Direct RNA sequencing reveals multi-layered regulation of the aging transcriptome

Direct RNA sequencing reveals multi-layered regulation of the aging transcriptome | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Direct RNA sequencing reveals multi-layered regulation of the aging transcriptome Sanqi An, Haisong Qin, Shujie Zhang, Lizi Cheng, Ke Ye, Yawen Qi, and 6 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7521909/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 11 You are reading this latest preprint version Abstract BACKGROUND: Aging involves complex molecular alterations across tissues; however, a comprehensive understanding of epitranscriptomic dynamics remains elusive. RESULTS: Here, leveraging high-quality dRNA-seq, we simultaneously constructed a multi-tissue (lung, skeletal muscle, heart, and hippocampus), cross-species (mouse and human) dynamics atlas of RNA features during aging, namely N 6-methyladenosine (m 6 A), 5-methylcytidine (m 5 C), pseudouridine (Ψ), inosine, alternative polyadenylation (APA), and splicing. Our analysis revealed widespread non-linear, age-dependent shifts across these RNA features, exhibiting both conserved patterns and striking tissue specificity. Strikingly, a pronounced, transient peak in these shift events of RNA features occurred at mid-life (12 months of age in mice, 46 years of age in human), and affected the expression of aging-related, mitochondrial, and metabolic genes. CONCLUSIONS: Collectively, we present a high-resolution atlas of m 6 A, m 5 C, pseudouridine and inosine modifications, alternative splicing and poly(A) site usage across tissues and aging stages, derived from direct RNA sequencing. This resource offers an unprecedented foundation for dissecting the RNA-centric mechanisms that govern mammalian aging. RNA modification aging Nanopore direct RNA sequencing alternative polyadenylation alternative splicing Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Aging is a complex and multifactorial process of physiological changes strongly associated with nearly all diseases, including nearly all chronic diseases and cancers[ 1 ]. Molecular alterations during aging have been documented at genomic (e.g., DNA methylation [ 2 , 3 ]), proteomic [ 4 – 6 ], metabolomic [ 7 , 8 ], and even in microbial community levels [ 9 – 11 ]. Yet, despite the well-known changes in RNA expression levels, our understanding of the aging patterns at the full spectrum of RNA characteristics (e.g., RNA modifications, RNA splicing, RNA editing) within individual multi-organs remains largely uncharted territories. Over 170 distinct types of chemical modifications on RNA molecules have been reported [ 12 ]. Among these, N 6-methyladenosine (m 6 A), 5-methylcytidine (m 5 C), pseudouridine (Ψ), and inosine are the most prevalent and extensively studied. These modifications precisely regulate gene expression in context-dependent manner, contributing to both organ-specific aging and systemic aging, and influencing the pathogenesis of tissue-specific diseases [ 13 , 14 ]. For instance, it has been revealed that METTL3-mediated m 6 A methylation maintains skeletal muscle and contributes to muscle degeneration [ 15 ]. Additionally, the m 6 A atlas of brain tissue across mice and human reveals conserved synaptic transcript hypomethylation, thereby accelerating cognitive decline in aging and Alzheimer’s disease [ 16 ]. Despite rapid progress in this field, our understanding of the complexity of RNA modifications and their potential crosstalk during aging remains fragmented. Beyond RNA modifications, alternative cleavage and polyadenylation (APA) is profoundly implicated in aging. APA governs the selection of polyadenylation sites (PASs) in precursor mRNAs (pre-mRNAs) and influences transcript stability, localization, and translation efficiency [ 17 , 18 ]. Of note, pervasive shortening of 3’UTRs through preferential proximal poly(A) site usage is a hallmark of aging across diverse tissues [ 19 , 20 ]. This phenomenon remodels post-transcriptional regulatory networks by eliminating miRNA and RBP binding sites, driving age-related functional decline(21). APA-mediated shortening of the 3’UTR, culminating in ribosomal dysfunction and progressive muscle atrophy [ 21 ]. Various methods for mapping PASs genome-wide exist (e.g., 3’RNA-seq [ 22 ], PAS-seq [ 23 ]). Analogous to APA, ASgenerates RNA isoform diversity in > 90% of human protein-coding genes, exponentially increasing proteome variation and critically regulating molecular pathways [ 24 , 25 ]. AS dysregulation is implicated in aging and age-related diseases [ 26 ]. In Frontotemporal Dementia, age-dependent mis-splicing of MAPT (tau isoforms) shifts the 3R/4R tau isoform balance toward aggregation-prone 4R-tau, driving neurofibrillary tangle formation in hippocampus [ 27 ]. Therefore, age-associated AS dysregulation in critical genes causes protein dysfunction and speeds up disease pathogenesis, especially in vulnerable tissues. Aging is intrinsically nonlinear and its trajectories differ across organs [ 28 ]. Ding et al [ 29 ] developed tissue-specific proteomic aging clocks spanning 13 human tissues, revealing that around 30 years of age represents the initial watershed in the aging trajectory, while the period of 45–55 years constitutes a milestone transition in the aging process. Other molecular markers of aging, such as epigenetic clocks [ 30 , 31 ], proteomic and metabolomic profiles [ 5 , 7 ], further demonstrate consistent, punctuated dysregulation rather than gradual decline [ 28 ]. These transitions correspond to periods when the risk and presentation of major aging-related diseases demonstrably accelerate. High-throughput transcriptome sequencing enables systematic quantification of molecular dynamics, providing a robust framework to elucidate fundamental aging mechanisms and their links to diseases. Yet, research has mainly focused on conventional transcriptome profiling methods, such as bulk or single-cell mRNA-seq, and RNA immunoprecipitation sequencing (MeRIP-seq), often obscuring native RNA information [ 32 , 33 ]. Nanopore direct RNA sequencing (DRS), a state-of-the-art long-read method that directly quantifies full-length transcripts without cDNA conversion, is highly sought after, and enables systematic interrogation of diverse RNA features in aging across species and biological contexts [ 34 ]. Our study is poised to establish a cross-species RNA epitranscriptomic atlas spanning lung, skeletal muscle, heart and hippocampus tissues across lifespans. Leveraging Nanopore direct RNA sequencing, we decoded both universal and tissue-specific aging patterns of six RNA features, i.e., m 6 A, m 5 C, Ψ, inosine, APA, and AS, and further illustrating their complex relationships and potential crosstalk during aging. Our analysis revealed widespread non-linear, age-dependent shifts across these RNA features, exhibiting both conserved patterns and striking tissue specificity. Notably, shift events across RNA features occurred at mid-life (12 months in mouse, 46 years in human), and concurrently affecting the expression of aging-related, mitochondrial, and metabolic genes. Overall, this work will broaden our comprehension of the epitranscriptomic regulation in aging. Results Molecular Characteristics and Landscape of m 6 A Modifications in Aging Across Different Tissues N6-methyladenosine (m 6 A) is the most abundant internal modification in eukaryotic RNA. Direct RNA sequencing (DRS) enables quantitative, single-base-resolution mapping of N6-methyladenosine (m 6 A), facilitating the detection of isoform-specific methylation differences across the transcriptome. Following high-quality quality control, we obtained a total of 710,789 m 6 A sites detected by DRS from mouse tissues (heart, hippocampus, lung, and skeletal muscle) at 6, 12, and 21 months of age, and human lung tissue at 30, 46, 60 years of age, with two biological replicates for each group ( Supplementary Table S1 ). The raw data were processed through a standard base calling pipeline, and alignment was conducted using Minimap2. As the most abundant RNA modification, m 6 A levels varied significantly across tissues and ages. In 21-month-old mice, m 6 A sites were enriched in the 5’UTR region. In contrast, in 6-month-old mice, m 6 A sites were more prevalent near the stop codon across the four tissues studied ( Fig. 1A ). For 12-month-old mice, the hippocampus and lung tissues exhibited a higher proportion of m 6 A sites within the CDS compared to other ages ( Fig. 1A-B ). Overall, our results revealed a heterogeneous distribution of m 6 A across different RNA regions. Furthermore, while the m 6 A motifs exhibited distinct patterns across tissues, they showed remarkable consistency across different ages ( Fig. 1C ). Given that m 6 A modifications are regulated by specific m 6 A regulators, we integrated proteomic data from corresponding tissues and ages to analyze the expression differences of key m 6 A regulators [35], including ALKBH5, CAPRIN1, and METTL3. Significant expression changes in these regulators were observed, suggesting potential age-specific regulation of m 6 A modifications ( Fig. 1D ). Correlation and Functional Clustering of Gene Expression and m 6 A Modifications Across Aging Tissues We performed hierarchical clustering on gene expression and m 6 A profiles across ages and tissues, revealing six distinct modules for each ( Fig. 2A-B ). For gene expression, distinct functional associations characterized each cluster ( Fig. 2A ). Heart tissue clusters were linked to heart contraction. In skeletal muscle, clusters at 12 months showed enrichment for muscle cell differentiation genes, while those at 6 and 21 months were prominent for metabolic processes. Hippocampal clusters at 12 months associated with learning and memory, an association weakened at 6 and 21 months. Lung clusters at 12 months exhibited high expression of cell-substrate adhesion genes, whereas immune-related genes were more prominent at 6 and 21 months. When analyzing m 6 A levels, we found no consistent age-dependent increase or decrease; instead, m 6 A levels changed non-linearly with age ( Fig. 2B ), This finding highlights a sophisticated and age-specific regulatory mechanism of m 6 A modification. In lung, substantial immune cell infiltration corresponded with higher abundance of immune-related m 6 A sites at 6 and 21 months compared to 12 months. Additionally, our analysis revealed age- and tissue-specific differences in other m 6 A clusters, including Cluster 1 (C1), which is related to neural development, and Cluster 2 (C2), associated with cell adhesion. To further investigate the functional implications of these m 6 A modifications, we performed a correlation analysis between m 6 A abundance and corresponding gene expression across four distinct tissues and three age groups ( Fig. 2C ). Heatmaps and correlation analysis revealed a predominance of positive correlations ( Fig. 2C-D ), suggesting m 6 A primarily promotes gene expression in these contexts. This finding strongly suggests that m 6 A plays a significant role in maintaining RNA transcript stability, especially in both skeletal muscle and lung tissues. Additionally, we found most m 6 A sites showed an upward trend with aging ( Fig. 2D ). Collectively, m 6 A exerts a pervasive influence on gene expression throughout the aging process. Dynamic Profiling of pseudouridine (Ψ), m 5 C, and inosine During mouse Aging Aanalogous to m 6 A, we simultaneously assessed the atlas of other three most abundant RNA modifications across tissues and ages. Ψ exhibits pronounced tissue specificity ( Fig. 3A ). Hippocampus and lung harbour the greatest number of Ψ peaks, whereas heart and skeletal muscle contain markedly fewer. Across ages, the total number of Ψ peaks declines modestly, yet the chromosomal distribution shifts dramatically ( Fig. 3A ). In lung, Ψ is concentrated proximal to transcription start sites, followed by heart and then skeletal muscle; hippocampus is largely devoid of promoter-proximal Ψ. Conversely, the 3’UTR displays a sharp hippocampus-specific Ψ peak that is absent in other tissues. Within individual organs, heart shows the largest age-dependent redistribution, whereas skeletal muscle is comparatively stable ( Fig. 3B ). Differential Ψ analysis confirms the greatest age-associated variability in hippocampus ( Fig. 3C ). Functional enrichment of age-variable Ψ sites reveals tissue-specific pathways; nevertheless, metabolic processes are over-represented across all tissues ( Fig. 3D–K ), implicating Ψ in metabolic control of aging. Ψ sequence motifs remain conserved with age but diverge between tissues ( Fig. 3L–O ). Relative to the tens of thousands of m 6 A and Ψ peaks detected per sample, m 5 C and inosine are far less abundant, averaging ~4,000 and ~200 sites, respectively ( Supplementary Figs. S1A and S2A ). Despite their scarcity, both modifications display clear tissue specificity--again enriched in lung and hippocampus--and modest age specificity ( Supplementary Figs. S1A, S2A ). Positional analysis reveals that m 5 C density is invariant at stop codons but varies markedly in 5’UTRs, CDS and 3’UTRs ( Supplementary Fig. S1B ), whereas inosine is preferentially localized to the 3’UTR terminus across tissues ( Supplementary Fig. S2B ). Differential modification analysis shows pronounced age-related changes for m 5 C ( Supplementary Figs. S1C – K ) but limited functional signal for inosine, consistent with its low site abundance ( Supplementary Figs. S2C–K ). Motif analysis indicated that both modifications retain age-stable yet tissue-variable sequence preferences ( Supplementary Figs. S1L–O and S2L–O ). To extend the relevance of our findings to human biology, we generated parallel DRS datasets from human lung sampled across three age strata ( Supplementary Table S2), and collected public DRS data for human brain tissue ( Supplementary Table S3). Age-dependent modification landscapes were readily resolved ( Supplementary Figs. S3A ). Differential analysis revealed that m 6 A, m 5 C, pseudouridine and inosine sites whose abundance changes with age are overwhelmingly enriched for gene sets controlling cell-cell adhesion and metabolic reprogramming ( Supplementary Figs. S3B, C ), underscoring the conservation of RNA-epigenetic control over these hallmarks of mammalian aging. Together, our data establish tissue identity as the dominant determinant of RNA modification patterns, with age sculpting tissue-specific regulatory networks--particularly those governing metabolism--through selective remodelling of Ψ, m 5 C and, to a lesser extent, inosine. Dynamic landscape of alternative polyadenylation in aging across distinct tissues Polyadenylation is essential for eukaryotic mRNA maturation, and APA is an important layer for aging regulation. To explore poly(A) tail length (PAL) changes during aging, we estimated PAL for each transcript across tissue samples using Dorado (v0.6.2), yielding an average PAL of 91.20 nt (IQR: 82.57-100.40) in mouse and an average PAL of 85.63 nt (IQR: 84.60-86.56) in human ( Supplementary Table S4-1 ). Notably, significant PAL alterations were observed during aging across tissues ( P <2×10 16 ) ( Supplementary Fig. S4A, B ). To delve into the dynamic choice of different polyadenylation sites in a gene during aging, we obtained a total of 48,460 PACs in mice and 31,053 PACs in human detected in ≥ 2 samples. These mouse PACs were annotated into 16,077 genes, of which 60.67% genes contain at least two poly(A) sites ( Fig. 4A ). In human, PACs were mapped