Skip to main content
Download PDF

Genome-wide DNA methylation patterns in Daphnia magna are not significantly associated with age

Epigenetics & Chromatin volume 18, Article number: 17 (2025) Cite this article

Abstract

Background

DNA methylation plays a crucial role in gene regulation and epigenetic inheritance across diverse organisms. Daphnia magna, a model organism in ecological and evolutionary research, has been widely used to study environmental responses, pharmaceutical toxicity, and developmental plasticity. However, its DNA methylation landscape and age-related epigenetic changes remain incompletely understood.

Results

In this study, we characterized DNA methyltransferases (DNMTs) and mapped DNA methylation across the D. magna genome using whole-genome bisulfite sequencing. Our analysis identified three DNMTs: a highly expressed but nonfunctional de novo methyltransferase (DNMT3.1), alongside lowly expressed yet functional de novo methyltransferase (DNMT3.2) and maintenance methyltransferase (DNMT1). D. magna exhibits overall low DNA methylation, targeting primarily CpG dinucleotides. Methylation is sparse at promoters but elevated in the first exons downstream of transcription start sites, with these exons showing hypermethylation relative to adjacent introns. To examine age-associated DNA methylation changes, we analyzed D. magna individuals across multiple life stages. Our results showed no significant global differences in DNA methylation levels between young, mature, and old individuals, nor any age-related clustering in dimensionality reduction analyses. Attempts to construct an epigenetic clock using machine learning models did not yield accurate age predictions, likely due to the overall low DNA methylation levels and lack of robust age-associated methylation changes.

Conclusions

This study provides a comprehensive characterization of D. magna’s DNA methylation landscape and DNMT enzymes, highlighting a distinct pattern of exon-biased CpG methylation. Contrary to prior studies, we found no strong evidence supporting age-associated epigenetic changes, suggesting that DNA methylation may have a limited role in aging in D. magna. These findings enhance our understanding of invertebrate epigenetics and emphasize the need for further research into the interplay between DNA methylation, environmental factors, and gene regulation in D. magna.

Background

Daphnia are planktonic crustaceans within the Phyllopoda subclass and the Branchiopoda class [1]. Among over 100 species within the Daphnia genus, Daphnia magna (D. magna) is widely recognized as a model organism in biological research, including studies on pharmaceutical toxicity, reproductive cycles, behavioral ethology, and phenotypic plasticity [2]. DNA methylation, the addition of a methyl group on the 5’ carbon of cytosine, plays a role in transcriptional regulation and phenotypic expression [3]. Previous studies have demonstrated that DNA methylation assessment aids in toxicity evaluation [4]. Given the significance of D. magna as a model organism, examining its epigenomics landscape, particularly DNA methylation pattern and regulation, could be critical [5].

CpG methylation is closely associated with gene silencing and is crucial for development [6, 7]. In Daphnia magna, cytosine methylation levels are relatively low, averaging around 1% across the genome, yet specific regions, such as exons, exhibit higher methylation, with exons 2 to 4 displaying the highest levels [3, 6, 8]. Previous studies, however, have tended to average methylation across different genes for each exon, overlooking the nuanced methylation patterns of individual genes and exons. To address this gap, our study aims to examine both average methylation patterns and detailed variations at the exon and gene levels across different genes.

DNA methyltransferases (DNMTs) are a family of enzymes that epigenetically regulate gene expression by establishing and maintaining CpG methylation patterns [9]. Maintenance methyltransferase DNMT1 replicates DNA methylation patterns to daughter cells during cell divisions, while the functional de novo methyltransferase DNMT3 establishes methyl groups at unmethylated cytosine sites [10, 11]. Our research aims to explore the functional domains of D. magna's DNMTs and their effect on genome-wide and gene-specific methylation patterns.

DNA methylation levels vary with age and influence the functional capability of organs in mammals, positioning DNA methylation-based biomarkers, also known as epigenetic clocks, as effective estimators of biological age [12,13,14]. Epigenetic clocks facilitate understanding the impact of both endogenous (epigenetic drift, etc.) and exogenous stressors on biological aging by comparing epigenetic age to chronological age [12, 14]. In a previous study, an epigenetic clock for D. magna was built with 12 clock CpGs [6]. However, this analysis was conducted using samples from only two age groups: 10-day and 50-day-old. Additionally, they incorporated Daphnia specimens from two different strains and included an interaction term for strain and age in their epigenetic clock. Both the large age gaps and an additional factor of strain potentially diminished the predictive accuracy of their regression model [6]. Thus, further research is needed to characterize age-related changes in DNA methylation and determine the validity of an epigenetic clock for D. magna.

This study aimed to investigate the targeting and age association of DNA methylation in D. magna. We cultured 17 D. magna samples of various ages, performed whole genome bisulfite sequencing, and characterized the DNA methylation patterns at the genome-wide, gene, and exon levels. To better understand the methylation patterns, we analyzed functional domains and mRNA expression levels of DNA methyltransferases. Lastly, we examined age-related changes in methylation patterns and attempted to build an epigenetic clock.

Methods

Sample collection

A heat-tolerant clone of D. magna, IL-MI-8, was originally acquired from the Ebert Lab at the University of Basel, Switzerland. This clone was sourced from a pond in Jerusalem, Israel, as part of the Ebert Lab stock collection. To ensure synchronized cohorts, neonates born within a two-day window were separated from their mothers. Sex was determined at 8- to 9-day-old animals. Female individuals were exclusively used in all experiments conducted in this study. All mothers received adequate nutrition and were cultured under identical conditions at 25 °C within an incubator. All cultures, including mother cultures, neonate cultures, and tank cultures, were maintained in ADaM water [15]. Animals were exposed to a light cycle consisting of 16 h of light followed by 8 h of darkness. Animals were fed a daily suspension of the green alga Scenedesmus obliquus at a concentration of 105 cells/ml (adjusted for population density, with one animal per 20 ml). Every sixth day, the water was replaced, and offspring were manually removed until animals were transferred to the culture platform. The operational protocols for the culture platform are akin to those described in previously published protocols [16].

DNA extraction and WGBS library preparation

Approximately 30–50 ng of purified genomic DNA in 50 µl has been subject to sonication using the Bioruptor Pico (Diagenode) for 15 cycles (30 ON; 90 OFF). Fragmentation patterns have been checked with the D1000HS Assay (Agilent Technologies) on a 4200 TapeStation. The volume of the fragmented DNA has been reduced to 20 µl using a Vacufuge (Eppendorf), then subject to bisulfite conversion using the EZ DNA Methylation-Lightning Kit (Zymo Research). The libraries have been prepared using the Accel-NGS Methyl kit (Swift Biosciences—now xGen Methyl-Seq Library Prep—IDT) according to the manufacturer’s recommendations except for a major modification. Briefly, the denatured BS-converted gDNA was subject to the adaptase, extension, and ligation reaction. Following the ligation purification, the DNA underwent primer extension (98 °C for 1 min; 70 °C for 2 min; 65 °C for 5 min; 4 °C hold) using oligos containing random UMI at the location of the i5 barcode. The extension using a UMI-containing primer allows the tagging of each molecule to remove PCR duplicates and correctly estimate DNA methylation levels. Following exonuclease I treatment and subsequent purification, the libraries were then amplified using a universal custom P5 primer and i7-barcoded P7 primers (initial denaturation: 98 °C for 30 s; 10 cycles of 98 °C for 10 s, 60 °C for 30 s, 68 °C for 60 s; final extension: 68 °C for 5 min; 4 °C hold). The resulting single-indexed libraries were then purified and quantified using the Qubit HS-dsDNA assay, and the quality was checked using the D1000-HS assay (Agilent—TapeStation 4200). The libraries were sequenced as 150 PE on the Illumina NovaSeq6000 platform.

