatin: genome mm10 after subtracting heterochromatin; transcriptionally inactive and unlabelled euchromatin: euchromatin after subtracting active promoters, active enhancers and poised enhancers. For each category, we compared the percentage of peaks between two conditions (Transwell 8 µm versus 3 µm, and cerebellum P15 control versus control and Lig4^ND1-cre). Significance was estimated by 95% bootstrap confidence interval (CIs) for the difference in means between conditions using a nonparametric bootstrap: for each chromatin category (for example, active promoters) the four observed values were pooled and resampled rows (value + group label) with replacement 5,000 times, computed the difference of means in each resample, and took the 2.5th and 97.5th percentiles (percentile CI). A comparison was called significant when the 95% bootstrap CI for the difference did not include 0, and not significant when CI included 0. All analyses were performed in R (v.4.3.2) using the boot package (v.1.3.32, R = 5,000; seed = 123).

CTCF and RAD21 ChIP–seq peaks were called using MACS (v.1.4.3) with the default parameters. The position weight matrix (PWM) of CTCF was taken from JASPAR database (v.2024)⁷² and significant CTCF motifs (P < 1 × 10⁻⁴) in the mouse genome were selected using the FIMO tool (v.4.12.0)⁷³. Significant CTCF motifs at double CTCF/RAD21-bound peaks were used for the analysis. For data visualization, bedgraphs generated with bedtools were converted to .bigwig files using bedGraphToBigWig (v.302.1)⁷⁴. Visualization of genomic profiles was performed using the UCSC browser⁷⁵. Genome browser profiles were normalized to present RPM. For aggregate plots around CTCF sites, TSSs and gene bodies, the signal was smoothed using smooth.spline function in R.

Behavioural tests

All behavioural tests were conducted during the light phase in the room where the mice were kept. Age- and weight-matched male mice were used for all behavioural tests. Mice of different genotypes were group-housed and tested in a random order. Investigators were blinded to genotype during behavioural experiments and data analysis. Control mice were included in each experimental session.

Footprint analysis

Each mouse was guided to walk along a straight runway lined with paper toward a dark box at the end. The forepaws and hindpaws were coated with non-toxic black and red paints, respectively. Each mouse completed three consecutive trials. A fresh sheet of white paper was placed on the floor of the runway for each run. Parameters measured included stride length, width based on forelimb or hindlimb placement, and the distance between the same-side forepaw and hindpaw prints (gait overlap)⁷⁶.

Balance beam test

Mice were encouraged to cross a narrow cylindrical plastic beam (12 mm diameter, 1 m length), placed horizontally 1 m above the bench surface, toward a dark goal box. Each mouse was trained for the task for 2 days before trial experiments. Mice performed three trials per day with a 5-min interval between trials. Mice were allowed to cross the beam voluntarily without external stimuli during trials. Hindlimb slips were recorded only while the mouse was passing through the middle 80 cm of the beam. The average number of slips from three trials was calculated for each mouse.

Rotarod test

Mice were trained to remain on a rod (Rota-rod treadmill, Muromachi Kikai, MK-600) rotating at a constant speed of 2 rpm for 180 s (ref. ⁷⁷). After a 1 h interval, a 180 s test session was conducted on a rotating rod with acceleration from 2 rpm to 20 rpm. One trial was run for each mouse per day for three consecutive days, with the training session performed only on day 1. The latency to fall (time spent on the rod) was recorded for each mouse in each session.

Grip strength measurement

Grip strength was assessed using a grip strength meter (force gauge A and D, AD-4932A-50N) with a rectangular grid handle. The mice were allowed to grasp the mesh of the grid handle with all four limbs. Once they had a firm grasp, they were slowly pulled horizontally by the tail until all limbs released the grid. The maximum force exerted during the trial was recorded. Five trials were performed for each mouse, with a 5 min rest period between the trials. The maximum force recorded from the five trials was used for analysis. Grip strength was normalized by calculating the ratio of grip strength to body weight, expressed as the raw grip strength (N)/weight (g) × 100.

Image analysis and data representation

All image processing and analysis was performed using FIJI (ImageJ), except for the images presented in Extended Data Fig. 6a, which were processed using Imaris (Oxford Instruments). For fluorescence images, maximum projections of multiple z stacks were used in all figures except the magnified viewed enlarged in the Extended Data Fig. 3e, a single z stack was used to show the mono-structure of γH2AX. StarDist v.0.3.0, a tool plug-in on FIJI, was used to count the number of nuclei.

Statistical analysis

Statistical analysis was performed using GraphPad Prism v.7 or R v.4.3.3. Details regarding the number of biological replicates are provided in the figure legends. Error bars in the graphs indicate the s.d. The statistical methods applied are specified in each figure legend.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.