Extended Data Fig. 13: Mutational Analysis at NPAS4-NuA4 sites during aging and Npas4 KO lifespan analysis.
From: A NPAS4–NuA4 complex couples synaptic activity to DNA repair

a. FACS sorting scheme to isolate NeuN-expressing neurons from aging Npas4 wild-type vs KO hippocampal tissue. Sorted cells from NeuN+ and NeuN- gates were re-analyzed for purity. See Supplementary Fig. 2 for gating scheme. b. Average normalized gene expression ± s.e.m. (relative to housekeeping gene Gapdh) of marker genes for neurons (NeuN+) or glial cell types (NeuN-). Each dot represents data from an individual mouse collected across 2 independent sorting days. ***P < 0.001; **P < 0.01; *P < 0.05. P values by unpaired, two-tailed t-tests with Benjamini-Hochberg multiple hypothesis correction. Individual P values and replicates as follows: Grin1 (n = 6): P = 2.06e-05, Grin2b (n = 3): P = 5.97e-03, Syapsin1 (n = 3): 5.72e-04, Rbfox3 (n = 6): 3.89e-04, Npas4 (n = 3): P = 5.21e-03, S100b (n = 6): P = 3.85e-06, Gfap (n = 3): P = 1.89e-02, Aldh1l1 (n = 3): P = 9.50e-03, Mog (n = 3): P = 2.29e-04, Pdgrfa (n = 6): P = 1.16e-05. c. Diagram of the SiMSen-seq amplicon sequencing approach used to identify somatic mutations that occur in individual DNA templates. d. Diagram of positive control experiment (top) used to test SiMSen-seq mutation detection method. Cultured cortical neurons were infected with lentivirus to express Cas9 with or without guide RNAs (Bdnf, Fos, Inhba, Scg2). SiMSen-seq was used to detect mutations at the gRNA cut sites at Inhba and Bdnf. Mutation frequency (that is, the number of insertions/deletion per UMI family) is plotted across a 40 bp window surrounding the cut site. e. Violin plots of read depth-normalized sequencing signal for NPAS4, MRE11, RAD50, H3K27ac CUT&RUN, γH2AX ChIP-seq and ATAC-seq in 2 h stimulated neurons at NPAS4-bound (24) and No NPAS4 (11) sites selected for mutational analysis. Each point represents a site. Inducible H3K27ac and Inducible ATAC plots show the distribution of fold changes (2 vs 0 h) in ATAC-seq and H3K27ac at these same sites. Dashed line represents the median; solid lines represent quartiles. ***P < 0.0005; **P < 0.005. P values by unpaired, two-tailed Wilcoxon rank sum tests. NPAS4: P = 0.002, MRE11: P = 0.0009, RAD50: P = 1.33e-04, γH2AX P = 0.0027, Percent GC P = 0.76; Percent AT P = 0.76; Inducible H3K27ac P = 0.25, H3K27ac P = 0.07; ATAC P = 0.19; Inducible ATAC P = 0.53. Replicates per factor included in Supplementary Table 2. f. Genomic annotations of NPAS4-bound (24) vs No NPAS4 (11) sites. g. Left panel: Average SNV frequency ± s.e.m. in NPAS4-Bound and No NPAS4 in young (3-month-old) animals. Each point represents a single site sampled from a mouse and data from 4 mice are shown. *P < 0.05; **P < 0.005, ***P < 0.001; T>A P = 0.99, T>G P = 0.99, T>C P = 0.0049, C>A P = 0.027, C>G P = 0.99, C>T P = 1.85e-09. P values by a one-way ANOVA with Holm-Sidak’s correction for multiple hypothesis testing. Right panel: Average Insertion/Deletion frequency ± s.e.m. in NPAS4-Bound and No NPAS4 sites in young (3-month-old) animals. Each point represents a single site sampled from a mouse and data from 4 mice are shown. ***P = 2.96e-05. P value by an unpaired, two-tailed Wilcoxon rank-sum test. h. Lifespan analysis on Npas4 wild-type (n = 25) vs Npas4 KO (n = 27) male littermates. Median lifespan KO = 12 months; Median lifespan of wild-type not determined. P = 8.67e-06 by two-tailed Gehan-Breslow-Wilcoxon test; P = 1.37e-06 by two-tailed log-rank Mantel-Cox test. i. Lifespan analysis on Npas4 wild-type (n = 28) vs Npas4 KO (n = 37) female littermates. Median lifespan KO = 11 months; Median lifespan of wild-type not determined. P = 1.01e-06 by two-tailed Gehan-Breslow-Wilcoxon test; P = 8.48e-08 by two-tailed log-rank Mantel-Cox test. j. Lifespan analysis on Npas4 wild-type (n = 16, Npas4+/+; Camk2a-Cre+; Sun1fl/+) vs Npas4 cKO (n = 9, Npas4fl/fl; Camk2a-Cre+; Sun1fl/+). Median lifespan cKO = 21.46 months; Median lifespan of wild-type = 29 months. P = 0.049 by two-tailed Gehan-Breslow-Wilcoxon test; P = 0.19 by a two-tailed log-rank Mantel-Cox test.