Bisulfite sequencing data processing

An alignment index incorporating a bisulfite-converted sequence for each reference strand was constructed from the Daphnia magna Xinb3 reference genome (BioProject ID: PRJNA624896, D. Ebert, personal communication) and the mitochondrial genome [17], using the BSBolt Index tool v1.6.0 [18]. The FASTQ files were aligned using BSBolt Align (default options) [18]. PCR duplicates were removed with samtools markdup v1.17 (option -r) [19]. Sequence alignment quality was assessed using the CIGAR (compact idiosyncratic gapped alignment report) strings, and sequences with a total number of matches below 50 were excluded from further analysis using samtools view and samtools index [19]. The number of mapped reads in each sample was quantified using samtools flagstat [19].

DNA methylation calling was performed with BSBolt CallMethylation (options: -IO) with the DNA alignment index as a reference, generating CGmap files [18, 20]. The CGmap files were subsequently filtered based on contig coverage and quality. Average coverage before and after filtering for cytosines across each chromosome was obtained using CGmaptools mec v0.1.2 [20]. Global DNA methylation levels were analyzed with CGmaptools mstat [20], and differences in global methylation levels of CpG and non-CpG cytosines were assessed using the Wilcoxon test. DNA methylation levels of genes, exons, and introns were calculated using CGmaptools mtr [20]. Meta-gene plots for genes and exons were generated using CGmaptools bed2fragreg and CGmaptools mfg [20]. Pathway enrichment analysis was performed using the R package clusterProfiler [21]. Genome browser snapshots were taken from Integrative Genomics Viewer (version IGV_2.18.2) [22].

The matrix of common CpG sites was generated using BSBolt AggregateMatrix function (options: -min-coverage 10 -min-sample 0.8 -CG) [18]. To handle missing data, CpG sites absent in some samples were imputed using the BSBolt Impute feature. UMAP was employed to visualize the clustering of samples based on methylated CpG sites. This approach allowed us to examine whether Daphnia samples from the same age group exhibited consistent methylation patterns, with an expectation of tighter clustering within age groups in the UMAP plot [23].

The CpG sites were filtered to only include the ones with the top 20% of the variability. The epigenetic clock model was built with the Lasso function from the Python module sklearn.linear_model and Leave-One-Out Cross Validation [24]. Differentially methylated regions (DMR) analysis was performed using the Metilene package v0.2–8 (options: -m 10 -d 0.001 -t 4 -f 1) [25].

Random permutation was used to assess the statistical significance of differences in methylation levels between age groups at each position within the meta-gene ridge plot. For each position, we calculated the differences in mean methylation levels between the two age groups. We conducted 100,000 random permutations and reassigned sample labels to form two new groups, ensuring that the size of these groups matched the age groups. In each permutation, we calculated the mean methylation difference between these newly assigned groups. p-values were calculated as the proportion of permuted differences as extreme as, or more extreme than, the observed differences, applying a two-tailed test. Significance at each site was defined as p-value < 0.05. This procedure was utilized across three comparisons: mature versus young, old versus young, and old versus mature.

RNA sequencing data processing

Details of an age-specific RNAseq experiment are reported elsewhere [26]. Briefly, RNA was extracted from somatic tissues (whole bodies) of Daphnia magna females from laboratory clone GB-EL75-69 (Basel University Daphnia Stock Collection, Switzerland) of different ages: young, reproducing (age 15–20 days), old, reproductively senescent, and old, reproductively rejuvenated (age both 130–175 days), in 4 replicates each. In all cases, Daphnia samples were sampled within 24 h of molting and egg-laying and freshly laid eggs (if any) were removed from the brood chamber before homogenizing Daphnia. RNAs were extracted using Qiagen RNeasyPlus Mini kit and RNA library preparation was performed using NEBNext Ultra II DirecMonal RNA Library Prep Kit (NEB, Lynn, MA) following the manufacturer's protocols. The libraries were sequenced with Illumina Novoseq 6000, S4 flow cell, PE100. Reads were mapped to D. magna Xinb3 reference transcriptome (BioProject ID: PRJNA624896; D. Ebert and P. Fields, personal communication), and genes with differential expression either between young vs. old Daphnia were identified using DEseq2 [27]. Statistical analysis of RNA expression levels between groups was performed using a two-sided t-test.

DNA methyltransferase analysis

The functional analysis of DNMT3.1, DNMT3.2, DNMT1, and UHRF1 was conducted using InterProScan [28]. Gene sequences were input into the tool to identify and characterize the functional domains within these methyltransferases. Nuclear localization signals (NLS) were predicted using NLStradamus, a Hidden Markov Model (HMM) tool, to assess their potential nuclear targeting [29]. The analysis was conducted using a 2-state HMM static model, and NLS sequences were identified based on a consensus between the Viterbi path and posterior probability estimates exceeding 70%.

Results

D. magna has three DNA methyltransferases

Based on the homology with other organisms’ DNA methyltransferases (DNMTs), D. magna harbors two functional and one non-functional DNMT (Table S1) [9]. It has two de novo DNMTs, DmaDNMT3.1 and DmaDNMT3.2 (Fig. 1A) [10].

Fig. 1
figure 1

The comparison of DNA (C5) methyltransferases and accessory proteins of D. magna and H. sapiens and the gene expression levels under different conditions. A Domain structures of de novo and maintenance DNA methyltransferases and accessory proteins in Daphnia magna and Homo sapiens were predicted using InterProScan and NLStradamus. Colored boxes represent the functional domains, with numbers indicating their respective positions along the protein sequence. Diagonal black-striped boxes indicate a predicted disordered consensus sequence. B Gene expression levels (TPM on log scale) of three DNMT genes (DmaDNMT1, DmaDNMT3.1, and DmaDNMT3.2) in young and old D. magna individuals. Bars represent the mean expression levels, and error bars denote standard deviation (SD). Statistical significance between conditions for each gene was assessed using a two-sided Student’s t-test, with significance indicated as follows: P < 0.001 (***) and nonsignificant (ns). Expression levels of DmaDNMT1 and DmaDNMT3.2 showed no significant differences between the groups, while DmaDNMT3.1 expression was significantly different in young individuals compared to the old group

Full size image

To better understand DmaDNMT and its relationship with the species’ methylation pattern, we used mammalian (human) DNMTs as a reference and compared the homologies between DmaDNMTs and HsDNMTs. Mammalian DNMT3a/b contains a Pro-Trp-Trp-Pro (PWWP) domain, an ATRX-DNMT3A-DNMT3L (ADD) domain, and a DNA-(cytosine-C5)-Methyltransferase (MTase) domain with 6 motifs (I to VI). Non-functional DmaDNMT3.1 is missing 5 critical motifs in the MTase domain, and it is thought to exert similar functions to mammalian DNMT3L. Both DmaDNMT3.1 and HsDNMT3L contain putatively functional ADD domains that do not bind H3K4me3. However, DmaDNMT3.1 owns the PWWP domain that is known to bind methylated lysines, specifically H3K36me2/3 in mammals [11].

