Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Jul 28;15(1):6357.
doi: 10.1038/s41467-024-50725-y.

Gene body DNA hydroxymethylation restricts the magnitude of transcriptional changes during aging

James R Occean  1 , Na Yang  1 , Yan Sun  2 , Marshall S Dawkins  1 , Rachel Munk  1 , Cedric Belair  1 , Showkat Dar  1 , Carlos Anerillas  1 , Lin Wang  1 , Changyou Shi  1 , Christopher Dunn  3 , Michel Bernier  4 , Nathan L Price  4 , Julie S Kim  2 , Chang-Yi Cui  1 , Jinshui Fan  5 , Moitrayee Bhattacharyya  6 , Supriyo De  5 , Manolis Maragkakis  1 , Rafael de Cabo  4 , Simone Sidoli  2 , Payel Sen  7
Affiliations

Gene body DNA hydroxymethylation restricts the magnitude of transcriptional changes during aging

James R Occean et al. Nat Commun. .

Abstract

DNA hydroxymethylation (5hmC), the most abundant oxidative derivative of DNA methylation, is typically enriched at enhancers and gene bodies of transcriptionally active and tissue-specific genes. Although aberrant genomic 5hmC has been implicated in age-related diseases, its functional role in aging remains unknown. Here, using mouse liver and cerebellum as model organs, we show that 5hmC accumulates in gene bodies associated with tissue-specific function and restricts the magnitude of gene expression changes with age. Mechanistically, 5hmC decreases the binding of splicing associated factors and correlates with age-related alternative splicing events. We found that various age-related contexts, such as prolonged quiescence and senescence, drive the accumulation of 5hmC with age. We provide evidence that this age-related transcriptionally restrictive function is conserved in mouse and human tissues. Our findings reveal that 5hmC regulates tissue-specific function and may play a role in longevity.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1
Fig. 1. Global accumulation of 5hmC in the aging liver.
A Schematic of experimental procedure and tissues used to profile global levels of 5mC and 5hmC by DNA mass spectrometry (nLC-MS/MS). Data are presented as mean ± SEM; statistical significance was assessed using two-sided unpaired Welch’s t-test with Holm-Šídák correction for multiple comparisons. B Dot blot for 5hmC using increasing amounts of gDNA isolated from young and old (n = 4 each) mouse livers; + control is 200 ng of young mouse hippocampus gDNA, – control is water. Signal quantifications are shown on the right. Data are presented as mean ± SD; statistical significance was assessed using two-way ANOVA with Šídák correction for multiple comparisons. AU represents arbitrary fluorescence units. C Representative immunofluorescence microscopy for 5hmC in young and old (n = 2 each) sex-matched liver sections. The mean 5hmC signal intensity per nucleus is quantified on the right, using data from 10 fields of view for each of the two young and two old biological replicates. Horizontal bar represents median; statistical significance was assessed using two-sided unpaired Welch’s t-test. Source data are provided as a Source Data file. Illustration credit: Endosymbiont GmbH.
Fig. 2
Fig. 2. 5hmC accumulates at genic regions associated with hepatic metabolism during aging.
A Principal component analysis (PCA) plot using input subtracted 5hmC bigWig files of young and old (n = 4 each) mice liver. B Volcano plot of differentially hydroxymethylated regions (DHMRs) between old and young (n = 4 each) mouse liver; identified by QSEA with an FDR < 0.05. Hypo DHMRs (FC ≤ −2) are regions with less enrichment in the old and hyper DHMRs (FC ≥ 2) are regions with higher enrichment in the old. C Metaplots of young and old (n = 4 each) mouse liver 5hmC signal at the DHMRs identified by QSEA. D Example genome browser tracks for mouse liver hyper DHMRs (Ppig and an intergenic region) and hypo DHMRs (Rbm47 and Car5a). E Gene ontology (GO) terms associated with the DHMRs from (B) using GREAT. The top 5 biological process terms with FDR < 0.05 are shown. F Pie charts showing CpG and genic/intergenic annotations of the DHMRs from (B). G Metaplots of young and old (n = 4 each) mouse liver 5hmC signal over the gene bodies of all mm10 genes; signal quantifications are shown on the side. Statistical significance was assessed using two-sided unpaired Welch’s t-test. For the box plot, the horizontal line within each box represents the 50th, while the bounds of the box depict the 25th and 75th percentile of the data. The whiskers extend to the minima (the smallest value within 1.5 times the interquartile range (IQR) below the first quartile, excluding outliers) and the maxima (the largest value within 1.5 times the IQR above the third quartile, excluding outliers). Source data are provided as a Source Data file.
Fig. 3
Fig. 3. Gene body 5hmC restricts the magnitude of transcriptional changes during aging.