to 16,077 genes, with 55.55% containing ≥ two PACs ( Fig. 4B ). Most PACs identified by DRS resided in 3’UTR, specifically 74.68% for mouse and 86.42% for human ( Fig. 4C, D and Supplementary Fig. S4C, D ). Motif enrichment analyses of NUE elements for all PACs showed conserved classical hexamer motif A[U/A]UAAA across species ( Supplementary Fig. S4E ). Reads from PACs showed a bimodal peak corresponding to proximal and distal poly(A) sites across 3’UTR regions in all tissue types ( Supplementary Fig. S4F ). Analysis of 21 published APA regulators expression revealed eight dysregulated during aging across tissues ( Supplementary Fig. S4G ), and this included Cpsf family member Cpsf1 , and Cpsf2 . To explore alternative 3’UTR patten during aging across tissue, hierarchical clusting of 4,376 mice and 947 human 3’UTR PACs revealed seven mice and six human clusters with significant age-dependent 3’UTR length variation ( Fig. 4E, F ). In mouse, a pivotal PAS shift at 12 months drove 3’UTR shortening in Cluster 1 and 5 (C1, C5), and lengthening in Cluster 2 (C2) across all tissues, converging on kinase signaling (e.g., serine, threonine, MAPK and GTPase) and autophagy-proteostasis pathways (e.g., macroautophagy and protein catabolism) ( Fig. 4E and Supplementary Table S4-4 ). Notably, tissue-specific patterns emerged. Hippocampus showed divergent regulation in Cluster 3 (C3, lengthening) and 4 (C4, shortening), which linked to synaptic vesicle transport and pyruvate metabolism; heart-specific Cluster 6 (C6) exhibited preferential proximal PAS usage yielding the shortest 3’UTRs. Lung-specific shortening occurred in Cluster 7 (C7) genes. Both genes in C6 and C7 were enriched for mitochondrial organization. In human, C5 showed age-related 3’UTR shortening whereas C6 displayed lengthening, with other clusters exhibiting non-monotonic dynamics. GO enrichment implicated core lung aging pathways including autophagy regulation, oxidative stress response, immune modulation, and cell fate determination (e.g., apoptosis, adipocyte differentiation) ( Fig. 4F and Supplementary Table S4-5 ). Of 1,140 mouse mitochondrial genes (MitoCarta3.0 [36]), 395 showed 3’UTR length changes; 12 shortened and 21 lengthened significantly at least one defined age stage in four mouse tissues ( Fig. 4G, H, Supplementary Fig. S4H-K and Supplementary Table S4-2 ), with 9 exhibiting tissue heterogeneity (e.g., Ndufa2 with APA shifts at 12 months and sustained downregulation of transcript and protein with aging). Fifty-three mitochondrial genes showed 3’UTR length divergence in human lung tissue, exemplified by age-dependent 3’UTR shortening in TIMM22 ( Supplementary Table S4-2 ). Interestingly, four mitochondrial genes, namely MCL1 , BNIP3L, SOD1 and STX17 , occurred APA shift events in cross-species lung tissue. Among aging-related genes curated in the database aging atlas [37], we found that 9 and 13 genes with significant 3’UTR shortening and lengthening, respectively, during at least one defined age stage in four mouse tissues ( Fig. 4G, H and Supplementary Fig. S4H-K ). Among pan-tissue 3’UTR shortening events, Ighm wasobserved 3’UTR shortening at 12 and 21 months of age in lung tissue, while Tpml in skeletal muscle. Notably, we found tissue-specific 3’UTR shortening, including heart-specific Alkbh5 ( Supplementary Table S4-2 ), and hippocampus-specific Etv1 , Gpm6a , Kcnt1 and Syne1 . Tissue-specific APA patterns were observed: seven skeletal muscle-specific (e.g., Foxo1 and Jun ), one heart-specific ( Mapt ), 5 hippocampus-specific (e.g., Trp53bp1 ), and 18 lung-specific (e.g., Ercc4 ) ( Supplementary Table S4-2 ). In human, eight aging-related genes exhibited non-linear changes at 46 years of age, including IL7R , LMNA , RORA , TIMP2 , VEGFA , GAPDH , NFE2L2 and CXCL12 ( Fig. 4G, I and Supplementary Fig. S4L ). Notably, we found that GSK3B and SP1 were lung-specific APA plasticity during aging across species ( Supplementary Table S4-2, 3 ). Collectively, these findings highlighted the dual nature of APA events, namely evolutionarily conserved yet exhibiting marked tissue specificity. Transcriptomic and splicing-level aging clocks across human and mouse organs reveal tissue-specific disease trajectories Leveraging the single-base accuracy of direct RNA sequencing (DRS), we quantified transcript-level abundance and systematically examined age-dependent alternative splicing across organs. Transcript-level profiling revealed that lung aging in humans is distinguished by an immune-centric signature, whereas the aging brain is dominated by neuronal and Alzheimer’s-disease-associated isoforms--suggesting that transcriptomic remodelling, rather than modification status, may be the more immediate driver of age-related dysfunction ( Fig. 5A ). Extending the comparison to mice, we found that the aging lung and hippocampus recapitulate the human immune and neurodegenerative programmes, respectively ( Fig. 5B ). Skeletal-muscle aging was marked by simultaneous rewiring of immunity and metabolism, while the heart displayed a transcriptome signature enriched for pathways that presage cardiovascular disease--mirroring epidemiological observations linking aging to heightened cardiac risk ( Fig. 5B ). Skipped exon (SE) events were markedly rare at 6 months of age, reached maximal prevalence at 12 months, and declined again by 21 months, indicating a transient but prominent role for SE-mediated regulation during mid-life ( Supplementary Fig. S5A ). Tissue specificity was also evident: the heart exhibited consistently low SE frequencies, whereas the lung displayed the highest burden at every time point. Focusing on the lung, differential SE analysis between age groups revealed a striking enrichment of affected genes in metabolic pathways ( Supplementary Fig. S5B ), suggesting that alternative splicing remodels metabolic networks as a conserved feature of mammalian aging. This temporal pattern was not restricted to SE events; analogous trends were observed for alternative 5’ splice sites (A5), alternative 3’ splice sites (A3), retained introns (RI) and mutually exclusive exons (MX) across all tissues examined ( Supplementary Fig. S6–9 ). Collectively, our data establish a mid-life surge in alternative splicing--particularly in metabolic genes--as a pervasive, organ-specific signature of aging. Methods RNA preparation For RNA preparation, the Total RNA Kit I (R6834) was used following the manufacturer’s instructions. Briefly, cells or tissues were lysed with TRK Lysis Buffer supplemented with β-mercaptoethanol. For these tissue samples, the lysis buffer was added directly to the culture vessel or homogenized with liquid nitrogen. The lysate was clarified by centrifugation at 14,000xg for 2 min, and RNA was purified using RNA binding columns. The RNA was washed sequentially with RNA Wash Buffers I and II, eluted with nuclease-free water, and quantified using a NanoDrop One spectrophotometer and Qubit 3.0 Fluorometer. RNA integrity was assessed by electrophoresis. Construction of nanopore sequencing Library and sequencing Prepared RNA of each sample was used for a DRS library preparation using the Oxford Nanopore DRS protocol (SQK-RNA004, Oxford Nanopore Technologies). For each sample, we obtained over 20 million raw reads range from 19,006,308 to 25,330,731, with basecalling Phred scores predominantly >8 and a substantial portion of transcripts exhibiting >10x coverage. Reads exceeding 1.2 kb in length, were subjected to quality control and retained reads with mean Q score ≥ 10 (default threshold) for downstream analysis. These results demonstrate that the datasets are of high quality and provide better outputs and coverage compared to datasets generated by MinION or GridION dRNA-seq platforms. For reversed connector connection, 8μL prepared RNA, 3μL NEBNext Quick Ligation Reaction Buffer (NEB), 1μL RT Adapter (RTA) (SQK-RNA004) and 1.5 μL T4 DNA Ligase (NEB) were mixed together and incubated under 25°C for 10 min. Afterwards, 8μL 5x first-strand buffer (NEB), 2μL 10 mMdINTPs (NEB), 9 μL Nuclease-free water, 4μL 0.1 M DTT (Thermo Fisher) and 2μL SuperScript II Reverse Transcriptase (Thermo Fisher Scientific, 18080044) were added into the above 15μL reaction system and incubated under 50°C for 50 min then 70°C for 10min. Reverse-transcribed mRNA was purified with 1.8* Agencourt RNAClean XP beads and washed with 23μl Nuclease-free water, and subsequently sequencing adapters were added using 8μL T4 Ligation Reaction Buffer, 6μL RNA Ligation Adapter and 4μL T4 DNA Ligase. The mix was purified and washed again as above, and 40μL RRB (SQK-RNA004) were added with 35μLNuclease-freewater. This final reaction system was loaded into Nanopore FLO-PRO004RA sequencing chip and sequenced for 48-72 hrs using PromethION sequencer (Oxford Nanopore Technologies). Base calling and alignment The raw nanopore sequencing data, stored in pod5 format, underwent basecalling using Dorado v.0.9.0 with the parameter --estimate-poly-a to generate FASTQ files. Subsequent quality filtering categorized the reads into ‘pass’ (Q ≥ 10) and ‘fail’ (Q < 10) groups based on their mean quality scores, with the ‘pass’ reads designated as clean reads for downstream analyses. These clean reads were then aligned to the reference genome (hg38 and mm10) using minimap2 v.2.17-r941 with parameters -ax splice -uf -k14. Finally, the alignment ratio of clean reads to the reference genes was calculated using Samtools (v1.10) RNA modification ratio estimation To estimate the RNA modification ratio (m 5 C, m 6 A, Ψ and inosine), we used the lastest Dorado basecaller (model: [email protected] ), which identifies modified sites at single nucleotide resolution with ultra-high precision (94.5%). Dorado can calculate the modification probability per read to determine the methylation status of the site. It counts the number of methylated and unmethylated reads and calculates the score value (Fraction) for the site using the following formula: Fraction = n(M)/(n(UM) + n(M)). Where: n(M): Number of methylated reads at a given site (reads identified as carrying the methylation mark). n(UM): Number of unmethylated reads at the same site (reads identified as lacking the methylation mark). n(UM) + n(M): The total number of valid reads covering the site (excluding missing or low-quality reads). Fraction: The methylation level (fraction methylated) at that site. We used modkit (v0.4.1) for differential methylation loci (DML) analysis and differential transcript analysis. The “effect size” was defined the difference between Group1_pct_modified of group 1 and Group2_pct_modified of group 2. The threshold for selecting differential modified sites is |effect_size| >= 0.1 and P = 1 and score > 3. Poly(A) sites analysis We employed the R package Quantifypoly(A) (v0.3.0) [38] to identify, cluster and annotate polyadenylation sites (poly(A) sites, or PAS) for each tissue sample. Adjacent PAS within a 24 nt side window [39] were merged to form new entities termed poly(A) site clusters (PACs). Reads mapping to each potential PAC were counted per sample for further analysis. Each PAC includes three functional elements: the far upstream element (FUE; 150 nt upstream), the near upstream element (NUE; 35 nt upstream), and the core element (CE; spanning -10 nt to +15 nt relative to the PAC). We extracted the NUE sequences from each PAC, which recruits the CPSF complex to precisely regulate the selection of APA sites, and conducted motif enrichment analyses by using HOMER (v4.1.1). To evaluate global alternative polyadenylation (APA) differences, we calculated poly(A) site usage rates for genes with multiple poly(A) sites. The Kolmogorov-Smirnov (KS) test was applied to compare these rates between groups. To calculate the preference for using proximal or distal PAC during aging, we selected the two most highly expressed PACs for each gene, and then calculated the the relative expression difference (RED) as: RED = log2(expression_proximal_PAC / expression_distal_PAC). Genes with RED ≥ 0.3 or RED ≤ -0.3 were considered indicative of PAC usage bias. To estimate 3’UTR shortening and lengthening during aging across tissues, we selected genes with differentially expressed and the highest abundant PACs within the 3’UTR. Using the weighted average 3’UTR length of each gene, we subsequently performed unsupervised hierarchical clusting analysis. Differential alternative splicing event analysis To identify differential alternative splicing events, we employed the latest Dorado model in conjunction with the suppa2 (DiffSplice) software. Through calculating ΔPSI (dPSI) values and conducting statistical significance tests (p-values), we detected significant alternative splicing events between groups. The ΔPSI (Cond1_Cond2_dPSI) was calculated using the formula: ΔPSI = PSI 2 − PSI 1 . Here, Cond1 and Cond2 represent group 1 and group 2, respectively. PSI 2 is the average PSI value of samples in group 2, PSI 1 is the average PSI value of samples in group 1, and ΔPSI denotes the difference in PSI values between the two groups. We first screened for differential alternative splicing events by selecting those with |ΔPSI| > 0, referred to as Diff_AS_num in the results. Further, we identified significant differential alternative splicing events as those with |ΔPSI| > 0.1 and p-value < 0.05, designated as Significant_Diff_AS_num in the results. Subsequent functional enrichment analysis was performed on the genes associated with these significant differential alternative splicing events. Differentially expressed isoform analysis Consensus sequences were generated using Flair v1.5.0 (-t 20) based on alignment results. Bam files were converted to bed12 format using bam2Bed12.py, followed by adding gene or transcript information via identify_gene_isoform.py. Redundancy was removed using the collapse program.Non-redundant transcripts were obtained by aligning consensus sequences to the reference genome and using StringTie v2.1.4 (--conservative -L -R) in merge mode (default parameters) for transcript reconstruction and redundancy removal.Coding regions of novel transcripts were predicted using TransDecoder v5.5.0 (-m 50, --single_best_only), which identifies potential CDS based on ORF length and log-likelihood scores. Gene Ontology (GO) enrichment analysis Gene Ontology (GO) enrichment analyses were conducted using the DAVID database to identify overrepresented biological processes, cellular components, and molecular functions among genes exhibiting significant differences in both expression levels and RNA modification patterns. Proteomic data analysis across mouse and human aging We mined the C57BL/6J aging proteome from the GEO repository (GSE225576), which encompasses high-resolution mass-spectrometry profiles of brain, heart, lung and skeletal muscle sampled at different age. Raw spectral intensities were log 2 -transformed and quantile-normalised. We then extracted the expression values of 34 well-characterised m 6 A-associated RNA-binding proteins (RBPs)--including writers (METTL3, METTL14, WTAP), erasers (FTO, ALKBH5) and readers (YTHDF1/2/3, IGF2BP1/2/3)--curated from the literature. Tissue- and age-stratified expression matrices were analysed in R v4.2.1. Group-wise means and s.d. were calculated, and tissue-specific trajectories were visualised with ggplot2 v3.4.0. Statistical significance was assessed by one-way ANOVA followed by Tukey HSD post-hoc testing (α = 0.05). The resulting atlas reveals pronounced tissue specificity and age-dependent remodelling of the m 6 A proteome, providing an orthogonal resource to our direct-RNA-sequencing