Conversely, the functional de novo DNA methyltransferase, DmaDNMT3.2, contains all six DNA methyltransferase motifs and the ADD domain, suggesting that it can sense the methylation status of H3K4 and bind unmethylated H3K4 histones [10]. However, it is missing the N-terminal PWWP domain present in the active mammalian DNMT3A and DNMT3B.

The other putatively active DNA methyltransferase in D. magna is the maintenance methyltransferase DmaDNMT1 (Fig. 1A). Compared to the mammalian counterpart, HsDNMT1, it is missing the N-terminal domains involved in the targeting of the replication foci, including the DNA methyltransferase 1-associated protein-binding domain (DMAP1-BD) and the proliferating cell nuclear antigen-binding domain (PCNA-BD). NLStradamus predicted DmaDNMT1’s nuclear localization signals (NLS) both upstream (position 403–434) and downstream (678–697) of the DRFtD domain, with the upstream NLS aligning with the conserved position in HsDNMT1. Additionally, a phosphatase-specific domain (PAP2-wunen) at the N-terminal region is unique to DmaDNMT1.

To understand the activity of the three DmaDNMTs, we analyzed RNA sequencing data across two age groups: young and old (Fig. 1B). Gene expression levels (TPM) varied significantly across the three DmaDNMTs by a two-sided t-test. The two functional enzymes, DmaDNMT1 and DmaDNMT3.2, exhibited consistently low expression levels with no significant differences between the age groups. By contrast, the non-functional DmaDNMT3.1 showed significantly higher expression in the old group compared to the young group. However, since DmaDNMT3.1 lacks the critical DNA methyltransferase motifs, its expression is unlikely to impact methylation levels.

High-coverage D. magna methylomes

To determine the DNA methylation profiles of Daphnia magna, we obtained two 45-day-old (mature-old) Xinb3 clone individuals, Dap_S1 and Dap_S2. Whole-genome bisulfite sequencing was performed as described in Methods, and subsequent sequence alignment to the D. magna genome and methylation calling was carried out. On average, approximately 79% of the reads mapped primarily to the genome, and this dropped to 56% after removing duplicates and 53% after filtering by CIGAR (compact idiosyncratic gapped alignment report) strings. The Daphnia magna genome comprises 56.6 million cytosine sites across both strands. Applying a minimum coverage threshold of 10 reads per site in the methylation calling step, we successfully identified 41.1 million cytosines, representing approximately 72.6% of the total cytosine sites in the genome. Although the average coverage across contigs was around 60, the distribution demonstrated large variations (Figure S1A).

The original genome assembly of Daphnia magna consisted of one mitochondrial genome [17] and 608 nuclear contigs (BioProject ID: PRJNA624896, D. Ebert, personal communication), many of which are short. Approximately 50% of these contigs are under 20,000 base pairs and 25% are below 10,000 base pairs in length. Some short contigs showed pronounced high coverage and hypermethylation at CpG sites, suggesting that they are outliers and possible contaminants. To improve the quality of methylome, alignment filtration was implemented, retaining only the CpN sites from the top-performing 97 nuclear contigs validated by genetic recombination map, Hi-C data, and optical mapping (BioProject ID: PRJNA624896, D. Ebert, personal communication). Following selective filtration, 38.7 million of these 41.1 million covered cytosine sites (94.1%) were retained for each sample. The mean contig length was 1.3 million base pairs, with an average coverage of approximately 60x (Figure S1B). This process reduced variability across contigs and confirmed the reliability of the remaining methylcytosine sites. Consequently, the cytosine sites chosen for our downstream analysis were situated on the top-performing contigs with a minimum coverage of 10x and an average coverage of 65x (Table S2, S3). These selected sites were included in the analysis even if they did not appear consistently across both samples.

Besides the two deeply sequenced D. magna samples (Dap_S1 and Dap_S2), an additional 15 D. magna samples denoted Dap_D1 to Dap_D16 (Dap_D7 excluded due to low quality) were bisulfite sequenced across different age groups: young (9 days), mature (22 to 27 days), and old (51 to 58 days). Similar approaches of alignment, methylation calling, and filtration were applied to the 15 samples. This approach yielded an average of 10.9 million cytosine sites per sample, each with a minimum coverage of 10x, for our downstream analysis (Table S2, S3).

Mitochondrial genome methylation as a negative control for non-conversion artifacts

Bisulfite treatment converts unmethylated cytosines to uracils, while leaving the methylated cytosines unchanged, allowing methylation to be accurately detected through sequencing. However, bisulfite conversion is not 100% effective, and a small fraction (typically around 1%) of unmethylated cytosines fail to deaminate and are mistakenly identified as methylated [30].

To distinguish between methylcytosines and unconverted cytosines we considered several criteria. If cytosine dinucleotides are methylated, we expect the methylation patterns at individual sites to be consistent across samples. To investigate this, we conducted correlation analyses between the Dap_S1 and Dap_S2 samples across different cytosine contexts. We observed a high correlation (r = 0.949) for CpG sites between the two samples, indicating reproducible methylation patterns. We also expect that methylated CpG dinucleotides are present on both strands, which is often referred to as symmetric methylation. In Dap_S1, we identified 1,100,329 pairs of CpG sites with non-zero methylation on both strands, of which 1,040,273 pairs (94.5%) had a difference in methylation of less than 0.05. Similarly, in Dap_S2, 1,204,966 CpG pairs had non-zero methylation, with 1,129,724 pairs (93.8%) showing a difference below 0.05. This high degree of symmetry between strands further supports the reliability of CpG methylation signals [31]. By contrast, non-CpG contexts exhibited much lower correlations, with CpA at r = 0.338, CpC at r = 0.222, and CpT at r = 0.249. The low correlation levels for non-CpG sites suggest that these are likely artifacts of incomplete bisulfite conversion rather than true methylation.

Moreover, to differentiate between bisulfite non-conversion and true cytosine methylation, we used the mitochondrial genome as a negative control, as previous studies have shown that mitochondrial DNA is generally unmethylated [32] and have utilized this approach to assess true methylation level [33]. Cytosine dinucleotides in the mitochondrial genome exhibited median coverage levels exceeding 100x (Figure S2), and this high coverage allows us to use the mitochondrial genome as a negative control. Dap_S1 and Dap_S2 displayed consistent methylation patterns across all cytosine contexts (Figure S3) with median methylation levels ranging from 0.2% to 0.5%. Non-CpG sites had third-quartile methylation levels between 1.0% and 1.2%, while CpG sites exhibited a slightly higher third quartile of 1.6%. To estimate the non-conversion rate in the nuclear genome, we compared the methylation levels of individual nuclear cytosine sites with those in the mitochondrial genome. The average methylation levels in the mitochondrial genome were comparable to or even higher than those observed in the nuclear genome (Figure S4). Moreover, non-CpG sites in the mitochondrial genome showed a higher percentage of methylation levels exceeding 20% compared to nuclear sites. However, the presence of nuclear mitochondrial sequences (nuMT) [34] in D. magna—mitochondrial DNA fragments inserted into the nuclear genome—may explain some of these effects [35]. These nuMT sequences are fragments of mitochondrial DNA and thus contribute ambiguity to interpreting the data and potentially skewing nuclear methylation measurements by retaining mitochondrial methylation patterns or contributing additional unmethylated cytosines, depending on their conversion rates. To estimate the proportion of nuMTs, we re-mapped reads originally aligned to the mitochondrial genome onto the nuclear genome. We found that 26.1% of mitochondrial-aligned reads in Dap_S1 and 22.7% in Dap_S2 mapped to the nuclear genome with a normalized Alignment Score greater than 90%. This score was calculated by dividing the alignment score from BSBolt Align by the read length. This substantial fraction suggests that a portion of what appears to be mitochondrial methylation may instead originate from nuMTs.

