IntroductionEccDNA (extrachromosomal circular DNA) of chromosomal origin is a latent source of unstable genetic variation in eukaryotic cells. eccDNA has been found to originate from any region of the eukaryotic genomes, including intergenic regions, fragments of genes, and intact genes1,2,3,4. The latter is especially important in the evolution of cancer because oncogenes can rapidly increase their copy number and expression levels if they are found within eccDNA (often referred to as ecDNA)5,6,7,8,9. However, little is known about the features that determine the formation of eccDNA in healthy animal tissues.eccDNA can be a result of DNA damage10. Genomic instability is regarded as a hallmark of aging11, whereby DNA mutations such as single nucleotide variants (SNVs) and indels accumulate with age12 and even lead to mammalian tissue-specific signatures13. It is still unknown whether the genomic burden of eccDNA increases in mammals during aging. Furthermore, studies in humans, mice, and birds have revealed a high frequency of eccDNA from genic regions2,14,15, suggesting an effect of transcription on eccDNA formation, which is in line with a similar relationship previously shown for selected genes in yeast16. Therefore, this study aims to investigate whether eccDNA is associated with transcription and how aging impacts the level and origin of eccDNA in healthy mammalian tissues. To achieve this, we generated a comprehensive atlas of eccDNA across different tissues and ages in the house mouse (Mus musculus).In this work, we assemble the atlas of eccDNA in mice and demonstrate that the number of eccDNA increases logarithmically with transcript levels, providing strong evidence that eccDNA formation is linked to transcriptional activity. Our study further reveals that some tissue-specific genes, which have numerous splice forms and high intron density, are partially protected from transcription-induced circularization. Additionally, we uncover that eccDNA does not accumulate in tissues as animals age. In summary, we provide insights into the mechanisms driving gene evolution and suggest important avenues for research in eccDNA.ResultsAn atlas of eccDNA in healthy miceWe adapted the Circle-Seq protocol2,17 to purify eccDNA from a total of 205 samples from wild-type C57BL/6NRj male mice. These samples included tissues isolated from the brain (cortex, n = 26; and hippocampus, n = 28), inguinal subcutaneous adipose tissue (SAT, n = 27), also referred here as Sfat, epididymal visceral adipose tissue (VAT, n = 27), also referred here as Vfat, liver (n = 27), pancreas (n = 24) and skeletal muscle (n = 25). Samples were collected from animals of three different ages: 3-month-old (young), 12-month-old (adult), and 22-month-old (old) mice (Fig. 1a and Supplementary Data 1). Additionally, we also purified eccDNA from mouse embryos stage E17.5 (neuronal tissue: n = 11, non-neuronal tissue: n = 10) from three different litters that showed no differences in biometric measurements (Supplementary Fig. 1a–d).Fig. 1: eccDNA atlas overview.a High Molecular Weight DNA was extracted from seven different tissues across 3, 12, and 22-month-old wildtype mice. Mitochondrial DNA was linearized using CRISPR-Cas9, and linear DNA was removed with exonuclease RecBCD enzyme. Remaining exonuclease resistant fraction was amplified using phi29 enzyme and sequenced using Circle-Seq. b Density size distribution of all detected eccDNA across samples, with a pie chart representing percentage in each size group. c Density size distribution of segment from 0 to 2 kilobase, with grey lines indicating periodicities for each peak. d Number of unique eccDNA detected in brain (cortex and hippocampus) and non-brain (muscle, liver, visceral adipose, subcutaneous adipose and pancreas) across embryo (n = 10, 11), 3 (n = 12, 49), 12 (n = 12, 40), and 22-month-old (n = 7, 13) age groups. The eccDNA number is normalized to million reads per million genomes and the number of genomes was estimated using averages between Qubit and qPCR estimates (see methods). Significance was assessed using two-side Kruskal–Wallis test with pairwise comparisons and Bonferroni correction (embryo: p = 0.057, 3-month-old: p = 0.98, 12-month-old: p = 0.32, 22-month-old: p = 0.99). The median is marked with a white line, with margins showing the interquartile range (IQR), whiskers extending up to 1.5 IQR, and outliers showing as individual data points. Only samples with ≥70% of reads mapped to the genome were included. e Number of unique eccDNA detected per tissue across 3 (n = 7, 5, 10, 11, 8, 9, 11), 12 (n = 5, 7, 6, 9, 7, 8, 10) and 22-month-old (n = 5, 2, 4, 2, 2, 2, 3) age groups. eccDNA number is normalized similarly. The eccDNA number is normalized to million reads per million genomes and the number of genomes was estimated using averages between Qubit and qPCR estimates. Significance was assessed using two-side Kruskal–Wallis test with pairwise comparisons and Bonferroni correction (tissue group: p = 0.09, age group: p = 0.33). f, g Number of eccDNA per each chromosome as a function of chromosome length (f) and number of eccDNA per gene as a function of gene length (g) with two-sided t test with Pearson regression. Pearson correlation (R), p-value, and gray shadow representing 95% confidence interval of the regression are shown.Full size imageThe 22-month-old group showed a distinctive aging phenotype compared to the 3-month-old and 12-month-old groups. The body weight of 22-month-old mice (44.4 ± 7.0 g) was significantly higher than 12-month-old mice (28.6 ± 1.8 g) and 3-month-old mice (25.9 ± 1.5 g) (Supplementary Fig. 1e). The aged animals also showed a decrease in white blood cells in circulation (3.35 × 106 ± 1.1 × 106 cells/mL) compared to 12-month-old mice (5.46 × 106 ± 0.92 cells/mL) and 3-month-old mice (5.1 × 106 ± 1.09 cells/mL) (Supplementary Fig. 1f).We included at least seven biological replicates from each of the 7 tissues and 3 ages. We extracted approximately 106 cells from each sample (Supplementary Fig. 1g). Nuclear eccDNA was enriched from the samples by linearization of mitochondrial DNA in vitro using a CRISPR-Cas9 system with two sgRNAs specific for mitochondrial DNA, and all linear chromosomal and mitochondrial DNA were subsequently removed using exonuclease V17. The remaining eccDNA was amplified by rolling circle amplification with the φ29 polymerase and sequenced with 150 bp-paired-end sequencing with at least 70 million reads per sample. The reads were subsequently aligned to the mouse reference genome (GRCm38/mm10) using the BWA-MEM algorithm18. eccDNA was detected by the Circle-Map pipeline19 and supported by at least two soft-clipped reads mapping around the circle junction (or one soft-clipped read and at least one discordant read) and at least 95% mean base coverage within the eccDNA detection coordinates. Circle-Map was chosen for the identification of eccDNA because it is the most commonly used eccDNA pipeline, and it out-performs other pipelines19. We tested whether the majority of the eccDNA in the samples had been detected by randomly sampling and mapping fractions of sequence reads (20%, 40%, 60%, 80%, 99%) from the liver, cortex, and hippocampus of three different ages. Saturation plots with the eccDNA count at each sequencing depth showed that the current sequencing depth was sufficient to record most of the eccDNA in the samples (Supplementary Fig. 1h–j).Chromosomes produce eccDNA numbers in proportion to their sizeWe detected 567,963 high-confidence eccDNAs distributed across different tissues and ages (Supplementary Data 2). Of these, 16.8% were below 2000 bp in size, which has previously been designated microDNA14. The average size of eccDNA detected in the study was 4786.5 bp (Fig. 1b), and the size range of eccDNA varied from