datasets. The human aging proteome was from the GSA repository HRA009355, HRA011245. Data processing and statistical analyses were performed identically to those used for the mouse datasets. Discussion Recent studies have begun to elucidate the role of RNA methylation in aging[15, 40]. For instance, Ricardo et al. demonstrated that reduced m 6 A-modified transcripts are associated with impaired synaptic protein synthesis and cognitive dysfunction, highlighting a potential mechanistic link between m 6 A and neuronal aging[41]. However, the reported effects of m 6 A-modifying enzymes during mammalian aging remain inconsistent. Notably, in long-lived endocrine mutant mice, elevated levels of key m 6 A regulator--including METTL3, METTL14, YTHDF1, and YTHDF2--have been observed, which enhance cap-independent translation and promote stress resistance and healthy aging[42]. We further assessed whether RNA modification enzymes are involved in the role of m 6 A during aging by analyzing proteomic data from mice at different ages[43], and found that ALKBH5 and METTL3 exhibited more pronounced differences than other m 6 A regulator in protein levels across age groups. Despite these insights, most previous studies have lacked precise, transcriptome-wide maps of RNA modifications. In this context, our use of the RNA004 direct RNA sequencing (DRS) technology provides a significant advance[44], enabling accurate, base-resolution detection of m 6 A and simultaneous profiling of multiple RNA modifications, including m 5 C, m 7 G, and Ψ (pseudouridine), in a single run. Compared to RNA002, RNA004 exhibits enhanced overall read accuracy at 93.5%, up from 92.1%. This improvement is mainly attributed to a reduction in mismatch and insertion errors. However, the deletion error rate remains elevated, indicating a need for further refinements[45]. Given the limited understanding of non-m 6 A RNA modifications in aging, our comprehensive multi-modification landscape offers a valuable resource for future mechanistic studies of RNA epigenetics in aged tissues. In this study, we employed dRNA-seq to conduct an in-depth analysis of the transcriptome and epitranscriptome in select senescent tissues of mice and humans. Our findings robustly corroborate prior studies. To our knowledge, this represents the first high-resolution, multi-modification atlas across aging-related samples, marking an important step toward deciphering the complexity of the epitranscriptome in the context of organismal aging. One limitation of our study lies in the relatively limited number of samples analyzed, particularly with respect to human-derived tissues. While the use of direct RNA sequencing (DRS) enabled high-resolution mapping of multiple RNA modifications, the current dataset may not fully capture the heterogeneity of RNA modification landscapes across individuals or populations. Expanding the sample size--especially by incorporating diverse human aging cohorts--will be essential to validate the generalizability of our findings and to better understand inter-individual variation in the epitranscriptomic regulation of aging. Declarations Ethics declarations The study received approval from the Ethics Committee of our institution (Approval Number: CY2023-032-01). Informed consent was obtained from all participants, and confidentiality was rigorously maintained. Conflict of interest statement The authors declare no competing interests. Author Contribution S.A. designed and performed most of the experiments, analyzed the data, and drafted the manuscript. H.Q. and S.Z. participated in experimental design, data collection, and manuscript revision. L.C., K.Y., Y.Q., J.M., W.Y. and W.L. assisted in sample preparation, sequencing experiments, and preliminary data analysis. K.M. provided technical support for RNA modification detection and bioinformatics analysis. L.Y. and Y.C. conceived and supervised the study, secured funding, revised the manuscript critically for important intellectual content, and gave final approval for publication. All authors read and approved the final manuscript. Acknowledgement This work was supported by grants from Noncommunicable Chronic Diseases-National Science and Technology Major Project (2023ZD0506102/2023ZD0506100), the National Natural Science Foundation of China (grant number 82203181); The Key Research and Development Program of Sichuan-Chongqing of the Chongqing Science and Technology Commission (grant number CSTB2022TIAD-CUX0001), the Guangxi Natural Science Foundation (2025GXNSFDA069011), and the Guangxi Medical University Training Program for Distinguished Young Scholars to Sanqi An. Data Availability All raw sequencing data of mice have been deposited to the GSA for Human in the Genome Sequence Archive (GSA) in National Genomics Data Center (https://ngdc.cncb.ac.cn/bioproject/browse/PRJCA036375). References Rudnicka E, Napierala P, Podfigurna A, Meczekalski B, Smolarczyk R, Grymowicz M. The World Health Organization (WHO) approach to healthy ageing. Maturitas. 2020;139:6-11. Pereira B, Correia FP, Alves IA, Costa M, Gameiro M, Martins AP, et al. Epigenetic Reprogramming as a Key to Reverse Ageing and Increase Longevity. Ageing Res Rev. 2024:102204. McCartney DL, Min JL, Richmond RC, Lu AT, Sobczyk MK, Davies G, et al. Genome-wide association studies identify 137 genetic loci for DNA methylation biomarkers of aging. Genome Biol. 2021;22(1):194. Moaddel R, Ubaida-Mohien C, Tanaka T, Lyashkov A, Basisty N, Schilling B, et al. Proteomics in aging research: A roadmap to clinical, translational research. Aging Cell. 2021;20(4):e13325. Argentieri MA, Xiao S, Bennett D, Winchester L, Nevado-Holgado AJ, Ghose U, et al. Proteomic aging clock predicts mortality and risk of common age-related diseases in diverse populations. Nat Med. 2024;30(9):2450-60. Kuo CL, Liu P, Drouard G, Vuoksimaa E, Kaprio J, Ollikainen M, et al. A proteomic signature of healthspan. Proc Natl Acad Sci U S A. 2025;122(23):e2414086122. Zhang S, Wang Z, Wang Y, Zhu Y, Zhou Q, Jian X, et al. A metabolomic profile of biological aging in 250,341 individuals from the UK Biobank. Nat Commun. 2024;15(1):8081. Panyard DJ, Yu B, Snyder MP. The metabolomics of human aging: Advances, challenges, and opportunities. Sci Adv. 2022;8(42):eadd6155. Shen L, Zhao H, Xi Y, Wang Z, Deng K, Gou W, et al. Mapping the gut microbial structural variations in healthy aging within the Chinese population. Cell Rep. 2024;43(11):114968. Kong F, Deng F, Li Y, Zhao J. Identification of gut microbiome signatures associated with longevity provides a promising modulation target for healthy aging. Gut Microbes. 2019;10(2):210-5. Pang S, Chen X, Lu Z, Meng L, Huang Y, Yu X, et al. Longevity of centenarians is reflected by the gut microbiome with youth-associated signatures. Nat Aging. 2023;3(4):436-49. Boccaletto P, Stefaniak F, Ray A, Cappannini A, Mukherjee S, Purta E, et al. MODOMICS: a database of RNA modification pathways. 2021 update. Nucleic Acids Res. 2022;50(D1):D231-D5. Wiener D, Schwartz S. The epitranscriptome beyond m(6)A. Nat Rev Genet. 2021;22(2):119-31. Roundtree IA, Evans ME, Pan T, He C. Dynamic RNA Modifications in Gene Expression Regulation. Cell. 2017;169(7):1187-200. Wu Z, Lu M, Liu D, Shi Y, Ren J, Wang S, et al. m(6)A epitranscriptomic regulation of tissue homeostasis during primate aging. Nat Aging. 2023;3(6):705-21. Castro-Hernandez R, Berulava T, Metelova M, Epple R, Pena Centeno T, Richter J, et al. Conserved reduction of m(6)A RNA modifications during aging and neurodegeneration is linked to changes in synaptic transcripts. Proc Natl Acad Sci U S A. 2023;120(9):e2204933120. Gruber AJ, Zavolan M. Alternative cleavage and polyadenylation in health and disease. Nat Rev Genet. 2019;20(10):599-614. Tang P, Yang Y, Li G, Huang L, Wen M, Ruan W, et al. Alternative polyadenylation by sequential activation of distal and proximal PolyA sites. Nat Struct Mol Biol. 2022;29(1):21-31. Lopez-Otin C, Blasco MA, Partridge L, Serrano M, Kroemer G. Hallmarks of aging: An expanding universe. Cell. 2023;186(2):243-78. Louros N, Ramakers M, Michiels E, Konstantoulea K, Morelli C, Garcia T, et al. Mapping the sequence specificity of heterotypic amyloid interactions enables the identification of aggregation modifiers. Nat Commun. 2022;13(1):1351. Mei H, Boom J, El Abdellaoui S, Abdelmohsen K, Munk R, Martindale JL, et al. Alternative Polyadenylation Utilization Results in Ribosome Assembly and mRNA Translation Deficiencies in a Model for Muscle Aging. J Gerontol A Biol Sci Med Sci. 2022;77(6):1130-40. Yoon OK, Brem RB. Noncanonical transcript forms in yeast and their regulation during environmental stress. RNA. 2010;16(6):1256-67. Shepard PJ, Choi EA, Lu J, Flanagan LA, Hertel KJ, Shi Y. Complex and dynamic landscape of RNA polyadenylation revealed by PAS-Seq. RNA. 2011;17(4):761-72. Li H, Wang Z, Ma T, Wei G, Ni T. Alternative splicing in aging and age-related diseases. Translational Medicine of Aging. 2017;1:32-40. Pal S, Gupta R, Davuluri RV. Alternative transcription and alternative splicing in cancer. Pharmacol Ther. 2012;136(3):283-94. Raj T, Li YI, Wong G, Humphrey J, Wang M, Ramdhani S, et al. Integrative transcriptome analyses of the aging brain implicate altered splicing in Alzheimer's disease susceptibility. Nat Genet. 2018;50(11):1584-92. Bowles KR, Silva MC, Whitney K, Bertucci T, Berlind JE, Lai JD, et al. ELAVL4, splicing, and glutamatergic dysfunction precede neuron loss in MAPT mutation cerebral organoids. Cell. 2021;184(17):4547-63 e17. Shen X, Wang C, Zhou X, Zhou W, Hornburg D, Wu S, et al. Nonlinear dynamics of multi-omics profiles during human aging. Nat Aging. 2024;4(11):1619-34. Ding Y, Zuo Y, Zhang B, Fan Y, Xu G, Cheng Z, et al. Comprehensive human proteome profiles across a 50-year lifespan reveal aging trajectories and signatures. Cell. 2025. Tarkhov AE, Lindstrom-Vautrin T, Zhang S, Ying K, Moqri M, Zhang B, et al. Nature of epigenetic aging from a single-cell perspective. Nat Aging. 2024;4(6):854-70. Ying K, Liu H, Tarkhov AE, Sadler MC, Lu AT, Moqri M, et al. Causality-enriched epigenetic age uncouples damage and adaptation. Nat Aging. 2024. Zhong ZD, Xie YY, Chen HX, Lan YL, Liu XH, Ji JY, et al. Systematic comparison of tools used for m(6)A mapping from nanopore direct RNA sequencing. Nat Commun. 2023;14(1):1906. Pratanwanich PN, Yao F, Chen Y, Koh CWQ, Wan YK, Hendra C, et al. Identification of differential RNA modifications from nanopore direct RNA sequencing with xPore. Nat Biotechnol. 2021;39(11):1394-402. Jain M, Abu-Shumays R, Olsen HE, Akeson M. Advances in nanopore direct RNA sequencing. Nat Methods. 2022;19(10):1160-4. An S, Huang W, Huang X, Cun Y, Cheng W, Sun X, et al. Integrative network analysis identifies cell-specific trans regulators of m6A. Nucleic Acids Res. 2020;48(4):1715-29. Rath S, Sharma R, Gupta R, Ast T, Chan C, Durham TJ, et al. MitoCarta3.0: an updated mitochondrial proteome now with sub-organelle localization and pathway annotations. Nucleic Acids Res. 2021;49(D1):D1541-D7. Aging Atlas C. Aging Atlas: a multi-omics database for aging biology. Nucleic Acids Res. 2021;49(D1):D825-D30. Ye C, Zhao D, Ye W, Wu X, Ji G, Li QQ, et al. QuantifyPoly(A): reshaping alternative polyadenylation landscapes of eukaryotes with weighted density peak clustering. Brief Bioinform. 2021;22(6). Tian B, Hu J, Zhang H, Lutz CS. A large-scale analysis of mRNA polyadenylation of human and mouse genes. Nucleic Acids Res. 2005;33(1):201-12. Fu YD, Jiang F, Zhang X, Pan YY, Xu R, Liang X, et al. Perturbation of METTL1-mediated tRNA N- methylguanosine modification induces senescence and aging. Nature Communications. 2024;15(1). Castro-Hernández R, Berulava T, Metelova M, Epple R, Centeno TP, Richter J, et al. Conserved reduction of m6A RNA modifications during aging and neurodegeneration is linked to changes in synaptic transcripts. P Natl Acad Sci USA. 2023;120(9). Sun J, Cheng BK, Su YK, Li M, Ma SY, Zhang Y, et al. The Potential Role of m6A RNA Methylation in the Aging Process and Aging-Associated Diseases. Frontiers in Genetics. 2022;13. Takasugi M, Nonaka Y, Takemura K, Yoshida Y, Stein F, Schwarz JJ, et al. An atlas of the aging mouse proteome reveals the features of age-related post-transcriptional dysregulation. Nature Communications. 2024;15(1). Liu-Wei W, van der Toorn W, Bohn P, Hölzer M, Smyth RP, von Kleist M. Sequencing accuracy and systematic errors of nanopore direct RNA sequencing. Bmc Genomics. 2024;25(1). Mears MC, Read QD, Bakre A. Comparison of direct RNA sequencing of using two different chemistries on the MinION platform. J Virol Methods. 2025;333. Additional Declarations No competing interests reported. Supplementary Files SupplementaryInformaton.zip Supplemental tables Table S1. m 6 A abundance across tissues and ages in mice. Table S2. DRS data form human lung. Table S3. DRS data form human brain. Table S4. Alternations of alternative cleavage and polyadenylation (APA) in aging across distinct tissues. Table S4-1 Statistics of poly(A) tail length in aging across mouse and human tissues. Table S4-2 3’UTR length changes across mouse four tissues. Table S4-3 3’UTR length changes across human lung tissues. Table S4-4 Gene Ontology (GO) biological processes (BP) significantly enriched by genes with aging-dependent 3’UTR length changes per cluster in mouse four aging tissues. Table S4-5 Gene Ontology (GO) biological processes (BP) significantly enriched by genes with aging-dependent 3’UTR length changes per cluster in human aging lung. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 07 Jan, 2026 Reviews received at journal 19 Dec, 2025 Reviews received at journal 11 Nov, 2025 Reviewers agreed at journal 30 Oct, 2025 Reviewers agreed at journal 29 Oct, 2025 Reviews received at journal 22 Oct, 2025 Reviewers agreed at journal 05 Oct, 2025 Reviewers invited by journal 02 Oct, 2025 Editor assigned by journal 04 Sep, 2025 Submission checks completed at journal 03 Sep, 2025 First submitted to journal 02 Sep, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7521909","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":528346479,"identity":"586d1447-8f52-4efb-bf88-a8167fda5abd","order_by":0,"name":"Sanqi An","email":"","orcid":"","institution":"Guangxi Medical University,","correspondingAuthor":false,"prefix":"","firstName":"Sanqi","middleName":"","lastName":"An","suffix":""},{"id":528346480,"identity":"b4add123-f57d-4179-aa7f-e9f0e3b02939","order_by":1,"name":"Haisong Qin","email":"","orcid":"","institution":"Guangxi Medical University,","correspondingAuthor":false,"prefix":"","firstName":"Haisong","middleName":"","lastName":"Qin","suffix":""},{"id":528346481,"identity":"b96dfe91-82d8-454b-9f74-62d3debab2a7","order_by":2,"name":"Shujie Zhang","email":"","orcid":"","institution":"Maternal and Child Health Hospital of Guangxi Zhuang Autonomous Region","correspondingAuthor":false,"prefix":"","firstName":"Shujie","middleName":"","lastName":"Zhang","suffix":""},{"id":528346482,"identity":"82c76f73-5074-4e12-af10-487bcb7f9fa5","order_by":3,"name":"Lizi Cheng","email":"","orcid":"","institution":"Guangxi Medical University,","correspondingAuthor":false,"prefix":"","firstName":"Lizi","middleName":"","lastName":"Cheng","suffix":""},{"id":528346483,"identity":"6a9ff501-9807-4c75-a30f-03464ad67eb5","order_by":4,"name":"Ke Ye","email":"","orcid":"","institution":"Sichuan University","correspondingAuthor":false,"prefix":"","firstName":"Ke","middleName":"","lastName":"Ye","suffix":""},{"id":528346484,"identity":"f2c46aaf-9b69-4a35-86cf-b5f1ed2c201c","order_by":5,"name":"Yawen Qi","email":"","orcid":"","institution":"Sichuan University","correspondingAuthor":false,"prefix":"","firstName":"Yawen","middleName":"","lastName":"Qi","suffix":""},{"id":528346485,"identity":"9d22622e-66e0-4f50-a2fd-fdc3e7e69105","order_by":6,"name":"Junhao Mu","email":"","orcid":"","institution":"the First Affiliated Hospital of Chongqing Medical University","correspondingAuthor":false,"prefix":"","firstName":"Junhao","middleName":"","lastName":"Mu","suffix":""},{"id":528346486,"identity":"f20818cd-f467-42e6-94e5-79e88fa1e88c","order_by":7,"name":"Weiyi Li","email":"","orcid":"","institution":"the First Affiliated Hospital of Chongqing Medical University","correspondingAuthor":false,"prefix":"","firstName":"Weiyi","middleName":"","lastName":"Li","suffix":""},{"id":528346487,"identity":"550b522d-bcfd-4b4b-93ef-781c045316fa","order_by":8,"name":"Wenli Yin","email":"","orcid":"","institution":"Maternal and Child Health Hospital of Guangxi Zhuang Autonomous Region","correspondingAuthor":false,"prefix":"","firstName":"Wenli","middleName":"","lastName":"Yin","suffix":""},{"id":528346489,"identity":"7fdbce34-5906-4d08-a732-80be49be0ed1","order_by":9,"name":"Kaikai Meng","email":"","orcid":"","institution":"Guangxi Subtropical Crops 21 Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Kaikai","middleName":"","lastName":"Meng","suffix":""},{"id":528346490,"identity":"d82e9eaa-8608-43e6-9653-5f3debf39dfc","order_by":10,"name":"Li Yang","email":"","orcid":"","institution":"the First Affiliated Hospital of Chongqing Medical University","correspondingAuthor":false,"prefix":"","firstName":"Li","middleName":"","lastName":"Yang","suffix":""},{"id":528346491,"identity":"eb289412-be72-4855-bdc6-2cb97e319e02","order_by":11,"name":"Yaxin Chen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABBklEQVRIiWNgGAWjYHACNiC24eHnbz7AwNgAEZIgQkuanOSMYwkkaTlsbHAgx4A4LQY3kp89+LiDObHhwJlvEj931MkbHGA+eJuHwS4Pp5Yzx8wNZ55hS2xs7t0m2XvmsOGGA2zJ1jwMycU4tRzvYZPmbeNJbGY4u+02Y9uBBIMDPGbSPAwHEhtwaTnMwyb9t00isY0h5xlQSx1QC/83/FpAtjC2GRjzMOSwAbUwg2xhw6tF8swxM8netgQ5CYlj5j972w4bzjzMZmw5xyAZpxY+YIhJ/Gz7z2N/vvmxwc+2Onm+480Pb7ypsMOpReEAhhAz2ME41AOBPC6zRsEoGAWjYBTAAQCmuFsJOuNCmAAAAABJRU5ErkJggg==","orcid":"","institution":"Sichuan University","correspondingAuthor":true,"prefix":"","firstName":"Yaxin","middleName":"","lastName":"Chen","suffix":""}],"badges":[],"createdAt":"2025-09-03 02:23:06","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7521909/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7521909/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":93602508,"identity":"f4eb4ce9-4f82-4bd8-83e5-e2b9979ca19e","added_by":"auto","created_at":"2025-10-15 14:50:34","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3178584,"visible":true,"origin":"","legend":"","description":"","filename":"DRSmanuscriptformatting.docx","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/ca51da0386440002c1f07142.docx"},{"id":93602503,"identity":"3fb8d289-71db-4de4-a863-afbc10e2bcd4","added_by":"auto","created_at":"2025-10-15 14:50:34","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":12403,"visible":true,"origin":"","legend":"","description":"","filename":"730d850c38c9463596bc4c52721ad0d7.json","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/4b5490f7dd0ce0e1b4756ab1.json"},{"id":93602499,"identity":"551af75a-cb41-401a-9209-ec14d5d84a9f","added_by":"auto","created_at":"2025-10-15 14:50:33","extension":"zip","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2186896,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryInformaton.zip","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/59bdb1ad0d1706e918cf5b2a.zip"},{"id":93602526,"identity":"d551cdac-3dee-4d03-9015-ec2e499c251d","added_by":"auto","created_at":"2025-10-15 14:50:36","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":129712,"visible":true,"origin":"","legend":"","description":"","filename":"730d850c38c9463596bc4c52721ad0d71enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/f3f60b90bc5fb4986094748f.xml"},{"id":93603554,"identity":"fcf7e953-05b5-4a7d-b9ea-40ea34846d01","added_by":"auto","created_at":"2025-10-15 14:58:35","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":229861,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/901853e4a361755809250307.jpeg"},{"id":93603547,"identity":"ce7b9c26-64c4-4321-be7f-ff5cb0aa7b58","added_by":"auto","created_at":"2025-10-15 14:58:34","extension":"jpeg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":189043,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage10.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/54099296801a7845e67e4932.jpeg"},{"id":93602533,"identity":"33f9dba8-5940-4e96-82bf-a26dc1783efd","added_by":"auto","created_at":"2025-10-15 14:50:41","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":120825,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage11.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/1a200c0dab8260d9ecbc7b36.jpeg"},{"id":93603926,"identity":"8336f0ec-0d45-4ae4-915a-c732d5b57002","added_by":"auto","created_at":"2025-10-15 15:06:34","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":120924,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage12.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/824e1106ab28f0ee5a03378f.jpeg"},{"id":93602529,"identity":"2f80883d-be27-4149-8519-7f12cbd1b26e","added_by":"auto","created_at":"2025-10-15 14:50:37","extension":"jpeg","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":119896,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage13.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/160f66db5d5a83903da3cfc3.jpeg"},{"id":93602516,"identity":"0360b1df-2f41-4d18-907f-c1bbb89e3357","added_by":"auto","created_at":"2025-10-15 14:50:35","extension":"jpeg","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":115368,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage14.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/4cfb190c019233632ddbb44f.jpeg"},{"id":93602538,"identity":"456b2534-0af1-48ff-bf42-07cc54e1874c","added_by":"auto","created_at":"2025-10-15 14:50:45","extension":"jpeg","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":285732,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/56a4e4eeac224835aecced08.jpeg"},{"id":93603927,"identity":"6b201627-df01-4d4e-bae3-703619b353e3","added_by":"auto","created_at":"2025-10-15 15:06:35","extension":"jpeg","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":221476,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/2b6d3015ef31ba9a0d999889.jpeg"},{"id":93602504,"identity":"4073f3f9-2541-4df9-a2ca-dab1a188c061","added_by":"auto","created_at":"2025-10-15 14:50:34","extension":"jpeg","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":323663,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/355be576aff5fe535c5b36b4.jpeg"},{"id":93603548,"identity":"417bcdce-cea6-45b2-a1ab-438e9438ad37","added_by":"auto","created_at":"2025-10-15 14:58:34","extension":"jpeg","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":218908,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/de5b971306998c7355bdf46e.jpeg"},{"id":93602519,"identity":"f2e1c756-5013-4e27-b554-4ef8c8d23709","added_by":"auto","created_at":"2025-10-15 14:50:35","extension":"jpeg","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":225613,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/af9155dc26d217935afb74b0.jpeg"},{"id":93603551,"identity":"ea38fd60-f7e5-4087-b406-ef42a0978c70","added_by":"auto","created_at":"2025-10-15 14:58:35","extension":"jpeg","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":240133,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage7.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/98090d32edebbdc6c1f2b3b7.jpeg"},{"id":93602524,"identity":"2b9516b4-3937-4df5-b13a-a6bc085fab9a","added_by":"auto","created_at":"2025-10-15 14:50:36","extension":"jpeg","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":171384,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage8.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/2f872d796809f310ebe319d6.jpeg"},{"id":93603553,"identity":"9fce3a8b-b170-4e38-a990-3fb4c508dbb0","added_by":"auto","created_at":"2025-10-15 14:58:35","extension":"jpeg","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":354477,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage9.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/8f66078e0f416ab0e5438d20.jpeg"},{"id":93602501,"identity":"d012e5d7-3a23-4e5f-893b-b49b271247fa","added_by":"auto","created_at":"2025-10-15 14:50:34","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":124481,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/d1205e85d9ce8233e2824b4b.png"},{"id":93602510,"identity":"4c8877c6-e775-46e3-a515-dd32fd3fe61d","added_by":"auto","created_at":"2025-10-15 14:50:35","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":87176,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage10.png","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/8aaab162a87507b2e2f3eb88.png"},{"id":93602539,"identity":"546ba02a-0f6b-4c24-8f34-e8d22ebc8c3f","added_by":"auto","created_at":"2025-10-15 14:50:45","extension":"png","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":44148,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage11.png","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/317528e67b92e70fce55a769.png"},{"id":93602530,"identity":"49f01544-618b-4bf3-a3a0-f6c92f6a71ee","added_by":"auto","created_at":"2025-10-15 14:50:37","extension":"png","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":43822,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage12.png","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/f7b3dc8a87ec43244e4c3da4.png"},{"id":93602522,"identity":"aabdd1cf-56bc-4e20-866b-c83529370f28","added_by":"auto","created_at":"2025-10-15 14:50:36","extension":"png","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":37014,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage13.png","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/3b6ae7b8d2838b3539ca9361.png"},{"id":93602535,"identity":"0391b4e5-72fe-4ff9-bff6-a54ce10249e7","added_by":"auto","created_at":"2025-10-15 14:50:41","extension":"png","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":30965,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage14.png","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/0793d6d68d6258c395bd69c3.png"},{"id":93602525,"identity":"4353d3db-8b23-486f-9b6a-5ef34882874c","added_by":"auto","created_at":"2025-10-15 14:50:36","extension":"png","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":222013,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/e05556b0b0b90bc965500f5e.png"},{"id":93602507,"identity":"0a886a1f-32a7-4343-b44c-a09557fa4e2a","added_by":"auto","created_at":"2025-10-15 14:50:34","extension":"png","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":130972,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/586d3b5d3e6e99c915b58d25.png"},{"id":93602528,"identity":"e69727b3-9ab9-4783-9967-38ee37853a2f","added_by":"auto","created_at":"2025-10-15 14:50:36","extension":"png","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":220304,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/e30450f3d2ce1a28d7849d08.png"},{"id":93602541,"identity":"696ef8ec-8d01-434c-872e-2f7f63e9abf7","added_by":"auto","created_at":"2025-10-15 14:50:46","extension":"png","order_by":27,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":123053,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/f8d74df7d5d334873c93e0a0.png"},{"id":93602532,"identity":"fabd1efc-b380-4f6d-b239-f237f42a4a3b","added_by":"auto","created_at":"2025-10-15 14:50:39","extension":"png","order_by":28,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":151378,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/5438ed163e1c70d4bf47d179.png"},{"id":93602513,"identity":"766cf358-f14c-46f7-8f69-241c33d1a854","added_by":"auto","created_at":"2025-10-15 14:50:35","extension":"png","order_by":29,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":149130,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/f54a43258a94bf8dd15e98bd.png"},{"id":93602523,"identity":"1abe6ed6-fc11-4a59-8ae5-f9de2566cf16","added_by":"auto","created_at":"2025-10-15 14:50:36","extension":"png","order_by":30,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":87397,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/8eae0f421e212a77949f1227.png"},{"id":93602540,"identity":"2c004c00-3978-4175-893d-72b4e586a9b2","added_by":"auto","created_at":"2025-10-15 14:50:45","extension":"png","order_by":31,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":217766,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/6e92c37ab49318e601acfa0d.png"},{"id":93602514,"identity":"7f90e896-d985-422d-a64a-0b85990a012c","added_by":"auto","created_at":"2025-10-15 14:50:35","extension":"xml","order_by":32,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":127590,"visible":true,"origin":"","legend":"","description":"","filename":"730d850c38c9463596bc4c52721ad0d71structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/db706d4f7f947c201d610fea.xml"},{"id":93602521,"identity":"12641637-ef84-4e7f-b81a-3260ee319fd5","added_by":"auto","created_at":"2025-10-15 14:50:36","extension":"html","order_by":33,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":138341,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/8d9195c57483cbe91f6a4605.html"},{"id":93603550,"identity":"04fbb001-5dad-4939-8cda-9707aa90858d","added_by":"auto","created_at":"2025-10-15 14:58:34","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":494866,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003em\u003c/strong\u003e\u003csup\u003e\u003cstrong\u003e6\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003eA and proteomics profile of aged mice. \u003c/strong\u003e(A) Distribution of m\u003csup\u003e6\u003c/sup\u003eA density along metagene coordinates in samples of different tissues and ages of mice. (B) Percentage distribution of m\u003csup\u003e6\u003c/sup\u003eA sites in UTR and CDS regions in samples of different tissues and ages of mice. (C) Sequence motif analyses of m\u003csup\u003e6\u003c/sup\u003eA sites across age groups. (D) Gene expression levels of m\u003csup\u003e6\u003c/sup\u003eA regulatory factors in different tissues and mouse age samples.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/ecdcbacbfbe9b5fd1964db69.png"},{"id":93602500,"identity":"08542d41-1769-40fe-98fa-25e3db14fc0b","added_by":"auto","created_at":"2025-10-15 14:50:33","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":620540,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe relationship of m\u003c/strong\u003e\u003csup\u003e\u003cstrong\u003e6\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003eA and gene expression\u003c/strong\u003e. (A) Trend graph, heat map and functional annotation analysis of gene expression patterns in different tissues and mouse age samples. (B) Trend graph, heat map and functional annotation analysis of m\u003csup\u003e6\u003c/sup\u003eA modification levels in different tissues and mouse age samples. (C) Heat maps display paired m\u003csup\u003e6\u003c/sup\u003eA modification levels (left panel) and corresponding gene expression (right panel) for individual samples from skeletal muscle, heart, hippocampus and lung. Samples are arranged identically along the horizontal axis, enabling direct, gene-by-gene comparison of methylation abundance and transcript expression within each tissue. (D) Differential m\u003csup\u003e6\u003c/sup\u003eA landscape between young and aged mice.The plot showing the number of statistical significance of m\u003csup\u003e6\u003c/sup\u003eA sites that differ most prominently between 6-month-old (young) and 21-month-old (aged) mice.