In contrast to non-CpG sites, CpG sites in the nuclear genome displayed higher methylation levels than those in the mitochondrial genome, suggesting that CpG sites in the nuclear genome are methylated rather than artifacts of non-conversion.

D. magna DNA CpG sites are methylated at low levels

To elucidate methylation at a global level, we computed the average methylation levels of the 97 contigs. CpT and CpC sites exhibited an average methylation level of approximately 1.25%, while CpA methylation was slightly lower with 1.1% methylation (Figure S5). By contrast, CpG sites displayed higher average methylation levels, around 2.05% across contigs, whose distribution of methylation was significantly different from that of the non-CpG sites (p-value < 0.0001 in Wilcoxon test).

We further assessed the methylation levels at individual CpN sites and categorized corresponding fractions of methylation levels (Fig. 2). We found that the two samples had comparable methylation levels, and the methylation levels of CpA, CpT, and CpC sites were distributed similarly. Around 40% of the non-CpG sites had a non-zero methylation level below 0.05, while around 55% had no methylation. Approximately 5% of the non-CpG sites had a methylation level above 0.05 and less than 0.2% above 0.20. By contrast, CpG sites are significantly hypermethylated compared to the three non-CpG cytosine contexts in both samples (p-value < 0.0001 in the Wilcoxon test). Approximately 6% of the CpG sites had a methylation level above 0.05 and around 1% of the sites above 0.20. Specifically, 0.5% of the CpG sites have a methylation level above 0.80. This higher methylation frequency in CpG sites could be attributed to the specificity of DNMTs, which primarily target CpG dinucleotides over non-CpG sites [36]. This pattern emphasizes the established role of CpG methylation in invertebrate transcriptional regulation [37], further distinguishing it from non-CpG methylation contexts.

Fig. 2
figure 2

Global DNA methylation levels at CpN sites of two deeply sequenced samples. The count distribution of methylation levels for individual CpN sites is shown. A CpA, B CpC, and D CpT sites exhibit comparably low methylation levels. C CpG sites display a significantly higher degree of methylation overall

Full size image

In addition to examining the overall methylation trends, we analyzed the methylation profiles within specific groups of sequences, including DNA transposons, retrotransposons, LTR retrotransposons, and non-LTR elements. While some short interspersed nuclear elements (e.g., SINE_U, SINE_B2, SINE_Alu) showed localized methylation upstream or downstream of the sequences, most repetitive elements exhibited constant low methylation levels.

D. magna DNA CpG sites are hypomethylated upstream of TSS

We next examined methylation patterns around genes, focusing on the range of 3 kilobases upstream of the transcription start site (TSS) to 3 kilobases downstream of the transcription end site (TES). Both samples, Dap_S1 and Dap_S2, showed consistent trends, highlighting the reproducibility of these patterns. Non-CpG sites maintained a uniform methylation level of approximately 1.5% across the gene and proximal regions (Fig. 3A, B, D), suggesting non-CpG methylation is mainly an artifact and represents background noise. In contrast, CpG methylation displayed greater variability at different positions along genes. Specifically, CpG methylation levels were lowest a few hundred bases upstream of the TSS (1.5% methylated) and sharply increased downstream of the TSS, peaking at nearly 4.5% (Fig. 3C). Subsequently, within the gene, CpG methylation levels gradually decreased, reaching 1.5%, underscoring the enrichment on the 5’ end of the genes.

Fig. 3
figure 3

Gene-level DNA CpN methylation levels in two deeply sequenced samples, spanning 3 kilobases (kb) upstream transcription start sites (TSS) and 3 kb downstream of transcription end sites (TES). Each line depicts the average gene methylation value for A CpA, B CpC, C CpG, and D CpT for a sample. Non-CpG sites exhibited uniformly low methylation levels, whereas CpG sites manifest a distinctive pattern. These sites have lower methylation levels at TSS compared to regions upstream and downstream

Full size image

To minimize positional artifacts from normalizing genes to a uniform distance and to provide more insight into dynamic CpG methylation within genes, we grouped genes into deciles based on their mean methylation levels (e.g., top 10%, 10–20%) and calculated methylation levels centered at the TSS and TES. Dap_S1 (Fig. 4) and Dap_S2 (Figure S6) displayed high methylation in the top 20% of genes (Figure S7) and minimal methylation in the lower 50% of genes (Figure S8). Around the TSS, the top 20% of genes showed markedly higher methylation levels than the rest, with methylation at the 5’ end exceeding 30% in the top 10% and nearly reaching 20% in the 10–20% group. At the same time, methylation in the top 10% of genes peaked at 10% upstream of the TES.

Fig. 4
figure 4

Gene-level CpG methylation patterns centered around TSS and TES for Dap_S1, spanning 3 kilobases (kb) upstream and downstream of transcription start sites (TSS) and transcription end sites (TES). Genes were divided into 10 groups based on their mean methylation levels. Each line represents the average methylation level within each group, centered around A the transcription start site (TSS) and B the transcription end site (TES). For example, Top 20% in legend represents the top 10–20% group

Full size image

We explored the functions of the highly methylated genes and unmethylated genes using pathway enrichment analysis. Human homologs of D. magna genes within the top 10% methylation category across both samples were significantly enriched for fundamental cellular processes in the Gene Ontology Biology Process (GOBP) database, including ribonucleoprotein complex biogenesis, cytoplasmic translation, mitochondrial gene expression, respiratory electron transport chain, methylation, and DNA replication. Unmethylated genes showed no enrichment for any specific pathway.

D. magna Exon CpG sites are more methylated than neighboring intron regions

To further explore the landscape of genomic methylation, we analyzed differential methylation patterns between exons and introns. Our findings reveal that exons contain a higher density of methylated CpG sites (Figure S9). On average, exons showed a methylation rate of approximately 0.17%, whereas nearby intronic regions were significantly less methylated, averaging 0.07% (p < 0.001 in the Wilcoxon test). Among the exons with a mean methylation rate greater than zero, the average methylation rate is 8.7%, significantly higher than the 4.9% observed in methylated introns (p < 0.001 in Wilcoxon test).

We also analyzed methylation levels across exons in genes with at least five exons to understand how exon position affects methylation. We observed that methylation varied significantly with exon positions (Fig. 5A). Specifically, Exon 1 shows a gradual increase in CpG methylation, peaking near its 3’ end, followed by a sharp drop at the transition to the adjacent intron. This pattern of exon hypermethylation and intron hypomethylation is primarily driven by the top 10% of methylated genes (Fig. 5B). Methylation levels continued to rise in Exon 2, reaching a peak of nearly 8% in Exon 3 before decreasing (Fig. 5A).

