Chemically induced reprogramming to reverse cellular aging

Abstract

A hallmark of eukaryotic aging is a loss of epigenetic information, a process that can be reversed. We have previously shown that the ectopic induction of the Yamanaka factors OCT4, SOX2, and KLF4 (OSK) in mammals can restore youthful DNA methylation patterns, transcript profiles, and tissue function, without erasing cellular identity, a process that requires active DNA demethylation. To screen for molecules that reverse cellular aging and rejuvenate human cells without altering the genome, we developed high-throughput cell-based assays that distinguish young from old and senescent cells, including transcription-based aging clocks and a real-time nucleocytoplasmic compartmentalization (NCC) assay. We identify six chemical cocktails, which, in less than a week and without compromising cellular identity, restore a youthful genome-wide transcript profile and reverse transcriptomic age. Thus, rejuvenation by age reversal can be achieved, not only by genetic, but also chemical means.

Keywords: epigenetics; information theory of aging; rejuvenation medicine; reprogramming; small molecules.

Figure 1

The NCC reporter system to monitor cellular senescence. (A) The NCC reporter system integrated in human fibroblasts. (B) NCC signals in quiescent fibroblasts. (C) Fluorescence intensity profiles corresponding to the path of the arrow in (B). (D) The colocalization of mCherry and eGFP signals in quiescent fibroblasts by Pearson correlation. (E) NCC signals in senescent fibroblasts. (F) Fluorescence intensity profiles corresponding to the path of the arrow in (E). (G) The colocalization of mCherry and eGFP signals in senescent fibroblasts by Pearson correlation. (H) Pearson correlation of quiescent and senescent fibroblasts. Data are mean ± SD. ^****p < 0.0001. Two-tailed Student’s t test.

Figure 2

OSK-mediated partial reprogramming ameliorates features of cellular senescence. (A) Heatmaps for mRNA levels of genes upregulated by senescence (n=3, p-adj < 0.01, FC > 2). (B) Percentage of genes changed by OSK (n=3, p-adj < 0.05) among those upregulated by senescence. (C) Heatmaps for mRNA levels of genes downregulated by senescence (p-adj < 0.01, FC > 2). (D) The percentage of genes changed by OSK (p-adj < 0.05) among those downregulated by senescence. (E) Top 20 gene ontology (GO) processes of genes upregulated by senescence. The red and blue bars indicate upregulation or downregulation by OSK, respectively. (F) Top 20 GO processes of genes downregulated by senescence. Red and blue bars indicate upregulation or downregulation by OSK, respectively. (G) Schematics of the Tet-On OSK system integrated in NCC reporter system fibroblasts. (H) NCC signals and track of the arrows in quiescent, senescent, or senescent + OSK fibroblasts. Scale bar, 50 μm. (I) Fluorescence intensity profiles corresponding to the arrow in (H). (J) EGFP intensities in the cytoplasm. Data are mean ± SD. ^*p < 0.05; ^***p < 0.001. One-way ANOVA-Bonferroni.

Figure 3

Reprogramming small molecule cocktails restore NCC alterations in senescent cells. (A) Chemical structures of small molecules of basal cocktails used to generate induced pluripotent stem cells (iPSCs) from mouse (left) or human (right) somatic cells. (B) Correlation heatmaps showing eGFP and mCherry colocalization in human senescent fibroblasts demonstrate the effects of 80 different combinations of small molecules (n=2). (C, D) Validation of six selected cocktails through independent experiments, showing colocalization (C) and representative images (D) of eGFP and mCherry signals. Scale bar, 50 μm. Data are mean ± SD. ^*p < 0.05; ^**p < 0.01; ^****p < 0.0001. One-way ANOVA-Bonferroni.

Figure 4

Transcriptomic rejuvenation by reprogramming small molecule cocktails. (A, B) Delta transcriptomic ages (tAgeΔ), as measured by a biological transcriptomic clock built on rodent and human transcriptomic data (A), or rodent data alone (B). (C) Delta ages, as measured by a chronological transcriptomic clock built using human data. n.s.: p > 0.05; ^*p < 0.05; ^**p < 0.01; ^***p < 0.001. (D) Correlation matrix of transcriptomic differences by cocktail treatment, iPSC reprogramming, or aging. ^*p-adj < 0.05; ^**p-adj < 0.01; ^***p-adj < 0.001. Benjamini-Hochberg approach. (E) Enrichment of pathways by cocktail treatment, iPSC reprogramming, or aging. Normalized Enrichment Score (NES) 0.05 < ^p-adj < 0.1; ^*p-adj < 0.05; ^**p-adj < 0.01; ^***p-adj < 0.001. Benjamini-Hochberg approach. (F) Pictograph of study results showing that both induction of OSK and treatment with C1-6 restore NCC integrity, transcript profiles, and reversing biomarkers of health. Created using https://www.biorender.com.