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/01b5ffbeda5943e82967b850.png"},{"id":93602517,"identity":"bf40a6c3-0281-4c5c-96b5-a36329d2a371","added_by":"auto","created_at":"2025-10-15 14:50:35","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":478718,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePseudouridine atlas across murine aging. \u003c/strong\u003e(A) Number of Ψ sites detected by direct RNA sequencing (DRS) in each tissue at 6, 12 and 21 months. (B) Metagene distribution of Ψ along RNA (5’UTR, CDS, 3’UTR) per sample. (C) Count of differential Ψ sites between age groups within each tissue. (D-K) Functional enrichment clusters for age-variable. (L-O) Sequence motif analyses of Ψ sites across tissues, revealing age-conserved yet tissue-specific preferences.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/07e751d2da6cdea6cd4dbb77.png"},{"id":93603928,"identity":"405d913c-f8aa-49a8-a134-5168ad67d88d","added_by":"auto","created_at":"2025-10-15 15:06:35","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":696250,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePolyadenylation sites and its dynamics usage during aging across tissue. (A, B)\u003c/strong\u003e Bar graphs showing numbers of genes categorized by the number of PAC identified within the gene by dRNA-seq in mouse (A) and human (B) tissues. \u003cstrong\u003e(C, D)\u003c/strong\u003e Gene distribution proportion of PACs in mouse four tissues (C) and human lung tissue (D). \u003cstrong\u003e(E, F)\u003c/strong\u003eUnsupervised hierarchical clusting analysis of differentially expressed highest-abundant 3’UTR PACs in aging tissues. Middle heatmap showing distinct clusters across aging mouse tissues (E) and human lung tissue (F). Heatmap right showing Gene Ontology (GO) biological processes (BP) analysis of genes whose 3’UTR shortening or lengthening during aging per cluster. Left boxplot showing normalized weighted 3’UTR length of genes in each cluster. (\u003cstrong\u003eG-J\u003c/strong\u003e) The number of commonly-detected and tissue-specific genes with 3’UTR shortening or lengthening in mouse (G, H) and human (I, J). PAC, poly(A) site cluster that merging adjacent polyadenylation site within a 24 nt side window. Hpc, hippocampus; SkM, skeletal muscle.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/1a62792c98db180d9e2a251a.png"},{"id":93603556,"identity":"e725483f-7c89-462a-b929-f3e20dd9d567","added_by":"auto","created_at":"2025-10-15 14:58:36","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":477568,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTranscriptomic signatures of aging in human and mouse tissues. (A)\u003c/strong\u003e Differentially expressed transcripts and pathway enrichment between young and aged human hippocampus and lung. (B) Corresponding differential transcriptomes and pathway enrichments in young versus aged mouse brain, lung, heart and skeletal muscle.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/99bc8d782d0c168b931b61cc.png"},{"id":93681930,"identity":"dc31ecdf-96cc-4310-abde-3d3b84702cd7","added_by":"auto","created_at":"2025-10-16 12:29:19","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4175671,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/b1659dfe-17d9-44f2-bb0b-bf1e9ad9f8ee.pdf"},{"id":93602534,"identity":"710852c2-fe39-4589-b0d6-003ac925cd84","added_by":"auto","created_at":"2025-10-15 14:50:41","extension":"zip","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":2186896,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplemental tables\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable S1. m\u003c/strong\u003e\u003csup\u003e\u003cstrong\u003e6\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003eA abundance across tissues and ages in mice.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable S2. DRS data form human lung.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable S3. DRS data form human brain.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable S4. Alternations of alternative cleavage and polyadenylation (APA) in aging across distinct tissues.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable S4-1 Statistics of poly(A) tail length in aging across mouse and human tissues.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable S4-2 3’UTR length changes across mouse four tissues.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable S4-3 3’UTR length changes across human lung tissues.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable S4-4 Gene Ontology (GO) biological processes (BP) significantly enriched by genes with aging-dependent 3’UTR length changes per cluster in mouse four aging tissues.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable S4-5 Gene Ontology (GO) biological processes (BP) significantly enriched by genes with aging-dependent 3’UTR length changes per cluster in human aging lung.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"SupplementaryInformaton.zip","url":"https://assets-eu.researchsquare.com/files/rs-7521909/v1/cfa51e6734ba7d2fe67690b6.zip"}],"financialInterests":"No competing interests reported.","formattedTitle":"Direct RNA sequencing reveals multi-layered regulation of the aging transcriptome","fulltext":[{"header":"Introduction","content":"\u003cp\u003eAging is a complex and multifactorial process of physiological changes strongly associated with nearly all diseases, including nearly all chronic diseases and cancers[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Molecular alterations during aging have been documented at genomic (e.g., DNA methylation [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]), proteomic [\u003cspan additionalcitationids=\"CR5\" citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e], metabolomic [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e], and even in microbial community levels [\u003cspan additionalcitationids=\"CR10\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Yet, despite the well-known changes in RNA expression levels, our understanding of the aging patterns at the full spectrum of RNA characteristics (e.g., RNA modifications, RNA splicing, RNA editing) within individual multi-organs remains largely uncharted territories.\u003c/p\u003e\u003cp\u003eOver 170 distinct types of chemical modifications on RNA molecules have been reported [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. Among these, \u003cem\u003eN\u003c/em\u003e6-methyladenosine (m\u003csup\u003e6\u003c/sup\u003eA), 5-methylcytidine (m\u003csup\u003e5\u003c/sup\u003eC), pseudouridine (Ψ), and inosine are the most prevalent and extensively studied. These modifications precisely regulate gene expression in context-dependent manner, contributing to both organ-specific aging and systemic aging, and influencing the pathogenesis of tissue-specific diseases [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. For instance, it has been revealed that METTL3-mediated m\u003csup\u003e6\u003c/sup\u003eA methylation maintains skeletal muscle and contributes to muscle degeneration [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Additionally, the m\u003csup\u003e6\u003c/sup\u003eA atlas of brain tissue across mice and human reveals conserved synaptic transcript hypomethylation, thereby accelerating cognitive decline in aging and Alzheimer\u0026rsquo;s disease [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Despite rapid progress in this field, our understanding of the complexity of RNA modifications and their potential crosstalk during aging remains fragmented.\u003c/p\u003e\u003cp\u003eBeyond RNA modifications, alternative cleavage and polyadenylation (APA) is profoundly implicated in aging. APA governs the selection of polyadenylation sites (PASs) in precursor mRNAs (pre-mRNAs) and influences transcript stability, localization, and translation efficiency [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Of note, pervasive shortening of 3\u0026rsquo;UTRs through preferential proximal poly(A) site usage is a hallmark of aging across diverse tissues [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. This phenomenon remodels post-transcriptional regulatory networks by eliminating miRNA and RBP binding sites, driving age-related functional decline(21). APA-mediated shortening of the 3\u0026rsquo;UTR, culminating in ribosomal dysfunction and progressive muscle atrophy [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. Various methods for mapping PASs genome-wide exist (e.g., 3\u0026rsquo;RNA-seq [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e], PAS-seq [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]). Analogous to APA, ASgenerates RNA isoform diversity in \u0026gt;\u0026thinsp;90% of human protein-coding genes, exponentially increasing proteome variation and critically regulating molecular pathways [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. AS dysregulation is implicated in aging and age-related diseases [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. In Frontotemporal Dementia, age-dependent mis-splicing of MAPT (tau isoforms) shifts the 3R/4R tau isoform balance toward aggregation-prone 4R-tau, driving neurofibrillary tangle formation in hippocampus [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. Therefore, age-associated AS dysregulation in critical genes causes protein dysfunction and speeds up disease pathogenesis, especially in vulnerable tissues.\u003c/p\u003e\u003cp\u003eAging is intrinsically nonlinear and its trajectories differ across organs [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. Ding et al [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e] developed tissue-specific proteomic aging clocks spanning 13 human tissues, revealing that around 30 years of age represents the initial watershed in the aging trajectory, while the period of 45\u0026ndash;55 years constitutes a milestone transition in the aging process. Other molecular markers of aging, such as epigenetic clocks [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e], proteomic and metabolomic profiles [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e], further demonstrate consistent, punctuated dysregulation rather than gradual decline [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. These transitions correspond to periods when the risk and presentation of major aging-related diseases demonstrably accelerate. High-throughput transcriptome sequencing enables systematic quantification of molecular dynamics, providing a robust framework to elucidate fundamental aging mechanisms and their links to diseases. Yet, research has mainly focused on conventional transcriptome profiling methods, such as bulk or single-cell mRNA-seq, and RNA immunoprecipitation sequencing (MeRIP-seq), often obscuring native RNA information [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. Nanopore direct RNA sequencing (DRS), a state-of-the-art long-read method that directly quantifies full-length transcripts without cDNA conversion, is highly sought after, and enables systematic interrogation of diverse RNA features in aging across species and biological contexts [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eOur study is poised to establish a cross-species RNA epitranscriptomic atlas spanning lung, skeletal muscle, heart and hippocampus tissues across lifespans. Leveraging Nanopore direct RNA sequencing, we decoded both universal and tissue-specific aging patterns of six RNA features, i.e., m\u003csup\u003e6\u003c/sup\u003eA, m\u003csup\u003e5\u003c/sup\u003eC, Ψ, inosine, APA, and AS, and further illustrating their complex relationships and potential crosstalk during aging. Our analysis revealed widespread non-linear, age-dependent shifts across these RNA features, exhibiting both conserved patterns and striking tissue specificity. Notably, shift events across RNA features occurred at mid-life (12 months in mouse, 46 years in human), and concurrently affecting the expression of aging-related, mitochondrial, and metabolic genes. Overall, this work will broaden our comprehension of the epitranscriptomic regulation in aging.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eMolecular Characteristics and Landscape of m\u003csup\u003e6\u003c/sup\u003eA Modifications in Aging Across Different Tissues\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eN6-methyladenosine (m\u003csup\u003e6\u003c/sup\u003eA) is the most abundant internal modification in eukaryotic RNA. Direct RNA sequencing (DRS) enables quantitative, single-base-resolution mapping of N6-methyladenosine (m\u003csup\u003e6\u003c/sup\u003eA), facilitating the detection of isoform-specific methylation differences across the transcriptome. Following high-quality quality control, we obtained a total of 710,789 m\u003csup\u003e6\u003c/sup\u003eA sites detected by DRS from mouse tissues (heart, hippocampus, lung, and skeletal muscle) at 6, 12, and 21 months of age, and human lung tissue at 30, 46, 60 years of age, with two biological replicates for each group (\u003cstrong\u003eSupplementary Table S1\u003c/strong\u003e). The raw data were processed through a standard base calling pipeline, and alignment was conducted using Minimap2. As the most abundant RNA modification, m\u003csup\u003e6\u003c/sup\u003eA levels varied significantly across tissues and ages. In 21-month-old mice, m\u003csup\u003e6\u003c/sup\u003eA sites were enriched in the 5’UTR region. In contrast, in 6-month-old mice, m\u003csup\u003e6\u003c/sup\u003eA sites were more prevalent near the stop codon across the four tissues studied (\u003cstrong\u003eFig.\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;1A\u003c/strong\u003e). For 12-month-old mice, the hippocampus and lung tissues exhibited a higher proportion of m\u003csup\u003e6\u003c/sup\u003eA sites within the CDS compared to other ages (\u003cstrong\u003eFig. 1A-B\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003eOverall, our results revealed a heterogeneous distribution of m\u003csup\u003e6\u003c/sup\u003eA across different RNA regions. Furthermore, while the m\u003csup\u003e6\u003c/sup\u003eA motifs exhibited distinct patterns across tissues, they showed remarkable consistency across different ages (\u003cstrong\u003eFig. 1C\u003c/strong\u003e). Given that m\u003csup\u003e6\u003c/sup\u003eA modifications are regulated by specific m\u003csup\u003e6\u003c/sup\u003eA regulators, we integrated proteomic data from corresponding tissues and ages to analyze the expression differences of key m\u003csup\u003e6\u003c/sup\u003eA regulators [35], including ALKBH5, CAPRIN1, and METTL3. Significant expression changes in these regulators were observed, suggesting potential age-specific regulation of m\u003csup\u003e6\u003c/sup\u003eA modifications (\u003cstrong\u003eFig. 1D\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCorrelation and Functional Clustering of Gene Expression and m\u003csup\u003e6\u003c/sup\u003eA Modifications Across Aging Tissues\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe performed hierarchical clustering on gene expression and m\u003csup\u003e6\u003c/sup\u003eA profiles across ages and tissues, revealing six distinct modules for each (\u003cstrong\u003eFig. 2A-B\u003c/strong\u003e). For gene expression, distinct functional associations characterized each cluster (\u003cstrong\u003eFig. 2A\u003c/strong\u003e). Heart tissue clusters were linked to heart contraction. In skeletal muscle, clusters at 12 months showed enrichment for muscle cell differentiation genes, while those at 6 and 21 months were prominent for metabolic processes. Hippocampal clusters at 12 months associated with learning and memory, an association weakened at 6 and 21 months. Lung clusters at 12 months exhibited high expression of cell-substrate adhesion genes, whereas immune-related genes were more prominent at 6 and 21 months.