Fig. 5
figure 5

Exon-level CpG methylation patterns in two deeply sequenced samples. A CpG methylation levels of Exon 1 to Exon 4 from genes with more than 5 exons are demonstrated. Exon 1 shows a gradual increase in CpG methylation, peaking near its 3’ end. Methylation levels continued to rise in Exon 2, reaching a peak of nearly 8% in Exon 3 before decreasing. B Genes were divided into 10 groups based on their mean methylation levels in Exon 1. Each line represents the average methylation level within each group, centered around Exon 1 end site, with 100b up- and downstream of each region. In top 10% genes, there is a sharp drop in methylation transitioning from Exon 1 to the adjacent intron

Full size image

D. magna's global and local CpG methylation patterns are weakly associated with age

To determine how the methylation pattern changes with age, we first conducted a global analysis of CpG methylation levels across three age groups: young, mature, and old. The average methylation levels were 3.426% in the young group, 3.386% in the mature group, and 3.311% in the old group, showing a slight decrease with age; however, none of these differences reached statistical significance (Fig. 6A). We also aggregated methylation values for CpG sites present in at least 80% of samples, imputed missing values, and constructed a CpG matrix. Visualization of this matrix using UMAP did not reveal clear clustering by age group, indicating a general absence of global age-related methylation patterns (Fig. 6B). We further filtered the CpG matrix to retain CpG sites in genes with at least 10% methylation (562 genes in total) in at least 80% of the 17 samples. However, UMAP visualization of the filtered matrix of the filtered matrix also showed no age-related methylation patterns (Figure S10). Furthermore, to identify specific CpG sites that could serve as indicators of age, we trained an epigenetic clock through Lasso Leave-One-Out cross-validation regression (Figure S11). However, the correlation between predicted and actual ages was not significant, indicating that age-associated changes in the methylation of CpG sites were not robust enough to build an accurate epigenetic clock.

Fig. 6
figure 6

CpG methylation patterns across age groups demonstrate weak relation to age. A Methylation levels of individual CpG sites in each group are presented as the frequency distribution. B The UMAP visualization of matrix with CpG sites present in at least 80% of the 17 samples does not reveal any specific age-related clustering. C Meta-gene plot across age groups demonstrates a similar pattern

Full size image

Next, we explored age-related changes in CpG methylation at the gene level. While meta-gene methylation analysis demonstrated a consistent pattern across age groups (Fig. 6C), closer examination of the differences between age groups revealed a weak trend of decreasing methylation with advancing age (Figure S12). However, random permutation tests at each position showed no statistically significant differences across all sites for the three pairwise age group comparisons. We also performed differentially methylated regions (DMR) analysis and linear regression analyses at the individual gene level, using the mean CpG methylation level as the dependent variable and age as the predictor. Neither any regions nor genes showed significant associations with age, indicating a lack of robust gene-level methylation changes.

Discussion

In this study, our primary focus was elucidating the genome-wide and gene-level methylation patterns in Daphnia magna and analyzing the age-associated variations in methylation levels. One key area of interest was the DNA methyltransferases in D. magna. Maintenance methyltransferase DmaDNMT1 and de novo methyltransferase DmaDNMT3.2 were both functional but expressed at low levels. In contrast, the non-functional DmaDNMT3.1 exhibited higher expression levels and a potential association with aging, suggesting functions beyond DNA methylation. For example, DNMT3.1 has been shown to regulate growth and reproduction under starvation conditions [10].

To explore the evolutionary relationships of invertebrate DNA methyltransferases, we compared DmaDNMTs with DNMTs from other invertebrates. Within the phylum Arthropoda, species in the subphylum Crustacea, including Daphnia pulex, Procambarus virginalis (marbled crayfish), Parhyale hawaiensis, Penaeus vannamei, and Hyalella azteca, retain at least one DNMT3. In contrast, several crustaceans, such as Armadillidium vulgare, Calanus finmarchicus, Eurytemora affinis, Lepeophtheirus salmonis, and Tigriopus californicus, have lost DNMT3 [38, 39]. Expanding to the Hexapoda subphylum, which shares a common arthropod ancestor with crustaceans, DNMT3 is largely conserved in bees and ants, with Solenopsis invicta (red imported fire ant) and Apis mellifera (honey bee) possessing one DNMT3 [39,40,41]. However, DNMT3 appears to be absent in most wasps, flies, butterflies, and moths, including Polistes dominula, Drosophila melanogaster, Bombyx mori (silkworm moth), and Danaus plexippus (monarch butterfly) [39, 42, 43]. Moving further from Arthropoda, DNMT3 is also present in other invertebrate phyla. In Mollusca, Crassostrea gigas (oyster) retains functional DNMT3 [44]. Similarly, in Annelida, Capitella teleta and Ophryotrocha fusiformis possess DNMT3, whereas Dinophilus gyrociliatus lacks it [45]. The observed diversity in the presence and functionality of DNMT3 among invertebrates reflects adaptive modifications to their unique genomic and environmental demands, with D. magna representing one variation of this evolutionary landscape.

Most invertebrates possess a maintenance methyltransferase DNMT1 with a functional MTase protein domain. However, certain species, including D. melanogaster, Aedes aegypti, Anopheles gambiae, and Caenorhabditis elegans lack DNMT1 altogether [11, 38,39,40,41,42,43,44,45]. As observed in D. magna, the N-terminal DMAP-1 binding domain has been lost in many invertebrates, including arthropods (A. mellifera), annelids (C. teleta and O. fusiformis), ctenophores (Pleurobrachia bachei), brachiopods (Lingula anatina) [45,46,47]. This supports previous findings that the DMAP-1 binding domain is conserved in chordates but largely absent in invertebrates [48].

We cultured 17 D. magna samples of different ages and conducted whole-genome bisulfite sequencing, achieving a high-coverage methylome with an approximate coverage of 60x. This allows us to analyze the methylation patterns across genome, gene, and exon levels. Globally, D. magna exhibited an average CpG methylation level of 2%. In comparison, mammals such as mice have an average CpG methylation level of around 80% [49]. One possible explanation for this low methylation level is the reduced expression of DmaDNMT1 and the absence of the DMAP-1 binding domain. The DMAP-1 binding domain facilitates DNMT1’s interaction with DMAP-1, a co-repressor that aids its methylation maintenance activity [50]. However, the correlation between the absence of DMAP-1 binding domain and reduced methylation appears weak and does not hold consistently across invertebrates. For example, A. mellifera lacks this domain and maintains a similarly low global CpG methylation levels of around 1% [51]. In contrast, the annelids C. teleta and O. fusiformis also lack the DMAP-1 binding domain but exhibit high methylation levels, with more than 15% of their CpG sites methylated above 80% [45]. These discrepancies suggest the factors contributing to D. magna’s low methylation levels remain unknown.

D. magna has an average gene body methylation of around 3%, slightly higher than the global methylation level. Gene body methylation refers to the addition of methyl groups, typically at CpG sites, within the coding regions of genes. It is often associated with actively transcribed genes, where it may help regulate splicing, reduce spurious transcription, and maintain genome stability [52]. Among eukaryotes, gene body methylation is widespread in animals and plants but is absent in fungi [33, 53]. The mosaic DNA methylation with a high level in gene bodies and lower level in intergenic regions was observed in many other invertebrates, including sea urchin, lancelet, honey bee A. mellifera, green peach aphid Myzus persicae, O. fusiformis, and C. teleta [45, 54,55,56]. However, D. magna’s gene body methylation is lower than the other invertebrates, which may be attributed to the absence of a functional PWWP domain in DNMT3.2. Since the PWWP domain binds to H3K36me2 and H3K36me3, markers typically found in transcribed regions, its absence could lead to a reduced recognition of H3K36-methylated gene bodies by the active DNMT3.2, resulting in low methylation levels [57,58,59]. Supporting this, H3K36me3 enrichment in gene bodies has been observed in other invertebrates, including Nematostella vectensis, Ciona intestinalis, A. mellifera, and Bombyx mori [60].

