The serum small non‐coding RNA (SncRNA) landscape as a molecular biomarker of age associated muscle dysregulation and insulin resistance in older adults

Abstract Small noncoding RNAs (sncRNAs) are implicated in age‐associated pathologies, including sarcopenia and insulin resistance (IR). As potential circulating biomarkers, most studies have focussed on microRNAs (miRNAs), one class of sncRNA. This study characterized the wider circulating sncRNA transcriptome of older individuals and associations with sarcopenia and IR. sncRNA expression including miRNAs, transfer RNAs (tRNAs), tRNA‐associated fragments (tRFs), and piwi‐interacting RNAs (piRNAs) was measured in serum from 21 healthy and 21 sarcopenic Hertfordshire Sarcopenia Study extension women matched for age (mean 78.9 years) and HOMA2‐IR. Associations with age, sarcopenia and HOMA2‐IR were examined and predicted gene targets and biological pathways characterized. Of the total sncRNA among healthy controls, piRNAs were most abundant (85.3%), followed by tRNAs (4.1%), miRNAs (2.7%), and tRFs (0.5%). Age was associated (FDR < 0.05) with 2 miRNAs, 58 tRNAs, and 14 tRFs, with chromatin organization, WNT signaling, and response to stress enriched among gene targets. Sarcopenia was nominally associated (p < .05) with 12 tRNAs, 3 tRFs, and 6 piRNAs, with target genes linked to cell proliferation and differentiation such as Notch Receptor 1 (NOTCH1), DISC1 scaffold protein (DISC1), and GLI family zinc finger‐2 (GLI2). HOMA2‐IR was nominally associated (p<0.05) with 6 miRNAs, 9 tRNAs, 1 tRF, and 19 piRNAs, linked with lysine degradation, circadian rhythm, and fatty acid biosynthesis pathways. These findings identify changes in circulating sncRNA expression in human serum associated with chronological age, sarcopenia, and IR. These may have clinical utility as circulating biomarkers of ageing and age‐associated pathologies and provide novel targets for therapeutic intervention.