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWhen analyzing m\u003csup\u003e6\u003c/sup\u003eA levels, we found no consistent age-dependent increase or decrease; instead, m\u003csup\u003e6\u003c/sup\u003eA levels changed non-linearly with age (\u003cstrong\u003eFig. 2B\u003c/strong\u003e), This finding highlights a sophisticated and age-specific regulatory mechanism of m\u003csup\u003e6\u003c/sup\u003eA modification. In lung, substantial immune cell infiltration corresponded with higher abundance of immune-related m\u003csup\u003e6\u003c/sup\u003eA sites at 6 and 21 months compared to 12 months. Additionally, our analysis revealed age- and tissue-specific differences in other m\u003csup\u003e6\u003c/sup\u003eA clusters, including Cluster 1 (C1), which is related to neural development, and Cluster 2 (C2), associated with cell adhesion.\u003c/p\u003e\n\u003cp\u003eTo further investigate the functional implications of these m\u003csup\u003e6\u003c/sup\u003eA modifications, we performed a correlation analysis between m\u003csup\u003e6\u003c/sup\u003eA abundance and corresponding gene expression across four distinct tissues and three age groups (\u003cstrong\u003eFig. 2C\u003c/strong\u003e). Heatmaps and correlation analysis revealed a predominance of positive correlations (\u003cstrong\u003eFig.\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;2C-D\u003c/strong\u003e), suggesting m\u003csup\u003e6\u003c/sup\u003eA primarily promotes gene expression in these contexts.\u0026nbsp;This finding strongly suggests that m\u003csup\u003e6\u003c/sup\u003eA plays a significant role in maintaining RNA transcript stability, especially in both skeletal muscle and lung tissues. Additionally, we found most m\u003csup\u003e6\u003c/sup\u003eA sites showed an upward trend with aging (\u003cstrong\u003eFig. 2D\u003c/strong\u003e).\u0026nbsp;Collectively, m\u003csup\u003e6\u003c/sup\u003eA exerts a pervasive influence on gene expression throughout the aging process.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDynamic Profiling of pseudouridine (Ψ), m\u003csup\u003e5\u003c/sup\u003eC, and inosine During mouse Aging\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAanalogous to m\u003csup\u003e6\u003c/sup\u003eA, we simultaneously assessed the atlas of other three most abundant RNA modifications across tissues and ages. Ψ exhibits pronounced tissue specificity (\u003cstrong\u003eFig. 3A\u003c/strong\u003e). Hippocampus and lung harbour the greatest number of Ψ peaks, whereas heart and skeletal muscle contain markedly fewer. Across ages, the total number of Ψ peaks declines modestly, yet the chromosomal distribution shifts dramatically (\u003cstrong\u003eFig. 3A\u003c/strong\u003e). In lung, Ψ is concentrated proximal to transcription start sites, followed by heart and then skeletal muscle; hippocampus is largely devoid of promoter-proximal Ψ. Conversely, the 3’UTR displays a sharp hippocampus-specific Ψ peak that is absent in other tissues. Within individual organs, heart shows the largest age-dependent redistribution, whereas skeletal muscle is comparatively stable (\u003cstrong\u003eFig. 3B\u003c/strong\u003e). Differential Ψ analysis confirms the greatest age-associated variability in hippocampus (\u003cstrong\u003eFig. 3C\u003c/strong\u003e). Functional enrichment of age-variable Ψ sites reveals tissue-specific pathways; nevertheless, metabolic processes are over-represented across all tissues (\u003cstrong\u003eFig. 3D–K\u003c/strong\u003e), implicating Ψ in metabolic control of aging. Ψ sequence motifs remain conserved with age but diverge between tissues (\u003cstrong\u003eFig. 3L–O\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003eRelative to the tens of thousands of m\u003csup\u003e6\u003c/sup\u003eA and Ψ peaks detected per sample, m\u003csup\u003e5\u003c/sup\u003eC and inosine are far less abundant, averaging ~4,000 and ~200 sites, respectively (\u003cstrong\u003eSupplementary Figs. S1A and S2A\u003c/strong\u003e). Despite their scarcity, both modifications display clear tissue specificity--again enriched in lung and hippocampus--and modest age specificity (\u003cstrong\u003eSupplementary Figs. S1A, S2A\u003c/strong\u003e). Positional analysis reveals that m\u003csup\u003e5\u003c/sup\u003eC density is invariant at stop codons but varies markedly in 5’UTRs, CDS and 3’UTRs (\u003cstrong\u003eSupplementary Fig. S1B\u003c/strong\u003e), whereas inosine is preferentially localized to the 3’UTR terminus across tissues (\u003cstrong\u003eSupplementary Fig. S2B\u003c/strong\u003e). Differential modification analysis shows pronounced age-related changes for m\u003csup\u003e5\u003c/sup\u003eC (\u003cstrong\u003eSupplementary Figs. S1C – K\u003c/strong\u003e) but limited functional signal for inosine, consistent with its low site abundance (\u003cstrong\u003eSupplementary Figs. S2C–K\u003c/strong\u003e). Motif analysis indicated that both modifications retain age-stable yet tissue-variable sequence preferences (\u003cstrong\u003eSupplementary Figs. S1L–O and S2L–O\u003c/strong\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo extend the relevance of our findings to human biology, we generated parallel DRS datasets from human lung sampled across three age strata (\u003cstrong\u003eSupplementary Table\u0026nbsp;\u003c/strong\u003eS2), and collected public DRS data for human brain tissue (\u003cstrong\u003eSupplementary Table\u0026nbsp;\u003c/strong\u003eS3). Age-dependent modification landscapes were readily resolved (\u003cstrong\u003eSupplementary\u003c/strong\u003e \u003cstrong\u003eFigs. S3A\u003c/strong\u003e). Differential analysis revealed that m\u003csup\u003e6\u003c/sup\u003eA, m\u003csup\u003e5\u003c/sup\u003eC, pseudouridine and inosine sites whose abundance changes with age are overwhelmingly enriched for gene sets controlling cell-cell adhesion and metabolic reprogramming (\u003cstrong\u003eSupplementary Figs. S3B, C\u003c/strong\u003e), underscoring the conservation of RNA-epigenetic control over these hallmarks of mammalian aging. Together, our data establish tissue identity as the dominant determinant of RNA modification patterns, with age sculpting tissue-specific regulatory networks--particularly those governing metabolism--through selective remodelling of Ψ, m\u003csup\u003e5\u003c/sup\u003eC and, to a lesser extent, inosine.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDynamic landscape of alternative polyadenylation in aging across distinct tissues\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePolyadenylation is essential for eukaryotic mRNA maturation, and APA is an important layer for aging regulation. To explore poly(A) tail length (PAL) changes during aging, we estimated PAL for each transcript across tissue samples using Dorado (v0.6.2), yielding an average PAL of 91.20 nt (IQR: 82.57-100.40) in mouse and an average PAL of 85.63 nt (IQR: 84.60-86.56) in human (\u003cstrong\u003eSupplementary Table S4-1\u003c/strong\u003e). Notably, significant PAL alterations were observed during aging across tissues (\u003cem\u003eP\u003c/em\u003e\u0026lt;2×10\u003csup\u003e16\u003c/sup\u003e) (\u003cstrong\u003eSupplementary Fig. S4A, B\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003eTo delve into the dynamic choice of different polyadenylation sites in a gene during aging, we obtained a total of 48,460 PACs in mice and 31,053 PACs in human detected in ≥ 2 samples. These mouse PACs were annotated into 16,077 genes, of which 60.67% genes contain at least two poly(A) sites (\u003cstrong\u003eFig. 4A\u003c/strong\u003e). In human, PACs were mapped to 16,077 genes, with 55.55% containing ≥ two PACs (\u003cstrong\u003eFig. 4B\u003c/strong\u003e). Most PACs identified by DRS resided in 3’UTR, specifically 74.68% for mouse and 86.42% for human (\u003cstrong\u003eFig. 4C, D\u0026nbsp;\u003c/strong\u003eand\u003cstrong\u003e\u0026nbsp;Supplementary Fig. S4C, D\u003c/strong\u003e). Motif enrichment analyses of NUE elements for all PACs showed conserved classical hexamer motif A[U/A]UAAA across species (\u003cstrong\u003eSupplementary Fig. S4E\u003c/strong\u003e). Reads from PACs showed a bimodal peak corresponding to proximal and distal poly(A) sites across 3’UTR regions in all tissue types (\u003cstrong\u003eSupplementary Fig. S4F\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003eAnalysis of 21 published APA regulators expression revealed eight dysregulated during aging across tissues (\u003cstrong\u003eSupplementary Fig. S4G\u003c/strong\u003e), and this included Cpsf family member \u003cem\u003eCpsf1\u003c/em\u003e, and\u003cem\u003e\u0026nbsp;Cpsf2\u003c/em\u003e. To explore alternative 3’UTR patten during aging across tissue, hierarchical clusting of 4,376 mice and 947 human 3’UTR PACs revealed seven mice and six human clusters with significant age-dependent 3’UTR length variation (\u003cstrong\u003eFig. 4E, F\u003c/strong\u003e). In mouse, a pivotal PAS shift at 12 months drove 3’UTR shortening in Cluster 1 and 5 (C1, C5), and lengthening in Cluster 2 (C2) across all tissues, converging on kinase signaling (e.g., serine, threonine, MAPK and GTPase) and autophagy-proteostasis pathways (e.g., macroautophagy and protein catabolism) (\u003cstrong\u003eFig. 4E\u0026nbsp;\u003c/strong\u003eand \u003cstrong\u003eSupplementary Table S4-4\u003c/strong\u003e). Notably, tissue-specific patterns emerged. Hippocampus showed divergent regulation in Cluster 3 (C3, lengthening) and 4 (C4, shortening), which linked to synaptic vesicle transport and pyruvate metabolism; heart-specific Cluster 6 (C6) exhibited preferential proximal PAS usage yielding the shortest 3’UTRs. Lung-specific shortening occurred in Cluster 7 (C7) genes. Both genes in C6 and C7 were enriched for mitochondrial organization. In human, C5 showed age-related 3’UTR shortening whereas C6 displayed lengthening, with other clusters exhibiting non-monotonic dynamics. GO enrichment implicated core lung aging pathways including autophagy regulation, oxidative stress response, immune modulation, and cell fate determination (e.g., apoptosis, adipocyte differentiation) (\u003cstrong\u003eFig. 4F\u003c/strong\u003e and \u003cstrong\u003eSupplementary Table S4-5\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003eOf 1,140 mouse mitochondrial genes (MitoCarta3.0 [36]), 395 showed 3’UTR length changes; 12 shortened and 21 lengthened significantly at least one defined age stage in four mouse tissues (\u003cstrong\u003eFig. 4G, H, Supplementary Fig. S4H-K\u003c/strong\u003e and \u003cstrong\u003eSupplementary Table S4-2\u003c/strong\u003e), with 9 exhibiting tissue heterogeneity (e.g., \u003cem\u003eNdufa2\u003c/em\u003e with APA shifts at 12 months and sustained downregulation of transcript and protein with aging). Fifty-three mitochondrial genes showed 3’UTR length divergence in human lung tissue, exemplified by age-dependent 3’UTR shortening in \u003cem\u003eTIMM22\u003c/em\u003e (\u003cstrong\u003eSupplementary Table S4-2\u003c/strong\u003e). Interestingly, four mitochondrial genes, namely \u003cem\u003eMCL1\u003c/em\u003e, \u003cem\u003eBNIP3L, SOD1\u003c/em\u003e and \u003cem\u003eSTX17\u003c/em\u003e, occurred APA shift events in cross-species lung tissue.\u003c/p\u003e\n\u003cp\u003eAmong aging-related genes curated in the database aging atlas [37], we found that 9 and 13 genes with significant 3’UTR shortening and lengthening, respectively, during at least one defined age stage in four mouse tissues (\u003cstrong\u003eFig. 4G, H and Supplementary Fig. S4H-K\u003c/strong\u003e). Among pan-tissue 3’UTR shortening events, \u003cem\u003eIghm\u0026nbsp;\u003c/em\u003ewasobserved 3’UTR shortening at 12 and 21 months of age in lung tissue, while \u003cem\u003eTpml\u003c/em\u003e in skeletal muscle. Notably, we found tissue-specific 3’UTR shortening, including heart-specific \u003cem\u003eAlkbh5\u0026nbsp;\u003c/em\u003e(\u003cstrong\u003eSupplementary Table S4-2\u003c/strong\u003e), and hippocampus-specific \u003cem\u003eEtv1\u003c/em\u003e, \u003cem\u003eGpm6a\u003c/em\u003e, \u003cem\u003eKcnt1\u003c/em\u003e and \u003cem\u003eSyne1\u003c/em\u003e. Tissue-specific APA patterns were observed: seven skeletal muscle-specific (e.g., \u003cem\u003eFoxo1\u003c/em\u003e and \u003cem\u003eJun\u003c/em\u003e), one heart-specific (\u003cem\u003eMapt\u003c/em\u003e), 5 hippocampus-specific (e.g.,\u0026nbsp;\u003cem\u003eTrp53bp1\u003c/em\u003e), and 18 lung-specific (e.g., \u003cem\u003eErcc4\u003c/em\u003e) (\u003cstrong\u003eSupplementary Table S4-2\u003c/strong\u003e). In human, eight aging-related genes exhibited non-linear changes at 46 years of age, including \u003cem\u003eIL7R\u003c/em\u003e, \u003cem\u003eLMNA\u003c/em\u003e, \u003cem\u003eRORA\u003c/em\u003e, \u003cem\u003eTIMP2\u003c/em\u003e, \u003cem\u003eVEGFA\u003c/em\u003e, \u003cem\u003eGAPDH\u003c/em\u003e, \u003cem\u003eNFE2L2\u003c/em\u003e and \u003cem\u003eCXCL12\u003c/em\u003e (\u003cstrong\u003eFig. 4G, I\u0026nbsp;\u003c/strong\u003eand \u003cstrong\u003eSupplementary Fig. S4L\u003c/strong\u003e). Notably, we found that \u003cem\u003eGSK3B\u0026nbsp;\u003c/em\u003eand \u003cem\u003eSP1\u003c/em\u003e were lung-specific APA plasticity during aging across species (\u003cstrong\u003eSupplementary Table S4-2, 3\u003c/strong\u003e). Collectively, these findings highlighted the dual nature of APA events, namely evolutionarily conserved yet exhibiting marked tissue specificity.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTranscriptomic and splicing-level aging clocks across human and mouse organs reveal tissue-specific disease trajectories\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eLeveraging the single-base accuracy of direct RNA sequencing (DRS), we quantified transcript-level abundance and systematically examined age-dependent alternative splicing across organs. Transcript-level profiling revealed that lung aging in humans is distinguished by an immune-centric signature, whereas the aging brain is dominated by neuronal and Alzheimer’s-disease-associated isoforms--suggesting that transcriptomic remodelling, rather than modification status, may be the more immediate driver of age-related dysfunction (\u003cstrong\u003eFig. 5A\u003c/strong\u003e). Extending the comparison to mice, we found that the aging lung and hippocampus recapitulate the human immune and neurodegenerative programmes, respectively (\u003cstrong\u003eFig. 5B\u003c/strong\u003e). Skeletal-muscle aging was marked by simultaneous rewiring of immunity and metabolism, while the heart displayed a transcriptome signature enriched for pathways that presage cardiovascular disease--mirroring epidemiological observations linking aging to heightened cardiac risk (\u003cstrong\u003eFig. 5B\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003eSkipped exon (SE) events were markedly rare at 6 months of age, reached maximal prevalence at 12 months, and declined again by 21 months, indicating a transient but prominent role for SE-mediated regulation during mid-life (\u003cstrong\u003eSupplementary Fig. S5A\u003c/strong\u003e). Tissue specificity was also evident: the heart exhibited consistently low SE frequencies, whereas the lung displayed the highest burden at every time point. Focusing on the lung, differential SE analysis between age groups revealed a striking enrichment of affected genes in metabolic pathways (\u003cstrong\u003eSupplementary Fig. S5B\u003c/strong\u003e), suggesting that alternative splicing remodels metabolic networks as a conserved feature of mammalian aging. This temporal pattern was not restricted to SE events; analogous trends were observed for alternative 5’ splice sites (A5), alternative 3’ splice sites (A3), retained introns (RI) and mutually exclusive exons (MX) across all tissues examined (\u003cstrong\u003eSupplementary Fig. S6–9\u003c/strong\u003e). Collectively, our data establish a mid-life surge in alternative splicing--particularly in metabolic genes--as a pervasive, organ-specific signature of aging.