In examining the methylation patterns across gene bodies and surrounding regions, we observed that the genes are hypomethylated upstream of the TSS and hypermethylated downstream of the TSS, with methylation levels decreasing along the genes. This pattern aligns with findings in the annelids O. fusiformis and C. teleta, where a similar trend of hypomethylation at TSS has been reported [45]. The hypomethylation at the TSS likely results from a mechanism similar to that described in mammals, where the ADD domain of active DNMT3 selectively binds to unmethylated H3K4 regions [61]. As mammalian promoters are typically rich in methylated H3K4, this prevents DNMT3 from binding and consequently blocks DNA methylation in mammalian systems [62]. A comparable mechanism may operate in Daphnia magna, as previous research has shown a high enrichment of H3K4me3 upstream of the actively transcribed Actin gene [9]. Furthermore, a strong correlation between gene expression and H3K4me3 levels at promoter regions in Daphnia pulex further supports this regulatory mechanism in D. magna [63].

We observed significantly higher methylation levels in exons compared to introns, which was also reported in O. fusiformis, C. gigas, and A. mellifera [44, 45, 49]. This differential methylation is likely influenced by nucleosome positioning, as DNA methyltransferases preferentially target nucleosome-bound DNA [64]. Consequently, the enrichment of nucleosomes on exons leads to higher methylation in these regions, suggesting a role for DNA methylation in exon definition and alternative splicing regulation [65, 66].

Our analysis of Daphnia magna samples revealed minor, insignificant decreases in global and gene-specific methylation levels with age as well as no significant age-related clustering. This contrasts with vertebrates, where DNA methylation patterns are highly dynamic and exhibit notable changes during aging [67]. Thus, the relatively constant DNA methylation level of D. magna across age underlines the possibility DNA methylation is less coupled to developmental changes in gene expression than in vertebrates, possibly due to the different structure of the DNA methyltransferases. At the same time, our attempt to build an epigenetic clock failed for D. magna. This failure is likely due to the low overall CpG methylation levels compared to mammals and the small magnitude of methylation changes with age, making it difficult to identify age-associated patterns necessary for clock construction. Indeed, only a few epigenetics clocks have been reported for invertebrates. Recently, an epigenetic clock on Nasonia vitripennis, the jewel wasp, using 19 age-predictive CpG sites was found to have a Spearman’s P of 0.94 for the correlation between predicted and actual age [68]. However, the fact that the clock CpG sites were pre-selected for their age association indicates possible overfitting to their samples. Thus we still lack robust evidence that epigenetic clocks can be generated for invertebrates.

Although no epigenetic clocks were established, we observed a slight decline in CpG site methylation levels with aging in D. magna. This pattern of epigenetic erosion is pronounced in invertebrates during development [69, 70]. For example, in A. mellifera (honey bee), exon methylation decreases from sperm and embryo to drone and worker larvae [71]. Similarly, annelids such as O. fusiformis and C. teleta experience global methylation loss from embryonic to adult ages [45], while deuterostomes like sea urchin and lancelet display lower methylation levels in adulthood compared to earlier developmental stages [54]. This feature of global hypomethylation has been reported as a consequence of heterochromatin loss, a hallmark of aging across diverse eukaryotes [72].

In this study, our primary emphasis was on the D. magna Xinb3 clone. The methylation profiles we elucidated may be slightly altered in other D. magna strains. Future research should explore the methylation patterns in additional clones to achieve a more comprehensive understanding. Given the limited correlation between DNA methylation and chronological age, our future attention will shift towards the influence of environmental factors, including temperature and water composition, on methylation dynamics.

Availability of data and materials

Bisulfite-seq data generated in this study has been deposited within the Gene Expression Omnibus (GEO) repository, accession number GSE267859. The code used for analysis can be found at https://github.com/rainy31liu/Daphnia_meth.

References

  1. Ebert D. Ecology, epidemiology, and evolution of parasitism in Daphnia. National Library of Medicine; 2005.

  2. Tkaczyk A, Bownik A, Dudka J, Kowal K, Åšlaska B. Daphnia magna model in the toxicity assessment of pharmaceuticals: a review. Sci Total Environ. 2021;763: 143038.

    Article  CAS  PubMed  Google Scholar 

  3. Vandegehuchte MB, Lemière F, Vanhaecke L, Berghe WV, Janssen CR. Direct and transgenerational impact on Daphnia magna of chemicals with a known effect on DNA methylation. Comp Biochem Physiol Part C Toxicol Pharmacol. 2010;151(3):278–85.

    Article  Google Scholar 

  4. Watson RE, McKim JM, Cockerell GL, Goodman JI. The value of DNA methylation analysis in basic, initial toxicity assessments. Toxicol Sci. 2004;79(1):178–88.

    Article  CAS  PubMed  Google Scholar 

  5. Bonini NM, Berger SL. The sustained impact of model organisms—in genetics and epigenetics. Genetics. 2017;205(1):1–4.

    Article  CAS  PubMed  Google Scholar 

  6. Hearn J, Plenderleith F, Little TJ. DNA methylation differs extensively between strains of the same geographical origin and changes with age in Daphnia magna. Epigenetics Chromatin. 2021;14:1–14.

    Article  Google Scholar 

  7. Smith ZD, Meissner A. DNA methylation: roles in mammalian development. Nat Rev Genet. 2013;14(3):204–20.

    Article  CAS  PubMed  Google Scholar 

  8. Kvist J, Gonçalves Athanàsio C, Shams Solari O, Brown JB, Colbourne JK, Pfrender ME, et al. Pattern of DNA methylation in Daphnia: evolutionary perspective. Genome Biol Evol. 2018;10(8):1988–2007.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. Lindeman LC, Thaulow J, Song Y, Kamstra JH, Xie L, Asselman J, et al. Epigenetic, transcriptional and phenotypic responses in two generations of Daphnia magna exposed to the DNA methylation inhibitor 5-azacytidine. Environ Epigenetics. 2019;5(3):016.

    Article  Google Scholar 

  10. Nguyen ND, Matsuura T, Kato Y, Watanabe H. DNMT3.1 controls trade-offs between growth, reproduction, and life span under starved conditions in Daphnia magna. Sci Rep. 2021;11(1):7326.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  11. Lyko F. The DNA methyltransferase family: a versatile toolkit for epigenetic regulation. Nat Rev Genet. 2018;19(2):81–92.

    Article  CAS  PubMed  Google Scholar 

  12. Duan R, Fu Q, Sun Y, Li Q. Epigenetic clock: a promising biomarker and practical tool in aging. Ageing Res Rev. 2022;81: 101743.

    Article  CAS  PubMed  Google Scholar 

  13. Horvath S. DNA methylation age of human tissues and cell types. Genome Biol. 2013;14:1–20.

    Article  Google Scholar 

  14. Horvath S, Raj K. DNA methylation-based biomarkers and the epigenetic clock theory of ageing. Nat Rev Genet. 2018;19(6):371–84.

    Article  CAS  PubMed  Google Scholar 

  15. Klüttgen B, Dülmer U, Engels M, Ratte H. ADaM, an artificial freshwater for the culture of zooplankton. Water Res. 1994;28(3):743–6.

    Article  Google Scholar 

  16. Cho Y, Jonas-Closs RA, Yampolsky LY, Kirschner MW, Peshkin L. Intelligent high-throughput intervention testing platform in Daphnia. Aging Cell. 2022;21(3): e13571.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Fields PD, Obbard DJ, McTaggart SJ, Galimov Y, Little TJ, Ebert D. Mitogenome phylogeographic analysis of a planktonic crustacean. Mol Phylogenet Evol. 2018;129:138–48.

