Bennett R et al. (2025), An epigenetic clock for Xenopus tropicalis.

XB-ART-61399

NPJ Aging 2025 May 22;111:38. doi: 10.1038/s41514-025-00236-x.

Show Gene links Show Anatomy links

An epigenetic clock for Xenopus tropicalis.

Bennett R , Morselli M , Petrova K , Peshkin L , Pellegrini M .

???displayArticle.abstract???
DNA methylation clocks have been widely used for accurate age prediction, but most studies have been carried out on mammals. Here we present an epigenetic clock for the aquatic frog Xenopus tropicalis, a widely used model organism in developmental biology and genomics. To construct the clock, we collected DNA methylation data from 192 frogs using targeted bisulfite sequencing at genomic regions containing CpG sites previously shown to have age-associated methylation in Xenopus. We found highly positively and negatively age-correlated CpGs are enriched in heterochromatic regions marked with H4K20me3 and H3K9me3. Positively age-correlated CpGs are enriched in bivalent chromatin and gene bodies with H3K36me3, and tend to be proximal to lowly expressed genes. These epigenetic features of aging are similar to those found in mammals, suggesting evolutionary conservation of epigenetic aging mechanisms. Our clock enables future aging biology experiments that leverage the unique properties of amphibians.

???displayArticle.pubmedLink??? 40404700
???displayArticle.pmcLink??? PMC12098715
???displayArticle.link??? NPJ Aging
???displayArticle.grants??? [+]

Species referenced: Xenopus tropicalis
GO keywords: DNA methylation involved in embryo development [+]

???attribute.lit??? ???displayArticles.show???

	Fig. 1. Overview of data collection workflow. Xenopus tropicalis frogs of different ages and strains were raised in tanks from birth, a skin punch from each frogs hindlimb webbing was collected for DNA extraction, and targeted bisulfite sequencing was performed to measure DNA methylation levels at a collection of predetermined genomic regions. These targeted regions were designed to capture CpG sites that were previously shown to be highly age-associated in a genome-wide study of Xenopus tropicalis.
	Fig. 2. Epigenetic clock performance.A–D DNA methylation clock nested cross validation results on n = 192 frogs. Models (B) and (D) were constructed to predict the natural logarithm of age. The dotted lines represent predicted age = actual age. MedAE is the median absolute error between predicted and actual age (in years). E Age predictions on a test set of 16 frogs based on a DNA methylation clock trained on the 192 frog training set.
	Fig. 3. Chromatin state analysis of highly age-associated CpGs.A ChromHMM automated chromatin state annotation emission parameters. B Chromatin states of selected CpG sites with high correlation between methylation and age. Log enrichment represents log2 of the fraction of + or - CpG sites in a chromatin state divided by the fraction of all CpG sites in that chromatin state. The reference background is all CpG sites from our TBS dataset with sequencing coverage of at least 10. An asterisk (*) indicates that the adjusted p-value < 0.05 for Fisher’s exact test.
	Fig. 4. Expression of genes proximal to highly age-associated CpGs.A Single-cell RNA sequencing atlas expression levels of the closest gene to significantly negatively correlated CpGs (red), significantly positively correlated CpGs (blue), and all genes (gray). Mean expression is calculated as the mean of log1pCP10k values over all cells in a tissue. Asterisks (*) indicate Benjamini–Hochberg adjusted p-value is < 0.05 for the Wilcoxon rank-sum test of mean expression in the gene list compared to mean expression of all genes. B Log2 fold change in mean expression of genes closest to significantly age-associated CpGs compared to all genes.