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eRNA preparation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor RNA preparation, the Total RNA Kit I (R6834) was used following the manufacturer\u0026rsquo;s instructions. Briefly, cells or tissues were lysed with TRK Lysis Buffer supplemented with\u0026nbsp;\u0026beta;-mercaptoethanol. For these tissue samples, the lysis buffer was added directly to the culture vessel or homogenized with liquid nitrogen. The lysate was clarified by centrifugation at 14,000xg for 2 min, and RNA was purified using RNA binding columns. The RNA was washed sequentially with RNA Wash Buffers I and II, eluted with nuclease-free water, and quantified using a NanoDrop One spectrophotometer and Qubit 3.0 Fluorometer. RNA integrity was assessed by electrophoresis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConstruction of nanopore sequencing Library and sequencing\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePrepared RNA of each sample was used for a DRS library preparation using the Oxford Nanopore DRS protocol (SQK-RNA004, Oxford Nanopore Technologies). For each sample, we obtained over 20 million raw reads range from 19,006,308 to 25,330,731, with basecalling Phred scores predominantly \u0026gt;8 and a substantial portion of transcripts exhibiting \u0026gt;10x coverage. Reads exceeding 1.2 kb in length, were subjected to quality control and retained reads with mean Q score \u0026ge; 10 (default threshold) for downstream analysis. These results demonstrate that the datasets are of high quality and provide better outputs and coverage compared to datasets generated by MinION or GridION dRNA-seq platforms.\u003c/p\u003e\n\u003cp\u003eFor reversed connector connection, 8\u0026mu;L prepared RNA, 3\u0026mu;L NEBNext Quick Ligation Reaction Buffer (NEB), 1\u0026mu;L RT Adapter (RTA) (SQK-RNA004) and 1.5\u0026nbsp;\u0026mu;L T4 DNA Ligase (NEB) were mixed together and incubated under 25\u0026deg;C for 10 min. Afterwards, 8\u0026mu;L 5x first-strand buffer (NEB), 2\u0026mu;L 10 mMdINTPs (NEB), 9\u0026nbsp;\u0026mu;L Nuclease-free water, 4\u0026mu;L 0.1 M DTT (Thermo Fisher) and 2\u0026mu;L SuperScript II Reverse Transcriptase (Thermo Fisher Scientific, 18080044) were added into the above 15\u0026mu;L reaction system and incubated under 50\u0026deg;C for 50 min then 70\u0026deg;C for 10min. Reverse-transcribed mRNA was purified with 1.8* Agencourt RNAClean XP beads and washed with 23\u0026mu;l Nuclease-free water, and subsequently sequencing adapters were added using 8\u0026mu;L T4 Ligation Reaction Buffer, 6\u0026mu;L RNA Ligation Adapter and 4\u0026mu;L T4 DNA Ligase. The mix was purified and washed again as above, and 40\u0026mu;L RRB (SQK-RNA004) were added with 35\u0026mu;LNuclease-freewater. This final reaction system was loaded into Nanopore FLO-PRO004RA sequencing chip and sequenced for 48-72 hrs using PromethION sequencer (Oxford Nanopore Technologies).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eBase calling and alignment\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe raw nanopore sequencing data, stored in pod5 format, underwent basecalling using Dorado v.0.9.0 with the parameter --estimate-poly-a to generate FASTQ files. Subsequent quality filtering categorized the reads into \u0026lsquo;pass\u0026rsquo; (Q \u0026ge; 10) and \u0026lsquo;fail\u0026rsquo; (Q \u0026lt; 10) groups based on their mean quality scores, with the \u0026lsquo;pass\u0026rsquo; reads designated as clean reads for downstream analyses. These clean reads were then aligned to the reference genome (hg38 and mm10) using minimap2 v.2.17-r941 with parameters -ax splice -uf -k14. Finally, the alignment ratio of clean reads to the reference genes was calculated using Samtools (v1.10)\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRNA modification ratio estimation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo estimate the RNA modification ratio (m\u003csup\u003e5\u003c/sup\u003eC, m\u003csup\u003e6\u003c/sup\u003eA, \u0026Psi; and inosine), we used the lastest Dorado basecaller (model: [email protected]), which identifies modified sites at single nucleotide resolution with ultra-high precision (94.5%). Dorado can calculate the modification probability per read to determine the methylation status of the site. It counts the number of methylated and unmethylated reads and calculates the score value (Fraction) for the site using the following formula:\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFraction = n(M)/(n(UM) + n(M)).\u003c/p\u003e\n\u003cp\u003eWhere:\u0026nbsp;\u003c/p\u003e\n\u003cul start=\"50\"\u003e\n \u003cli\u003en(M): Number of methylated reads at a given site (reads identified as carrying the methylation mark).\u0026nbsp;\u003c/li\u003e\n \u003cli\u003en(UM): Number of unmethylated reads at the same site (reads identified as lacking the methylation mark).\u0026nbsp;\u003c/li\u003e\n \u003cli\u003en(UM) + n(M): The total number of valid reads covering the site (excluding missing or low-quality reads).\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eFraction: The methylation level (fraction methylated) at that site.\u0026nbsp;\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eWe used modkit (v0.4.1) for differential methylation loci (DML) analysis and differential transcript analysis. The \u0026ldquo;effect size\u0026rdquo; was defined the difference between Group1_pct_modified of group 1 and Group2_pct_modified of group 2. The threshold for selecting differential modified sites is |effect_size| \u0026gt;= 0.1 and \u003cem\u003eP\u003c/em\u003e \u0026lt; 0.05. The threshold for selecting differential modified transcripts is |log2fc| \u0026gt;= 1 and score \u0026gt; 3.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePoly(A) sites analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe employed the R package \u003cem\u003eQuantifypoly(A)\u0026nbsp;\u003c/em\u003e(v0.3.0) [38] to identify, cluster and annotate polyadenylation sites (poly(A) sites, or PAS) for each tissue sample. Adjacent PAS within a 24 nt side window [39] were merged to form new entities termed poly(A) site clusters (PACs). Reads mapping to each potential PAC were counted per sample for further analysis. Each PAC includes three functional elements: the far upstream element (FUE; 150 nt upstream), the near upstream element (NUE; 35 nt upstream), and the core element (CE; spanning -10 nt to +15 nt relative to the PAC). We extracted the NUE sequences from each PAC, which recruits the CPSF complex to precisely regulate the selection of APA sites, and conducted motif enrichment analyses by using HOMER (v4.1.1).\u003c/p\u003e\n\u003cp\u003eTo evaluate global alternative polyadenylation (APA) differences, we calculated poly(A) site usage rates for genes with multiple poly(A) sites. The Kolmogorov-Smirnov (KS) test was applied to compare these rates between groups. To calculate the preference for using proximal or distal PAC during aging, we selected the two most highly expressed PACs for each gene, and then calculated the the relative expression difference (RED) as: RED = log2(expression_proximal_PAC / expression_distal_PAC). Genes with RED \u0026ge; 0.3 or RED \u0026le; -0.3 were considered indicative of PAC usage bias.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo estimate 3\u0026rsquo;UTR shortening and lengthening during aging across tissues, we selected genes with differentially expressed and the highest abundant PACs within the 3\u0026rsquo;UTR. Using the weighted average 3\u0026rsquo;UTR length of each gene, we subsequently performed unsupervised hierarchical clusting analysis.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDifferential alternative splicing event analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo identify differential alternative splicing events, we employed the latest Dorado model in conjunction with the suppa2 (DiffSplice) software. Through calculating\u0026nbsp;\u0026Delta;PSI (dPSI) values and conducting statistical significance tests (p-values), we detected significant alternative splicing events between groups.\u003c/p\u003e\n\u003cp\u003eThe \u0026Delta;PSI (Cond1_Cond2_dPSI) was calculated using the formula: \u0026Delta;PSI = PSI\u003csub\u003e2\u003c/sub\u003e \u0026minus; PSI\u003csub\u003e1\u003c/sub\u003e. Here, Cond1 and Cond2 represent group 1 and group 2, respectively. PSI\u003csub\u003e2\u003c/sub\u003e is the average PSI value of samples in group 2, PSI\u003csub\u003e1\u003c/sub\u003e is the average PSI value of samples in group 1, and\u0026nbsp;\u0026Delta;PSI denotes the difference in PSI values between the two groups.\u003c/p\u003e\n\u003cp\u003eWe first screened for differential alternative splicing events by selecting those with |\u0026Delta;PSI| \u0026gt; 0, referred to as Diff_AS_num in the results. Further, we identified significant differential alternative splicing events as those with |\u0026Delta;PSI| \u0026gt; 0.1 and p-value \u0026lt; 0.05, designated as Significant_Diff_AS_num in the results. Subsequent functional enrichment analysis was performed on the genes associated with these significant differential alternative splicing events.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDifferentially expressed isoform analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConsensus sequences were generated using Flair v1.5.0 (-t 20) based on alignment results. Bam files were converted to bed12 format using bam2Bed12.py, followed by adding gene or transcript information via identify_gene_isoform.py. Redundancy was removed using the collapse program.Non-redundant transcripts were obtained by aligning consensus sequences to the reference genome and using StringTie v2.1.4 (--conservative -L -R) in merge mode (default parameters) for transcript reconstruction and redundancy removal.Coding regions of novel transcripts were predicted using TransDecoder v5.5.0 (-m 50, --single_best_only), which identifies potential CDS based on ORF length and log-likelihood scores.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGene Ontology (GO) enrichment analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGene Ontology (GO) enrichment analyses were conducted using the DAVID database to identify overrepresented biological processes, cellular components, and molecular functions among genes exhibiting significant differences in both expression levels and RNA modification patterns.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eProteomic data analysis across mouse and human aging\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe mined the C57BL/6J aging proteome from the GEO repository (GSE225576), which encompasses high-resolution mass-spectrometry profiles of brain, heart, lung and skeletal muscle sampled at different age. Raw spectral intensities were log\u003csub\u003e2\u003c/sub\u003e-transformed and quantile-normalised. We then extracted the expression values of 34 well-characterised m\u003csup\u003e6\u003c/sup\u003eA-associated RNA-binding proteins (RBPs)--including writers (METTL3, METTL14, WTAP), erasers (FTO, ALKBH5) and readers (YTHDF1/2/3, IGF2BP1/2/3)--curated from the literature. Tissue- and age-stratified expression matrices were analysed in R v4.2.1. Group-wise means and s.d. were calculated, and tissue-specific trajectories were visualised with \u003cem\u003eggplot2\u003c/em\u003e v3.4.0. Statistical significance was assessed by one-way ANOVA followed by Tukey HSD post-hoc testing (\u0026alpha; = 0.05). The resulting atlas reveals pronounced tissue specificity and age-dependent remodelling of the m\u003csup\u003e6\u003c/sup\u003eA proteome, providing an orthogonal resource to our direct-RNA-sequencing datasets. The human aging proteome was from the GSA repository HRA009355, HRA011245. Data processing and statistical analyses were performed identically to those used for the mouse datasets.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eRecent studies have begun to elucidate the role of RNA methylation in aging[15, 40]. For instance, Ricardo et al. demonstrated that reduced m\u003csup\u003e6\u003c/sup\u003eA-modified transcripts are associated with impaired synaptic protein synthesis and cognitive dysfunction, highlighting a potential mechanistic link between m\u003csup\u003e6\u003c/sup\u003eA and neuronal aging[41]. However, the reported effects of m\u003csup\u003e6\u003c/sup\u003eA-modifying enzymes during mammalian aging remain inconsistent. Notably, in long-lived endocrine mutant mice, elevated levels of key m\u003csup\u003e6\u003c/sup\u003eA regulator--including METTL3, METTL14, YTHDF1, and YTHDF2--have been observed, which enhance cap-independent translation and promote stress resistance and healthy aging[42]. We further assessed whether RNA modification enzymes are involved in the role of m\u003csup\u003e6\u003c/sup\u003eA during aging by analyzing proteomic data from mice at different ages[43], and found that ALKBH5 and METTL3 exhibited more pronounced differences than other m\u003csup\u003e6\u003c/sup\u003eA regulator in protein levels across age groups.\u003c/p\u003e\n\u003cp\u003eDespite these insights, most previous studies have lacked precise, transcriptome-wide maps of RNA modifications. In this context, our use of the RNA004 direct RNA sequencing (DRS) technology provides a significant advance[44], enabling accurate, base-resolution detection of m\u003csup\u003e6\u003c/sup\u003eA and simultaneous profiling of multiple RNA modifications, including m\u003csup\u003e5\u003c/sup\u003eC, m\u003csup\u003e7\u003c/sup\u003eG, and\u0026nbsp;Ψ\u0026nbsp;(pseudouridine), in a single run. Compared to RNA002, RNA004 exhibits enhanced overall read accuracy at 93.5%, up from 92.1%. This improvement is mainly attributed to a reduction in mismatch and insertion errors. However, the deletion error rate remains elevated, indicating a need for further refinements[45]. Given the limited understanding of non-m\u003csup\u003e6\u003c/sup\u003eA RNA modifications in aging, our comprehensive multi-modification landscape offers a valuable resource for future mechanistic studies of RNA epigenetics in aged tissues. In this study, we employed dRNA-seq to conduct an in-depth analysis of the transcriptome and epitranscriptome in select senescent tissues of mice and humans. Our findings robustly corroborate prior studies. To our knowledge, this represents the first high-resolution, multi-modification atlas across aging-related samples, marking an important step toward deciphering the complexity of the epitranscriptome in the context of organismal aging.