Fig 2: Stalled DNA replication, DSBs and G2/M arrests of Nde1-/- NPCs are associated with compromised H4K20me3(A) Diagram of IdU-CldU sequential labeling scheme for analyzing S-phase progression in embryonic cortical tissue. Pregnant dams were given IdU followed by CldU before embryos were fixed at E12.5. Embryonic NPCs were analyzed by IdU-CldU double IH staining. DNA were stained by Hoechst 33,342.(B) Representative IH images of H4K20me3 in S-phase cortical NPCs of WT and Nde1-/-embryos sequentially labeled by IdU and CldU as described in (A). DNA were stained by Hoechst 33,342. VZ: ventricular zone; CP: cortical plate; S: S-phase zone; G2/M: G2 and M-phase zone. Bar: 50um.(C) Representative IH images of H4K20me3 in NPCs at early S-phase (IdU-;CldU+), mid S-phase (IdU+;CldU+), late S-phase (IdU+; sparse CldU+ foci), or stalled S-phase (IdU+; CldU- and in the S-phase zone of basal VZ as opposed to where G2/M nuclei localize at the apical VZ) at E12.5. DNA were stained by Hoechst 33,342. H4K20me3 and Hoechst 33,342 double stained images are included to reveal the H4K20me3 occupancy at chromocenters. Nuclear territory is marked by red dots on H4K20me3 images according to DNA stains by Hoechst 33,342 (middle column). Note the weak and diffuse H4K20me3 immunosignals in late S-phase and DNA replication-stalled Nde1-/- NPCs. Bar: 10um.(D) Numbers of H4K20me3+ foci per nucleus in NPCs of E12.5 embryos at early, mid, late, and stalled S-phase (mean ± SD). p values by two-sample Kolmogorov-Smirnov tests are indicated.(E) Ratios of H4K20me3 to Hoechst 33,342+ foci in nuclei of early, mid, and late S-phase or DNA replication stalled cortical NPCs at E12.5. Mean ± SD *p< 0.0001, by student’s t tests.(F) Representative IH images of E12.5 cortical sections that were doubled labeled by antibodies against phospho-Histone H3 Ser10 (H3S10p, in green) and H4K20me3 (in red). DNA was stained by Hoechst 33,342 (in blue). Diagram to the left illustrates cell cycle-dependent interkinetic nuclear migration (INM) of cortical NPCs. Bar: 50um.(G) High-magnification IH images of H4K20me3 in selected H3S10p+ nuclei shown in (F) (indicated by circles and numbers). Note the reduced number and intensity of H4K20me3 foci in Nde1-/- nuclei retained in basal ventricular zone. Bar:10um.(H) IB of histone extracts from cortical tissue at E12.5, E14.5 and P0, showing that Nde1 LOF results in decreased H4K20me3 with increased DSBs (?H2A.X) and G2/M arrest (H3S10p). See also Figure S2.