A Metaplots of merged young and old (n = 4 each) mouse liver 5hmC signal over gene bodies with low (n = 6,340), intermediate (n = 6,339), and high (n = 6,339) average mRNA counts for young (left) and old (right) samples (n = 3 each). B Correlation between average gene body 5hmC signal (100 ranked groups) and variance in mRNA FC (old vs young) among the genes per group for young (left) and old (right). ρ = Spearman’s correlation coefficient, p-values were derived from Spearman’s rank correlation. C Box plots showing mRNA FC of old vs young (n = 3 each) for genes downregulated or upregulated with age. D Metaplots of young and old (n = 4 each) mouse liver 5hmC signal over gene bodies in (C). Quantifications are depicted below the plot; statistical significance was assessed using two-sided unpaired Welch’s t-test. E Same as (C) except for genes with minimal or maximal expression change between old and young (n = 3 each). F Metaplots of young and old (n = 4 each) mouse liver 5hmC signal over gene bodies in (E). Quantifications are depicted below the plot; statistical significance was assessed using two-sided unpaired Welch’s t-test. G Metaplots of young and old (n = 4 each) mouse liver 5mC signal over gene bodies in (E) with minimal (left) and maximal (right) expression change with age. Quantifications are depicted below the plot; statistical significance was assessed using two-sided unpaired Welch’s t-test. H Box plots showing the distribution of various genic features for the genes with minimal and maximal expression changes between old vs young (n = 3 each) mice. Statistical significance was assessed using two-sided unpaired Welch’s t-test. For all box plots (C–H), the horizontal line within each box represents the 50th, while the bounds of the box depict the 25th and 75th percentile of the data. The whiskers extend to the minima (the smallest value within 1.5 times the IQR below the first quartile, excluding outliers) and the maxima (the largest value within 1.5 times the IQR above the third quartile, excluding outliers). Source data are provided as a Source Data file.
Fig. 4
Fig. 4. Alternative splicing mediates 5hmC’s transcriptionally restrictive function through decreased binding of splicing factors.
A Schematic of oligo mass-spec experimental procedure. B Volcano plot showing differentially enriched proteins in old mice for the 5hmC oligo 1 pull-down vs input (n = 4 each). Some significantly enriched or de-enriched interactors are labeled. C Same as (B) except for the 5hmC oligo 2 pull-down vs input (n = 4 each) in the old. D GO terms associated with proteins depleted/de-enriched or enriched for oligo 1 in the old 5hmC vs young C and mC (n = 4 each) comparison. E GO terms associated with proteins depleted/de-enriched or enriched for oligo 2 in the old 5hmC vs old C and mC (n = 4 each) comparison (top) and the old 5hmC vs young C and mC (n = 4 each) comparison (bottom). F Number of differential splicing events detected in RNA-seq data between old and young (n = 3 each) samples at p < 0.05 using rMATS (top). Number of events and the unique number of genes are indicated. Number of differential splicing events grouped by increasing gene body 5hmC signal in old (middle). Number of differential splicing events grouped by minimal or maximal expression change with age (bottom). G Bar plots showing differential isoform usage from dRNA-seq results with young and old (n = 4 each) samples for genes with minimal and maximal expression changes with age at indicated p-value thresholds derived from the rMATS statistical model (H) Transcript length (top) and poly A length (bottom) distribution for genes that undergo minimal or maximal expression changes between old and young (n = 4 each); statistical significance was assessed using Mann–Whitney U test. For (B-E), statistical differences for each protein were assessed using Welch’s t-test (if the F-test p-value was <0.05); otherwise, the standard Student’s t-test was used. Source data are provided as a Source Data file. Illustration credit: Endosymbiont GmbH.
Fig. 5
Fig. 5. Quiescence and senescence drive the increase of 5hmC with age and impact ATP production.
A Normalized mRNA counts in young and old (n = 3 each) mouse liver for 5hmC-relevant enzymes. Data are presented as mean ± SEM; statistical significance was assessed using multiple two-sided unpaired t-test with FDR correction (Benjamini, Krieger, and Yekutieli). B TET activity assay in young and old (n = 4 each) mouse liver lysates. C Dot blot for 5hmC signal using gDNA isolated from proliferating and contact inhibition-induced quiescent HepG2 cells (n = 3 independent cell cultures sourced from the same vial).  + control is 200 ng of young mouse hippocampus gDNA, – control is water. Quantifications are depicted below. D ATP production assay using proliferating and contact-inhibited quiescent HepG2 cells (n = 3 technical replicates for each of 3 independent cell cultures sourced from the same vial). E TMRM mitoprobe assay with proliferating and contact-inhibited quiescent HepG2 cells (n = 3 independent cell cultures sourced from the same vial). F Dot blot of 5hmC using gDNA isolated from proliferating and IRIS, ETIS, and OSIS WI-38 cells (n = 3 independent cell cultures sourced from a single vial). +control is 50 ng of young mouse hippocampus gDNA, – control is water. Quantifications are shown below. G Dot blot for 5hmC using gDNA isolated from HepG2 cells treated with vitamin C (n = 2 independent cell cultures sourced from a single vial). + control is 200 ng of young mouse