| INTRODUCTION
Ageing is characterized by a gradual loss of general function, and reduced repair capacity which leads to an increased risk of mortality and susceptibility to multiple age-related pathologies.Muscle ageing is associated with a progressive impairment in metabolic function and a loss of muscle mass.As skeletal muscle is critical for posture and is the primary organ implicated in glucose clearance, age related changes in muscle have been linked to both sarcopenia and insulin resistance.However, the underlying molecular mechanisms leading to sarcopenia or insulin resistance remain unclear.
Small non-coding RNAs (sncRNAs) have been increasingly recognized as important regulators of many biological processes.sncRNA populations include microRNAs (miRNA), piwi-interacting RNAs (piRNAs), transfer RNAs (tRNA), and tRNA-associated fragments (tRFs).To date, the majority of research has focused on miRNAs which either repress translation or induce mRNA degradation of target transcripts through sequence-specific binding to the 3′UTR.In skeletal muscle, miRNAs play key roles in muscle homeostasis, controlling muscle mass, function, and metabolism.For example, miR-675-3p and miR-675-5p promote muscle differentiation and regeneration by repressing the bone morphogenetic protein (BMP) pathway through targeting the anti-differentiation SMAD transcription factors, SMAD1 and SMAD5 and the cell division control protein 6 (CDC6), 1 miR-1 modulates muscle cell growth by regulating IGF/PI3K/Akt signaling by directly targeting IGF-1 2 whilst miR-223 regulates glucose uptake by inhibiting Glut4 in muscle tissue. 3miR-NAs have also been implicated in muscle ageing through regulation of key genes including insulin-like growth factors, FOXO transcription factors, myostatin, NADdependent protein deacetylase sirtuin-1 (SIRT1), and transforming growth factor-β signaling pathways. 4iRNAs can be actively secreted from cells, either bound to RNA binding proteins, 5 high-density lipoproteins, 6 or released during cell death, 7 regulating mRNA targets in recipient cells and mediating cross talk between organs.The detection of circulating miRNAs has led to considerable interest in the use of miRNAs as circulating biomarkers of age-related pathologies as well as targets for novel intervention strategies.For example, the first observation of altered circulating miRNA levels during aging was miR-34a, which was elevated in the plasma of older mice. 8miR-34a was also increased in peripheral blood mononuclear cells (PBMCs) and brains of older mice, with a reciprocal decrease of its target Sirt1 mRNA.Furthermore, recent human studies have reported differential miRNA expression between young and aged individuals in several different peripheral fluids, including serum, [9][10][11][12] plasma, 13,14 and saliva. 15A number of studies have also reported differences in circulating miRNAs with respect to age associated pathologies including sarcopenia and insulin resistance.For example, eight miRNAs in plasma (miR-10a-3p,−92a-3p,−185-3p,−194-3p,−326,−532-5p,−576-5p, and −760), 16 and two miRNAs in serum (miR-21 and −203a-3p) 17 have been shown to be associated with sarcopenia status, while miR-29a, miR-34a, miR-375, miR-103, miR-107, miR-132, miR-142-3p, and miR-144 have been identified as potential circulating biomarkers of type 2 diabetes. 18owever, most studies investigating associations between circulating miRNAs and age-associated pathologies Of the total sncRNA among healthy controls, piRNAs were most abundant (85.3%), followed by tRNAs (4.1%), miRNAs (2.7%), and tRFs (0.5%).Age was associated (FDR < 0.05) with 2 miRNAs, 58 tRNAs, and 14 tRFs, with chromatin organization, WNT signaling, and response to stress enriched among gene targets.Sarcopenia was nominally associated (p < .05)with 12 tRNAs, 3 tRFs, and 6 piRNAs, with target genes linked to cell proliferation and differentiation such as Notch Receptor 1 (NOTCH1), DISC1 scaffold protein (DISC1), and GLI family zinc finger-2 (GLI2).HOMA2-IR was nominally associated (p<0.05) with 6 miRNAs, 9 tRNAs, 1 tRF, and 19 piRNAs, linked with lysine degradation, circadian rhythm, and fatty acid biosynthesis pathways.These findings identify changes in circulating sncRNA expression in human serum associated with chronological age, sarcopenia, and IR.These may have clinical utility as circulating biomarkers of ageing and ageassociated pathologies and provide novel targets for therapeutic intervention.

K E Y W O R D S
ageing, epigenetics, HOMA2-IR, insulin resistance, noncoding RNA, sarcopenia, serum have utilized targeted approaches to investigate altered miRNA expression, and the presence of sncRNA populations other than miRNAs in serum, and their potential as putative biomarkers has received little research attention.piRNAs function to maintain genome stability; in addition to their role in translation, tRNAs can regulate gene expression and have been shown to act as sensors of nutritional stress, [19][20][21] while tRFs, generated through enzymatic cleavage of the 5′ or 3′ ends of tRNAs, demonstrate miRNA like activity, and have been reported to regulate gene expression by binding to the promoters of target genes. 22,23Given the importance of these different classes of sncRNAs, we carried out global small RNA sequencing from serum samples from community dwelling older individuals with or without sarcopenia, to firstly characterize the sncRNA landscape of serum from aged individuals and subsequently investigate the associations of sncRNA expression levels with sarcopenia and insulin resistance; to date there have been limited studies investigating the association between circulating sncRNAs and these phenotypes.In addition, where possible we identified downstream gene targets of the sncRNAs, and pathways enriched amongst the target genes to gain insights into the underlying regulatory mechanisms altered during ageing and associated pathologies.

| Study participants
5][26][27] This study received ethical approval from the Hertfordshire Research Ethics Committee (number 07/Q0204/68) and was conducted in accordance with the 1964 Declaration of Helsinki and its later amendments.Sarcopenia status was defined according to the European Working Group on Sarcopenia in Older People (EWGSOP) 2010 definition, 28 with the following thresholds: ALMi (ALM/ height 2 ) ≤ 7.23 kg/m 2 for men and ≤5.67 kg/m 2 for women; grip strength < 30 kg for men and <20 kg for women; and walking speed ≤ 0.8 m/s.