Fig 3: Nde1 and Brap LOF result in nuclear architecture aberrations in heterochromatin-associated domains(A) Transmission electron microscopy (TEM) images of representative cortical NPCs and neurons in WT or Nde1-/- embryos. Arrows indicate nuclear lamina aberrations. Bar: 5um.(B) IH images of individual nuclei from cortical sections of mice at weaning age (P22-28), showing diverse nuclear architecture and heterochromatin (marked by H4K20me3) aberrations in cortical neurons (NeuN+) of WT, Nde1-/-, Nde1-/-p53-/-, and BrapcKO mice. Shown are representative images of nuclei from cortical sections of mice of indicated genotypes. n= 5 for each genotype. Bar:10um.(C) Violin plots of H4K20me3 foci size distribution on IH images of cortical neurons in mice of genotypes indicated at weaning age (P22-P28). The total number of foci quantified (by Image-ProPremier) for each genotype is represented as median plus quartiles (n = 3 mice per genotype). ***p<0.001 by two-sample Kolmogorov-Smirnov tests.(D) Quantification of inverted nuclei (%) in cortical tissues of mice of genotype indicated at weaning (1M) and 4 months (4M) of ages. Shown are Mean ± SD and distributions of the percentage of inverted nuclei from different mice (n = 3 mice per genotype). ***p<0.001 by student’s t-tests.(E) IH images of diverse nuclear architecture and nuclear lamina (marked by Lamin B1) in cortical neurons (NeuN+) of WT, Nde1-/-, Nde1-/-p53-/-, and BrapcKO mice at weaning age (P22-28). Shown are representative images of nuclei from cortical sections of mice of indicated genotypes (n = 5 mice for each genotype). Bar:10um.(F) Quantification of nuclear inversion, nuclear lamina aberration, and the coincidence of nuclear inversion and nuclear lamina aberration in neocortical neurons of mice of indicated genotypes at 3-4 months of age (data are represented as mean from n = 3 mice for each genotype). The percentages of four types of nuclei were scored. Type I: conventional nuclei with normal nuclear lamina; Type II: inverted nuclei with normal nuclear lamina; Type III: conventional nuclei with aberrant nuclear lamina; Type IV: inverted nuclei with aberrant nuclear lamina.(G) Transmission electron microscopy (TEM) images of representative nuclei of neocortical neurons of Nde1-/-, BrapcKO, Nde1-/-BrapcKO, and Control (Brapflox/wt Cre-) adult mice (4-8 months of age). Arrows indicate nuclear lamina aberrations. Higher magnification images show decreased heterochromatin abundance at the nuclear periphery of mutant neurons. Bars: 5um or 2um as indicated.(H) IH images of H4K20me3-Lamin B1 double stained neocortical sections of WT, Nde1-/-, BrapcKO, and Nde1-/-BrapcKO mice of weaning age (P22-P28). Note the phenocopy and synergy of Nde1 and Brap LOF. Bar: 50um.(I) Frequency of nuclear buds or micronuclei in cortical neurons (NeuN+) of mice of genotypes indicated at weaning age (P22-P28). Mean ± SD, **p<0.01, ***p<0.001 by student’s t-tests.DNA was visualized by staining with Hoechst 33,342. See also Figure S4.

Fig 4: Nde1 LOF induces progeria and aging associated histone PTMs, neuroinflammation, and neurodegeneration(A) IB of histone extracts from cortical tissues of mice of indicated genotypes at 4 months of age. Shown are representative images. Note the substantial increase in H4K20me3 and H2Aub in the cortical tissues with Nde1 or Brap LOF.(B) Ubiquitin IBs of cortical tissue total protein extracts from mice of indicated genotypes at 4 months of age. Note the accumulation of poly-ubiquitinated proteins in the cortical tissues with Nde1 or Brap LOF.(C) Proteasome activity analysis of WT and Nde1-/- cortical tissues. Samples were prepared from 8-10 monthold WT and Nde1-/- mice. 20S proteasome activity (mean ± SD) was determined by assessing the amount of Suc-LLVY-AMC cleavage.(D) IB of cortical tissue total protein extracts from mice of indicated genotypes at 10 months of age. Shown are representative images. Note the increased levels of GFAP and phospho-tau in Nde1-/- compared to WT mice.(E) IH images of GFAP (green) labeled cortical sections of WT and Nde1-/- mice at 12 months of age. Shown are representative images. Note the increased and enlarged GFAP+ cells in Nde1-/- mice compared to their WT littermates. Bar: 100um.(F) IH images of GFAP (green)-Lamin A/C (red) double immuno-labeled cortical sections of mice of indicated genotypes at 6 months of age. Shown are representative images. High-magnification views of the boxed areas are included. Note the spatial coincidence of aberrant nuclei and reactive astrocytes. Bars: 50um or 10um as indicated.(G) IH images of Iba (green) labeled cortical sections of WT and Nde1-/- mice at 12 months of age. Shown are representative images. High-magnification views of the boxed areas are included. Note the morphological difference of microglia between WT and Nde1-/- mice. Bars: 50um or 10um as indicated.(H) IH images of Iba (green)-Lamin A/C (red) double immuno-labeled mouse cortical sections of indicated genotypes at 6 months of age. Note the spatial coincidence of aberrant nuclei and activated microglia with ameboid morphology. Bar: 50um.(I) High-magnification views of aberrant nuclei (marked by Lamin A/C, red) embraced by activated microglia (Iba, green). Bar: 10um.(J) Quantification of dendritic spine density (Mean ± SD) and representative images of Golgi-cox analyses of WT and Nde1-/- mice at 3 and 12 months of age, showing age-dependent reduction in spine density in cortical layer ? pyramidal neurons of Nde1-/- mice. *p<0.01, ***p<0.0001 by Student’s t tests. Bars: 100um or 10um as indicated.Nuclear DNA was stained with Hoechst 33,342 in blue in all IH images.