hippocampus gDNA, – control is water. Quantifications are depicted below. H ATP production assay using proliferating and vitamin C treated HepG2 cells (n = 3 technical replicates for each of 3 independent cell cultures sourced from the same vial). I TMRM mitoprobe assay with proliferating and vitamin C treated HepG2 cells (n = 3 independent cell cultures sourced from the same vial); statistical significance was assessed using two-sided unpaired Welch’s t-test. For panels (B–I), data are presented as mean ± SD. Statistical significance was assessed using two-sided unpaired Welch’s t-test, except for dot blots (C, F, G), which used two-way ANOVA with Tukey’s multiple comparisons post-hoc test. AU represents arbitrary fluorescence units. Source data are provided as a Source Data file.
Fig. 6
Fig. 6. Altering 5hmC levels affects transcriptional magnitude.
A Schematic for 70% partial hepatectomy. B Relative global 5hmC signal for young and old (n = 3 each) mouse liver samples at the indicated times. Data are presented as mean ± SD. Statistical significance was assessed using two-way ANOVA with Geisser-Greenhouse correction. C Metaplot of 5hmC signal over bodies of all mm10 genes for indicated groups. Quantifications are shown on the side. D Metaplot of 5hmC signal across genes bodies with minimal (left) and maximal (right) expression changes between old pre-surgery vs young pre-surgery (n = 3 each) mRNA comparisons. Quantifications are shown below. For (C-D), statistical significance was assessed using one-way ANOVA with Tukey’s multiple comparisons post-hoc test. E Box plots showing transcriptional changes for “genes with minimal change with age” (from D, left). Statistical significance was assessed using Mann–Whitney U test. F Box plots showing transcriptional changes for indicated group comparisons. Statistical significance was assessed using a Mann–Whitney U test with FDR correction (Benjamini-Hochberg). G Schematic showing vitamin C treatment in T24 bladder cancer cells from Peng et al.. H Metaplot of 5hmC signal over the bodies of all hg19 genes in vitamin C-treated T24 cells and untreated controls (n = 1 each). I Box plots showing absolute mRNA FC distribution of genes with minimal and maximal expression changes in vitamin C-treated T24 cells vs untreated controls (n = 2 independent cell cultures). Below, metaplot of 5hmC signal across gene bodies with minimal (left) and maximal (right) expression changes. J Box plots showing transcriptional changes for all genes in vitamin C-treated T24 cells vs untreated controls (n = 2 independent cell cultures per group). Statistical significance was assessed using one-way ANOVA. For all box plots, the horizontal line within each box represents the 50th, while the bounds of the box depict the 25th and 75th percentile of the data. The whiskers extend to the minima (the smallest value within 1.5 times the IQR below the first quartile, excluding outliers) and the maxima (the largest value within 1.5 times the IQR above the third quartile, excluding outliers). Source data are provided as a Source Data file. Illustration credit: Endosymbiont GmbH.
Fig. 7
Fig. 7. Human tissues also show 5hmC-mediated transcriptional restriction.
A Schematic outline of procedure, age groups, and tissues chosen from the GTEx project. B Metaplots of 5hmC signal from He et al. over gene bodies of brain-differential and brain-specific genes, C heart-differential and heart-specific genes, and (D) liver-differential and liver-specific genes; statistical significance was assessed using Mann–Whitney U test. E Metaplots of 5hmC signal from Cui et al. over gene bodies of brain-differential and brain-specific genes, (F) heart-differential and heart-specific genes, and (G) liver-differential and liver-specific genes; statistical significance was assessed using Mann–Whitney U test. H Model illustrating the transcriptionally restrictive role of 5hmC and its propensity to downregulate tissue-specific functions with increasing age. Source data are provided as a Source Data file. Illustration credit: Endosymbiont GmbH.
See this image and copyright information in PMC

Update of

Similar articles

See all similar articles

References

    1. Kayo, T., Allison, D. B., Weindruch, R. & Prolla, T. A. Influences of aging and caloric restriction on the transcriptional profile of skeletal muscle from rhesus monkeys. Proc. Natl Acad. Sci. USA98, 5093–5098 (2001). 10.1073/pnas.081061898 - DOI - PMC - PubMed
    1. Linford, N. J. et al. Transcriptional response to aging and caloric restriction in heart and adipose tissue. Aging Cell6, 673–688 (2007). 10.1111/j.1474-9726.2007.00319.x - DOI - PubMed
    1. Roy, A. K. et al. Impacts of transcriptional regulation on aging and senescence. Ageing Res. Rev.1, 367–380 (2002). 10.1016/S1568-1637(02)00006-5 - DOI - PubMed
    1. Soreq, L. et al. Major shifts in glial regional identity are a transcriptional hallmark of human brain aging. Cell Rep.18, 557–570 (2017). 10.1016/j.celrep.2016.12.011 - DOI - PMC - PubMed
    1. Booth, L. N. & Brunet, A. The aging epigenome. Mol. Cell62, 728–744 (2016). 10.1016/j.molcel.2016.05.013 - DOI - PMC - PubMed

LinkOut - more resources