\u003c/p\u003e\n\u003cp\u003eOne limitation of our study lies in the relatively limited number of samples analyzed, particularly with respect to human-derived tissues. While the use of direct RNA sequencing (DRS) enabled high-resolution mapping of multiple RNA modifications, the current dataset may not fully capture the heterogeneity of RNA modification landscapes across individuals or populations. Expanding the sample size--especially by incorporating diverse human aging cohorts--will be essential to validate the generalizability of our findings and to better understand inter-individual variation in the epitranscriptomic regulation of aging.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eEthics declarations\u003c/h2\u003e\n\u003cp\u003eThe study received approval from the Ethics Committee of our institution (Approval Number: CY2023-032-01). Informed consent was obtained from all participants, and confidentiality was rigorously maintained.\u003c/p\u003e\n\u003ch2\u003eConflict of interest statement\u003c/h2\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\n\u003cp\u003eS.A. designed and performed most of the experiments, analyzed the data, and drafted the manuscript. H.Q. and S.Z. participated in experimental design, data collection, and manuscript revision. L.C., K.Y., Y.Q., J.M., W.Y. and W.L. assisted in sample preparation, sequencing experiments, and preliminary data analysis. K.M. provided technical support for RNA modification detection and bioinformatics analysis. L.Y. and Y.C. conceived and supervised the study, secured funding, revised the manuscript critically for important intellectual content, and gave final approval for publication. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003ch2\u003eAcknowledgement\u003c/h2\u003e\n\u003cp\u003eThis work was supported by grants from Noncommunicable Chronic Diseases-National Science and Technology Major Project (2023ZD0506102/2023ZD0506100), the National Natural Science Foundation of China (grant number 82203181); The Key Research and Development Program of Sichuan-Chongqing of the Chongqing Science and Technology Commission (grant number CSTB2022TIAD-CUX0001), the Guangxi Natural Science Foundation (2025GXNSFDA069011), and the Guangxi Medical University Training Program for Distinguished Young Scholars to Sanqi An.\u003c/p\u003e\n\u003ch2\u003eData Availability\u003c/h2\u003e\n\u003cp\u003eAll raw sequencing data of mice have been deposited to the GSA for Human in the Genome Sequence Archive (GSA) in National Genomics Data Center (https://ngdc.cncb.ac.cn/bioproject/browse/PRJCA036375).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eRudnicka E, Napierala P, Podfigurna A, Meczekalski B, Smolarczyk R, Grymowicz M. The World Health Organization (WHO) approach to healthy ageing. Maturitas. 2020;139:6-11.\u003c/li\u003e\n\u003cli\u003ePereira B, Correia FP, Alves IA, Costa M, Gameiro M, Martins AP, et al. Epigenetic Reprogramming as a Key to Reverse Ageing and Increase Longevity. Ageing Res Rev. 2024:102204.\u003c/li\u003e\n\u003cli\u003eMcCartney DL, Min JL, Richmond RC, Lu AT, Sobczyk MK, Davies G, et al. Genome-wide association studies identify 137 genetic loci for DNA methylation biomarkers of aging. Genome Biol. 2021;22(1):194.\u003c/li\u003e\n\u003cli\u003eMoaddel R, Ubaida-Mohien C, Tanaka T, Lyashkov A, Basisty N, Schilling B, et al. Proteomics in aging research: A roadmap to clinical, translational research. Aging Cell. 2021;20(4):e13325.\u003c/li\u003e\n\u003cli\u003eArgentieri MA, Xiao S, Bennett D, Winchester L, Nevado-Holgado AJ, Ghose U, et al. Proteomic aging clock predicts mortality and risk of common age-related diseases in diverse populations. Nat Med. 2024;30(9):2450-60.\u003c/li\u003e\n\u003cli\u003eKuo CL, Liu P, Drouard G, Vuoksimaa E, Kaprio J, Ollikainen M, et al. A proteomic signature of healthspan. Proc Natl Acad Sci U S A. 2025;122(23):e2414086122.\u003c/li\u003e\n\u003cli\u003eZhang S, Wang Z, Wang Y, Zhu Y, Zhou Q, Jian X, et al. A metabolomic profile of biological aging in 250,341 individuals from the UK Biobank. Nat Commun. 2024;15(1):8081.\u003c/li\u003e\n\u003cli\u003ePanyard DJ, Yu B, Snyder MP. The metabolomics of human aging: Advances, challenges, and opportunities. Sci Adv. 2022;8(42):eadd6155.\u003c/li\u003e\n\u003cli\u003eShen L, Zhao H, Xi Y, Wang Z, Deng K, Gou W, et al. Mapping the gut microbial structural variations in healthy aging within the Chinese population. Cell Rep. 2024;43(11):114968.\u003c/li\u003e\n\u003cli\u003eKong F, Deng F, Li Y, Zhao J. Identification of gut microbiome signatures associated with longevity provides a promising modulation target for healthy aging. Gut Microbes. 2019;10(2):210-5.\u003c/li\u003e\n\u003cli\u003ePang S, Chen X, Lu Z, Meng L, Huang Y, Yu X, et al. Longevity of centenarians is reflected by the gut microbiome with youth-associated signatures. Nat Aging. 2023;3(4):436-49.\u003c/li\u003e\n\u003cli\u003eBoccaletto P, Stefaniak F, Ray A, Cappannini A, Mukherjee S, Purta E, et al. MODOMICS: a database of RNA modification pathways. 2021 update. Nucleic Acids Res. 2022;50(D1):D231-D5.\u003c/li\u003e\n\u003cli\u003eWiener D, Schwartz S. The epitranscriptome beyond m(6)A. Nat Rev Genet. 2021;22(2):119-31.\u003c/li\u003e\n\u003cli\u003eRoundtree IA, Evans ME, Pan T, He C. Dynamic RNA Modifications in Gene Expression Regulation. Cell. 2017;169(7):1187-200.\u003c/li\u003e\n\u003cli\u003eWu Z, Lu M, Liu D, Shi Y, Ren J, Wang S, et al. m(6)A epitranscriptomic regulation of tissue homeostasis during primate aging. Nat Aging. 2023;3(6):705-21.\u003c/li\u003e\n\u003cli\u003eCastro-Hernandez R, Berulava T, Metelova M, Epple R, Pena Centeno T, Richter J, et al. Conserved reduction of m(6)A RNA modifications during aging and neurodegeneration is linked to changes in synaptic transcripts. Proc Natl Acad Sci U S A. 2023;120(9):e2204933120.\u003c/li\u003e\n\u003cli\u003eGruber AJ, Zavolan M. Alternative cleavage and polyadenylation in health and disease. Nat Rev Genet. 2019;20(10):599-614.\u003c/li\u003e\n\u003cli\u003eTang P, Yang Y, Li G, Huang L, Wen M, Ruan W, et al. Alternative polyadenylation by sequential activation of distal and proximal PolyA sites. Nat Struct Mol Biol. 2022;29(1):21-31.\u003c/li\u003e\n\u003cli\u003eLopez-Otin C, Blasco MA, Partridge L, Serrano M, Kroemer G. Hallmarks of aging: An expanding universe. Cell. 2023;186(2):243-78.\u003c/li\u003e\n\u003cli\u003eLouros N, Ramakers M, Michiels E, Konstantoulea K, Morelli C, Garcia T, et al. Mapping the sequence specificity of heterotypic amyloid interactions enables the identification of aggregation modifiers. Nat Commun. 2022;13(1):1351.\u003c/li\u003e\n\u003cli\u003eMei H, Boom J, El Abdellaoui S, Abdelmohsen K, Munk R, Martindale JL, et al. Alternative Polyadenylation Utilization Results in Ribosome Assembly and mRNA Translation Deficiencies in a Model for Muscle Aging. J Gerontol A Biol Sci Med Sci. 2022;77(6):1130-40.\u003c/li\u003e\n\u003cli\u003eYoon OK, Brem RB. Noncanonical transcript forms in yeast and their regulation during environmental stress. RNA. 2010;16(6):1256-67.\u003c/li\u003e\n\u003cli\u003eShepard PJ, Choi EA, Lu J, Flanagan LA, Hertel KJ, Shi Y. Complex and dynamic landscape of RNA polyadenylation revealed by PAS-Seq. RNA. 2011;17(4):761-72.\u003c/li\u003e\n\u003cli\u003eLi H, Wang Z, Ma T, Wei G, Ni T. Alternative splicing in aging and age-related diseases. Translational Medicine of Aging. 2017;1:32-40.\u003c/li\u003e\n\u003cli\u003ePal S, Gupta R, Davuluri RV. Alternative transcription and alternative splicing in cancer. Pharmacol Ther. 2012;136(3):283-94.\u003c/li\u003e\n\u003cli\u003eRaj T, Li YI, Wong G, Humphrey J, Wang M, Ramdhani S, et al. Integrative transcriptome analyses of the aging brain implicate altered splicing in Alzheimer\u0026apos;s disease susceptibility. Nat Genet. 2018;50(11):1584-92.\u003c/li\u003e\n\u003cli\u003eBowles KR, Silva MC, Whitney K, Bertucci T, Berlind JE, Lai JD, et al. ELAVL4, splicing, and glutamatergic dysfunction precede neuron loss in MAPT mutation cerebral organoids. Cell. 2021;184(17):4547-63 e17.\u003c/li\u003e\n\u003cli\u003eShen X, Wang C, Zhou X, Zhou W, Hornburg D, Wu S, et al. Nonlinear dynamics of multi-omics profiles during human aging. Nat Aging. 2024;4(11):1619-34.\u003c/li\u003e\n\u003cli\u003eDing Y, Zuo Y, Zhang B, Fan Y, Xu G, Cheng Z, et al. Comprehensive human proteome profiles across a 50-year lifespan reveal aging trajectories and signatures. Cell. 2025.\u003c/li\u003e\n\u003cli\u003eTarkhov AE, Lindstrom-Vautrin T, Zhang S, Ying K, Moqri M, Zhang B, et al. Nature of epigenetic aging from a single-cell perspective. Nat Aging. 2024;4(6):854-70.\u003c/li\u003e\n\u003cli\u003eYing K, Liu H, Tarkhov AE, Sadler MC, Lu AT, Moqri M, et al. Causality-enriched epigenetic age uncouples damage and adaptation. Nat Aging. 2024.\u003c/li\u003e\n\u003cli\u003eZhong ZD, Xie YY, Chen HX, Lan YL, Liu XH, Ji JY, et al. Systematic comparison of tools used for m(6)A mapping from nanopore direct RNA sequencing. Nat Commun. 2023;14(1):1906.\u003c/li\u003e\n\u003cli\u003ePratanwanich PN, Yao F, Chen Y, Koh CWQ, Wan YK, Hendra C, et al. Identification of differential RNA modifications from nanopore direct RNA sequencing with xPore. Nat Biotechnol. 2021;39(11):1394-402.\u003c/li\u003e\n\u003cli\u003eJain M, Abu-Shumays R, Olsen HE, Akeson M. Advances in nanopore direct RNA sequencing. Nat Methods. 2022;19(10):1160-4.\u003c/li\u003e\n\u003cli\u003eAn S, Huang W, Huang X, Cun Y, Cheng W, Sun X, et al. Integrative network analysis identifies cell-specific trans regulators of m6A. Nucleic Acids Res. 2020;48(4):1715-29.\u003c/li\u003e\n\u003cli\u003eRath S, Sharma R, Gupta R, Ast T, Chan C, Durham TJ, et al. MitoCarta3.0: an updated mitochondrial proteome now with sub-organelle localization and pathway annotations. Nucleic Acids Res. 2021;49(D1):D1541-D7.\u003c/li\u003e\n\u003cli\u003eAging Atlas C. Aging Atlas: a multi-omics database for aging biology. Nucleic Acids Res. 2021;49(D1):D825-D30.\u003c/li\u003e\n\u003cli\u003eYe C, Zhao D, Ye W, Wu X, Ji G, Li QQ, et al. QuantifyPoly(A): reshaping alternative polyadenylation landscapes of eukaryotes with weighted density peak clustering. Brief Bioinform. 2021;22(6).\u003c/li\u003e\n\u003cli\u003eTian B, Hu J, Zhang H, Lutz CS. A large-scale analysis of mRNA polyadenylation of human and mouse genes. Nucleic Acids Res. 2005;33(1):201-12.\u003c/li\u003e\n\u003cli\u003eFu YD, Jiang F, Zhang X, Pan YY, Xu R, Liang X, et al. Perturbation of METTL1-mediated tRNA N- methylguanosine modification induces senescence and aging. Nature Communications. 2024;15(1).\u003c/li\u003e\n\u003cli\u003eCastro-Hern\u0026aacute;ndez R, Berulava T, Metelova M, Epple R, Centeno TP, Richter J, et al. Conserved reduction of m6A RNA modifications during aging and neurodegeneration is linked to changes in synaptic transcripts. P Natl Acad Sci USA. 2023;120(9).\u003c/li\u003e\n\u003cli\u003eSun J, Cheng BK, Su YK, Li M, Ma SY, Zhang Y, et al. The Potential Role of m6A RNA Methylation in the Aging Process and Aging-Associated Diseases. Frontiers in Genetics. 2022;13.\u003c/li\u003e\n\u003cli\u003eTakasugi M, Nonaka Y, Takemura K, Yoshida Y, Stein F, Schwarz JJ, et al. An atlas of the aging mouse proteome reveals the features of age-related post-transcriptional dysregulation. Nature Communications. 2024;15(1).\u003c/li\u003e\n\u003cli\u003eLiu-Wei W, van der Toorn W, Bohn P, H\u0026ouml;lzer M, Smyth RP, von Kleist M. Sequencing accuracy and systematic errors of nanopore direct RNA sequencing. Bmc Genomics. 2024;25(1).\u003c/li\u003e\n\u003cli\u003eMears MC, Read QD, Bakre A. Comparison of direct RNA sequencing of using two different chemistries on the MinION platform. J Virol Methods. 2025;333.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"genome-biology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"gbio","sideBox":"Learn more about [Genome Biology](https://genomebiology.biomedcentral.com/)","snPcode":"13059","submissionUrl":"https://submission.springernature.com/new-submission/13059/3","title":"Genome Biology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"RNA modification, aging, Nanopore direct RNA sequencing, alternative polyadenylation, alternative splicing","lastPublishedDoi":"10.21203/rs.3.rs-7521909/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7521909/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBACKGROUND: \u003c/strong\u003eAging involves complex molecular alterations across tissues; however, a comprehensive understanding of epitranscriptomic dynamics remains elusive.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRESULTS: \u003c/strong\u003eHere, leveraging high-quality dRNA-seq, we simultaneously constructed a multi-tissue (lung, skeletal muscle, heart, and hippocampus), cross-species (mouse and human) dynamics atlas of RNA features during aging, namely \u003cem\u003eN\u003c/em\u003e6-methyladenosine (m\u003csup\u003e6\u003c/sup\u003eA), 5-methylcytidine (m\u003csup\u003e5\u003c/sup\u003eC), pseudouridine (Ψ), inosine, alternative polyadenylation (APA), and splicing. Our analysis revealed widespread non-linear, age-dependent shifts across these RNA features, exhibiting both conserved patterns and striking tissue specificity. Strikingly, a pronounced, transient peak in these shift events of RNA features occurred at mid-life (12 months of age in mice, 46 years of age in human), and affected the expression of aging-related, mitochondrial, and metabolic genes.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCONCLUSIONS: \u003c/strong\u003eCollectively, we present a high-resolution atlas of m\u003csup\u003e6\u003c/sup\u003eA, m\u003csup\u003e5\u003c/sup\u003eC, pseudouridine and inosine modifications, alternative splicing and poly(A) site usage across tissues and aging stages, derived from direct RNA sequencing. This resource offers an unprecedented foundation for dissecting the RNA-centric mechanisms that govern mammalian aging.\u003c/p\u003e","manuscriptTitle":"Direct RNA sequencing reveals multi-layered regulation of the aging transcriptome","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-15 14:50:06","doi":"10.21203/rs.3.rs-7521909/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-01-07T08:14:48+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-20T03:41:15+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-11T15:40:50+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"465380565089267756307030719958758649","date":"2025-10-31T02:10:27+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"125939182462715906921468686739419238632","date":"2025-10-29T10:20:38+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-22T09:13:23+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"51686673364473720867787145007628278354","date":"2025-10-06T02:00:43+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-10-02T14:06:19+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-09-04T11:07:22+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-09-03T06:22:00+00:00","index":"","fulltext":""},{"type":"submitted","content":"Genome Biology","date":"2025-09-03T02:09:12+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"genome-biology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"gbio","sideBox":"Learn more about [Genome Biology](https://genomebiology.biomedcentral.com/)","snPcode":"13059","submissionUrl":"https://submission.springernature.com/new-submission/13059/3","title":"Genome Biology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"7a2d4861-ea6d-4aec-9bb3-a76a6e28aa64","owner":[],"postedDate":"October 15th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-01T21:23:15+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-15 14:50:06","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7521909","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7521909","identity":"rs-7521909","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}