Fig 5: Nde1 and Brap LOF results in heterochromatin DSBs and increased transcription of pericentromeric satellite repeats(A and B) Representative IH images of ?H2A.X-NeuN double immuno-labeled neocortical sections of 4 monthold WT, Nde1-/-, and Nde1-/-p53-/- mice. Note that ?H2A.X is shown specifically in the nucleus of NeuN+ cells (arrow heads). Bars: 50um or 10um asindicated.(C) IB of total cortical protein extracts from 4 month old mice of indicated genotypes. Note the substantial elevation of Brca1 and 53BP1 in Nde1-/- and Nde1-/-p53-/- cortical tissues.(D) IBs of histone extracts with anti-?H2A.X (upper image) and anti-ubiquitin (lower image) antibodies, showing increased ?H2A.X (~17kDa) and penta-ubiquitinated ?H2A.X (boxed, ~60 kDa) in cortical tissues of Nde1-/- and Nde1-/-p53-/- mice of 4 months of age. The blot was re-probed with an anti-ubiquitin antibody after stripping off ?H2A.X signals, confirming that the higher molecular weight band of ~60 kDa recognized by anti-?H2A.X (boxed) contains ubiquitin modification. Loading was normalized to histone H3.(E) Quantification (%) of neurons (NeuN+) showing ?H2A.X and inverted nuclei in cortical tissue of 4 monthsold mice of genotypes indicated. Mean ± SD, *p< 0.001 by student’s t tests.(F) Representative IH images of selected nuclei in ?H2A.X-LaminB1 double immune-labeled neocortical neurons in adult (3-4 months old) WT, Nde1-/-, Nde1-/-p53-/-, and BrapcKO mice. Note the coincidence of large ?H2A.X foci in neurons with aberrant nuclear lamina. Bar: 10um.(G) Representative IH images of selected nuclei in ?H2A.X-H4K20me3 double immuno-labeled neocortical neurons in adult (3-4 month old) WT, Nde1-/-, Nde1-/-p53-/-, and BrapcKO mice. Bar: 10um.(H) Representative line scans of ?H2A.X, H4K20me3, and Hoechst 33,342 fluorescence intensities of selected individual nuclei, revealing the spatial proximity of ?H2A.X with heterochromatin condensates.(I) Immuno-FISH analysis with a probe for major satellite DNA (green) and an antibody to H4K20me3 (red) on mouse neocortical sections. Shown are representative images, demonstrating the presence of pericentromeric major satellite repeats in the micronuclei of Nde1-Brap double deficient neocortical cells. Note the weaker H4K20me3 in the micronuclei (arrows). Bar: 50um.(J) PCR, cloning, and sequencing analyses of adult (2-4 months) cerebral cortical genomic DNA, showing increased polymorphic length variations of pericentromeric major satellite repeats in Nde1-/-, Nde1-/-p53-/- cortices. Data were from n = 3 mice per genotype. A total of 100 PCR clones were sequenced from each genotype. Data present all clones containing PCR inserts with MajSat sequences. p values by two-sample Kolmogorov-Smirnov tests are indicated.(K) Northern blot analysis of pericentromeric major satellite RNA in cerebral cortical tissues of 8–10 months old mice of indicated genotypes. The integrity and the amount of RNA loaded are shown by ethidium bromide stained gel below. See also Table S1. Major satellite length distribution.