    Article  CAS  PubMed  Google Scholar 

  18. Farrell C, Thompson M, Tosevska A, Oyetunde A, Pellegrini M. BiSulfite Bolt: a bisulfite sequencing analysis platform. Gigascience. 2021;10(5):033.

    Article  Google Scholar 

  19. Danecek P, Bonfield JK, Liddle J, Marshall J, Ohan V, Pollard MO, et al. Twelve years of SAMtools and BCFtools. Gigascience. 2021;10(2):008.

    Article  Google Scholar 

  20. Guo W, Zhu P, Pellegrini M, Zhang MQ, Wang X, Ni Z. CGmapTools improves the precision of heterozygous SNV calls and supports allele-specific methylation detection and visualization in bisulfite-sequencing data. Bioinformatics. 2018;34(3):381–7.

    Article  CAS  PubMed  Google Scholar 

  21. Yu G, Wang LG, Han Y, He QY. clusterProfiler: an R package for comparing biological themes among gene clusters. Omics J Integr Biol. 2012;16(5):284–7.

    Article  CAS  Google Scholar 

  22. Thorvaldsdóttir H, Robinson JT, Mesirov JP. Integrative Genomics Viewer (IGV): high-performance genomics data visualization and exploration. Brief Bioinform. 2013;14(2):178–92.

    Article  PubMed  Google Scholar 

  23. McInnes L, Healy J, Melville J. Umap: Uniform manifold approximation and projection for dimension reduction. ArXiv Prepr ArXiv180203426. 2018;

  24. Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, et al. Scikit-learn: Machine learning in Python. J Mach Learn Res. 2011;12:2825–30.

    Google Scholar 

  25. Jühling F, Kretzmer H, Bernhart SH, Otto C, Stadler PF, Hoffmann S. Metilene: fast and sensitive calling of differentially methylated regions from bisulfite sequencing data. Genome Res. 2016;26(2):256–62.

    Article  PubMed  PubMed Central  Google Scholar 

  26. Dua I, Pearson AC, Lowman RL, Peshkin L, Yampolsky LY. Post-senescence reproductive rebound in Daphnia associated with reversal of age-related transcriptional changes. GeroScience. 2024;1:1–20.

    Google Scholar 

  27. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15:1–21.

    Article  Google Scholar 

  28. Jones P, Binns D, Chang HY, Fraser M, Li W, McAnulla C, et al. InterProScan 5: genome-scale protein function classification. Bioinformatics. 2014;30(9):1236–40.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  29. Nguyen Ba AN, Pogoutse A, Provart N, Moses AM. NLStradamus: a simple Hidden Markov Model for nuclear localization signal prediction. BMC Bioinformatics. 2009;10:1–11.

    Article  Google Scholar 

  30. Genereux DP, Johnson WC, Burden AF, Stöger R, Laird CD. Errors in the bisulfite conversion of DNA: modulating inappropriate-and failed-conversion frequencies. Nucleic Acids Res. 2008;36(22):e150–e150.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Bird AP. Use of restriction enzymes to study eukaryotic DNA methylation: II. The symmetry of methylated sites supports semi-conservative copying of the methylation pattern. J Mol Biol. 1978;118(1):49–60.

    Article  CAS  PubMed  Google Scholar 

  32. Mechta M, Ingerslev LR, Fabre O, Picard M, Barrès R. Evidence suggesting absence of mitochondrial DNA methylation. Front Genet. 2017;8:166.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Zemach A, McDaniel IE, Silva P, Zilberman D. Genome-wide evolutionary analysis of eukaryotic DNA methylation. Science. 2010;328(5980):916–9.

    Article  CAS  PubMed  Google Scholar 

  34. Bensasson D, Zhang DX, Hartl DL, Hewitt GM. Mitochondrial pseudogenes: evolution’s misplaced witnesses. Trends Ecol Evol. 2001;16(6):314–21.

    Article  CAS  PubMed  Google Scholar 

  35. Kowal K, Tkaczyk A, Pierzchała M, Bownik A, Ślaska B. Identification of mitochondrial dna (Numts) in the nuclear genome of daphnia magna. Int J Mol Sci. 2020;21(22):8725.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  36. Moore LD, Le T, Fan G. DNA methylation and its basic function. Neuropsychopharmacology. 2013;38(1):23–38.

    Article  CAS  PubMed  Google Scholar 

  37. Suzuki MM, Kerr AR, De Sousa D, Bird A. CpG methylation is targeted to transcription units in an invertebrate genome. Genome Res. 2007;17(5):625–31.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Lewis SH, Ross L, Bain SA, Pahita E, Smith SA, Cordaux R, et al. Widespread conservation and lineage-specific diversification of genome-wide DNA methylation patterns across arthropods. PLoS Genet. 2020;16(6): e1008864.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  39. Engelhardt J, Scheer O, Stadler PF, Prohaska SJ. Evolution of DNA methylation across ecdysozoa. J Mol Evol. 2022;90(1):56–72.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  40. Wang Y, Jorda M, Jones PL, Maleszka R, Ling X, Robertson HM, et al. Functional CpG methylation system in a social insect. Science. 2006;314(5799):645–7.

    Article  CAS  PubMed  Google Scholar 

  41. Kay S, Skowronski D, Hunt BG. Developmental DNA methyltransferase expression in the fire ant Solenopsis invicta. Insect Sci. 2018;25(1):57–65.

    Article  CAS  PubMed  Google Scholar 

  42. Krauss V, Reuter G. DNA methylation in Drosophila—a critical evaluation. Prog Mol Biol Transl Sci. 2011;101:177–91.

    Article  CAS  PubMed  Google Scholar 

  43. Planques A, Kerner P, Ferry L, Grunau C, Gazave E, Vervoort M. DNA methylation atlas and machinery in the developing and regenerating annelid Platynereis dumerilii. BMC Biol. 2021;19(1):148.

    Article  PubMed  PubMed Central  Google Scholar 

  44. Wang X, Li Q, Lian J, Li L, Jin L, Cai H, et al. Genome-wide and single-base resolution DNA methylomes of the Pacific oyster Crassostrea gigas provide insight into the evolution of invertebrate CpG methylation. BMC Genomics. 2014;15:1–12.

    Article  Google Scholar 

  45. Guynes K, Sarre LA, Carrillo-Baltodano AM, Davies BE, Xu L, Liang Y, et al. Annelid methylomes reveal ancestral developmental and aging-associated epigenetic erosion across Bilateria. Genome Biol. 2024;25(1):204.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  46. Dabe EC, Sanford RS, Kohn AB, Bobkova Y, Moroz LL. DNA methylation in basal metazoans: insights from ctenophores. Integr Comp Biol. 2015;55(6):1096–110.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  47. Gerdol M, La Vecchia C, Strazzullo M, De Luca P, Gorbi S, Regoli F, et al. Evolutionary history of DNA methylation related genes in Bivalvia: new insights from Mytilus galloprovincialis. Front Ecol Evol. 2021;9: 698561.

    Article  Google Scholar 

  48. Albalat R. Evolution of DNA-methylation machinery: DNA methyltransferases and methyl-DNA binding proteins in the amphioxus Branchiostoma floridae. Dev Genes Evol. 2008;218:691–701.

