Extended Data Fig. 9: Suppression of DNA methylation in young AT2 cells and LUAD cells recapitulates ageing-associated changes in gene expression.
From: Ageing limits stemness and tumorigenesis by reprogramming iron homeostasis
(a) Quantification of 5-methylcytosine (5-mC) immunofluorescence in lungs of young mice administered the DNMT1 inhibitor GSK3685032 for 8 days or vehicle controls (n = 4 mice). AT2 cells were identified as SPC+ cells. (b) Scatter plot showing the correlation between the fold change of DEGs from the signature shared by AT2 cells and LUAD cells (Extended Data Fig. 3i; Supplementary Table 2) in AT2 cells isolated from young mice administered DNMT1-i or vehicle (x-axis) vs. aged or young vehicle-treated mice (y-axis). Pearson correlation was calculated. The regression line (blue) and its 95% confidence interval (light grey) were plotted using a linear model. (c) Quantification of Nupr1 mRNA molecules in Sftpc+ AT2 cells in young mice administered vehicle (blue) or DNMT1-i (orange) for 8 days, compared to aged mice that received vehicle (red) (n = 353, 438, and 429 AT2 cells per condition, respectively). Scale bar: 20 µm. (d) Ex vivo transformation of AT2 cells from aged and young KP mice with or without DNMT1 inhibition (n = 3 biological replicates). (e-f) Expression of Nupr1 and Lcn2 in KP LUAD cell lines with or without DNMT1 inhibition (n = 4 independent KP lines). The mRNA levels are normalized to the housekeeping gene Gapdh and further normalized to vehicle control. Note that data of (e) was plotted as a heatmap in Fig. 4f. (g) Experimental strategy for transcriptomic profiling of AT2 cells following shRNA-mediated knockdown of Dnmt1 compared to Renilla control in vivo. (h) Validation of shRNA targeting Dnmt1 by quantitative PCR (n = 3 technical replicates from one representative experiment, repeated twice). (i) Quantification of 5-mC immunofluorescence in AT2 cells of young Rosa26mTmG mice transduced with lentiviral PGK-Cre plus shRNAs targeting Dnmt1 or Renilla control (n = 4 and 7 mice). Given that successful delivery of the shRNAs is marked by switch of endogenous fluorescence from tdTomato to GFP, 5-mC only in the GFP+ AT2 cells was quantified. (j) Contour plot showing correlation between the fold change (FC) of DEGs in AT2 cells isolated from young mice administered shDNMT1s or control shRNA (shRenilla) (x-axis) compared to aged vs. young AT2 cells expressing shRenilla control (y-axis). Grayscale represents the probability distribution. Pearson correlation was calculated. The percentage of genes in each quadrant is significantly different from random (binomial test p < 2.2 ×10−16). The regression line (blue) was plotted using a linear model. (k) Scatter plot showing the correlation between the fold change of DEGs from the signature shared by AT2 cells and LUAD cells (Extended Data Fig. 3i; Supplementary Table 2) in AT2 cells expressing shDnmt1 or shRenilla (x-axis) isolated from young mice vs. aged or young shRenilla-treated mice (y-axis). Pearson correlation was calculated. The regression line (blue) and its 95% confidence interval (light grey) were plotted using a linear model. (l) Expression of Nupr1 in KP LUAD cell lines expressing shRNAs targeting Dnmt1 or Renilla control (n = 4 independent KP lines). The mRNA level is normalized to the housekeeping gene Gapdh and further normalized to shRenilla. Mean with SD is shown in (a), (d-f), (h-i) and (l). Mean with SEM is shown in (c). Two-tailed Student’s t test was used in (a) and (i). One-way ANOVA was used in (c-d) and (h). Two-tailed paired Student’s t test was used in (e-f) and (l). Pearson correlation test was used in (b) and (k). The illustration in (g) was created using BioRender (https://biorender.com).