    Article  CAS  PubMed  Google Scholar 

  49. Feng S, Cokus SJ, Zhang X, Chen PY, Bostick M, Goll MG, et al. Conservation and divergence of methylation patterning in plants and animals. Proc Natl Acad Sci. 2010;107(19):8689–94.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  50. Lee GE, Kim JH, Taylor M, Muller MT. DNA methyltransferase 1-associated protein (DMAP1) is a co-repressor that stimulates DNA methylation globally and locally at sites of double strand break repair. J Biol Chem. 2010;285(48):37630–40.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  51. Bewick AJ, Vogel KJ, Moore AJ, Schmitz RJ. Evolution of DNA methylation across insects. Mol Biol Evol. 2017;34(3):654–65.

    CAS  PubMed  Google Scholar 

  52. Jones PA. Functions of DNA methylation: islands, start sites, gene bodies and beyond. Nat Rev Genet. 2012;13(7):484–92.

    Article  CAS  PubMed  Google Scholar 

  53. Bewick AJ, Hofmeister BT, Powers RA, Mondo SJ, Grigoriev IV, James TY, et al. Diversity of cytosine methylation across the fungal tree of life. Nat Ecol Evol. 2019;3(3):479–90.

    Article  PubMed  PubMed Central  Google Scholar 

  54. Skvortsova K, Bertrand S, Voronov D, Duckett PE, Ross SE, Magri MS, et al. Active DNA demethylation of developmental cis-regulatory regions predates vertebrate origins. Sci Adv. 2022;8(48):2258.

    Article  Google Scholar 

  55. Suzuki MM, Bird A. DNA methylation landscapes: provocative insights from epigenomics. Nat Rev Genet. 2008;9(6):465–76.

    Article  CAS  PubMed  Google Scholar 

  56. Klughammer J, Romanovskaia D, Nemc A, Posautz A, Seid CA, Schuster LC, et al. Comparative analysis of genome-scale, base-resolution DNA methylation profiles across 580 animal species. Nat Commun. 2023;14(1):232.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  57. Dukatz M, Holzer K, Choudalakis M, Emperle M, Lungu C, Bashtrykov P, et al. H3K36me2/3 binding and DNA binding of the DNA methyltransferase DNMT3A PWWP domain both contribute to its chromatin interaction. J Mol Biol. 2019;431(24):5063–74.

    Article  CAS  PubMed  Google Scholar 

  58. Kolasinska-Zwierz P, Down T, Latorre I, Liu T, Liu XS, Ahringer J. Differential chromatin marking of introns and expressed exons by H3K36me3. Nat Genet. 2009;41(3):376–81.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  59. Weinberg DN, Papillon-Cavanagh S, Chen H, Yue Y, Chen X, Rajagopalan KN, et al. The histone mark H3K36me2 recruits DNMT3A and shapes the intergenic DNA methylation landscape. Nature. 2019;573(7773):281–6.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  60. Nanty L, Carbajosa G, Heap GA, Ratnieks F, van Heel DA, Down TA, et al. Comparative methylomics reveals gene-body H3K36me3 in Drosophila predicts DNA methylation and CpG landscapes in other invertebrates. Genome Res. 2011;21(11):1841–50.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  61. Otani J, Nankumo T, Arita K, Inamoto S, Ariyoshi M, Shirakawa M. Structural basis for recognition of H3K4 methylation status by the DNA methyltransferase 3A ATRX–DNMT3–DNMT3L domain. EMBO Rep. 2009;10(11):1235–41.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  62. Bernstein BE, Mikkelsen TS, Xie X, Kamal M, Huebert DJ, Cuff J, et al. A bivalent chromatin structure marks key developmental genes in embryonic stem cells. Cell. 2006;125(2):315–26.

    Article  CAS  PubMed  Google Scholar 

  63. Kvist J, Athanàsio CG, Pfrender ME, Brown JB, Colbourne JK, Mirbahai L. A comprehensive epigenomic analysis of phenotypically distinguishable, genetically identical female and male Daphnia pulex. BMC Genomics. 2020;21:1–17.

    Article  Google Scholar 

  64. Chodavarapu RK, Feng S, Bernatavichute YV, Chen PY, Stroud H, Yu Y, et al. Relationship between nucleosome positioning and DNA methylation. Nature. 2010;466(7304):388–92.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  65. Ehrlich M, Lacey M. DNA methylation and differentiation: silencing, upregulation and modulation of gene expression. Epigenomics. 2013;5(5):553–68.

    Article  CAS  PubMed  Google Scholar 

  66. Shayevitch R, Askayo D, Keydar I, Ast G. The importance of DNA methylation of exons on alternative splicing. RNA. 2018;24(10):1351–62.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  67. Meehan RR, Stancheva I. DNA methylation and control of gene expression in vertebrate development. Essays Biochem. 2001;37:59–70.

    Article  CAS  PubMed  Google Scholar 

  68. Brink K, Thomas C, Jones A, Chan T, Mallon E. Exploring the ageing methylome in the model insect, Nasonia vitripennis. BMC Genomics. 2024;25(1):305.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  69. Kane AE, Sinclair DA. Epigenetic changes during aging and their reprogramming potential. Crit Rev Biochem Mol Biol. 2019;54(1):61–83.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  70. Yu G, Wu Q, Gao Y, Chen M, Yang M. The epigenetics of aging in invertebrates. Int J Mol Sci. 2019;20(18):4535.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  71. Harris KD, Lloyd JP, Domb K, Zilberman D, Zemach A. DNA methylation is maintained with high fidelity in the honey bee germline and exhibits global non-functional fluctuations during somatic development. Epigenetics Chromatin. 2019;12:1–18.

    Article  CAS  Google Scholar 

  72. Ciccarone F, Tagliatesta S, Caiafa P, Zampieri M. DNA methylation dynamics in aging: how far are we from understanding the mechanisms? Mech Ageing Dev. 2018;174:3–17.

    Article  CAS  PubMed  Google Scholar 

Download references

Funding

L.P. was supported by the quadraScope Ventures fund and McKenzie Family Charitable Trust.

Author information

Authors and Affiliations

  1. Department of Molecular, Cell and Developmental Biology, University of California, Los Angeles, CA, 90024, USA

    Ruoshui Liu & Matteo Pellegrini

  2. Department of Chemistry, Life Sciences, and Environmental Sustainability, University of Parma, Parma, Italy

    Marco Morselli

  3. Department of Biological Sciences, East Tennessee State University, Johnson City, TN, 37614, USA

    Lev Y. Yampolsky

  4. Systems Biology, Harvard Medical School, Boston, MA, 02115, USA

    Leonid Peshkin

Authors
  1. Ruoshui Liu
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Marco Morselli
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Lev Y. Yampolsky
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Leonid Peshkin
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Matteo Pellegrini
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

R.L. wrote the main manuscript text and prepared figures. M.M. generated library and sequenced data. L.Y. provided gene expression data. L.P. conducted sample collection. L.P. and M.P. supervised analysis and the writing of manuscript. All authors reviewed the manuscript.

Corresponding authors

Correspondence to Leonid Peshkin or Matteo Pellegrini.

Ethics declarations

Competing interest

The authors declare no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Liu, R., Morselli, M., Yampolsky, L.Y. et al. Genome-wide DNA methylation patterns in Daphnia magna are not significantly associated with age. Epigenetics & Chromatin 18, 17 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13072-025-00580-y

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13072-025-00580-y

Keywords

Epigenetics & Chromatin

ISSN: 1756-8935