RNA extraction
To examine the association between circulating sncRNAs and ageing associated pathologies with a particular focus on muscle dysregulation, serum samples (n = 42) were selected from the HSSe cohort; this included serum samples from 21 women with sarcopenia, and from 21 women without sarcopenia, selected to be matched for age and HOMA2-IR levels (Table 1).As the number of men with sarcopenia and sufficient serum for RNA extraction was limited, only serum samples from HSSe women participants were assessed in this study.Total RNA was extracted from serum samples using the miRNeasy Kit (Qiagen, UK) following manufacturers guidelines.RNA concentration and quality were quantified using Nanodrop and Qubit assays (Thermo Fisher, UK).Extracted RNA was treated with DNAse1 (Sigma, UK), snap frozen on dry ice, and stored at −80° until sequencing analysis.

| sncRNA sequencing
Total RNA was used to perform small RNA-seq (average approximately 18 million reads per sample) using the small RNA workflow (Oxford Genomics, OGC).RNA was  quantified using RiboGreen (Invitrogen) on the FLUOstar OPTIMA plate reader (BMG Labtech) and the size profile and integrity analyzed on the 2200 or 4200 TapeStation (Agilent, RNA ScreenTape).Small RNA library preparation was completed using NEBNext Small RNA kit (NEB) following manufacturer's instructions and applying the low input protocol modifications.Libraries were amplified (15 cycles) on a Tetrad (Bio-Rad) using in-house unique dual indexing primers (based on DOI: 10.1186/1472-6750-13-104). Size selection was performed using Pippin Prep instrument (Sage Science) using the 3% Agarose, dye free gel with internal standards (size selection: 125 to 160 bp).Individual libraries were normalized using Qubit, and size profile analyzed on the 2200 or 4200 TapeStation.The pooled library was diluted (~10 nM) for storage.The 10 nM library was denatured and further diluted prior to loading on the sequencer.Single end sequencing was performed using NextSeq500 platform (Illumina, NextSeq 500/550 v2.5 Kits, 75 cycles).BCL files were demultiplexed using bcl2fastq and fastq files generated.Adapters were trimmed using Trim Galore 29 discarding reads that were <10 bp after trimming and any reads longer than 50 bp.For miRNA analysis, the miRDeep2 30 pipeline was run.Briefly, reads were collapsed and aligned to the hg19 genome using the mapper.plscript, after which the miRDeep2.plscript was run to count both mature and hairpin miRNA species.Reads were aligned to tRFs using MINTmap. 31For all other sncRNA subtypes, fastq files were aligned to the hg19 genome using the short-read aligner Bowtie 32 using the following options: -q -k 10 -v 0 -S -t --best --strata.Gene counts were generated using featureCounts.  .5 | sncRNA predicted targets and biological significance 2.5.1 | miRNAs target information and pathway enrichment miRNA target genes and their related functional pathways were identified using DIANA-mirPath v3.0 (http:// snf-515788.vm.okean os.grnet.gr/ ). 34Predicted and/or experimentally validated gene targets were identified within mirPath software using Tarbase v7.0, the largest manually curated, experimentally validated miRNA-gene interactions database. 35Identified targets were subsequently used for KEGG and Gene Ontology pathway analysis.

| tRF target information and gene ontologies
To determine the functional role of tRFs, differentially expressed tRFs were entered into the tRF data base (tRFdb) (http:// genome.bioch.virgi nia.edu/ trfdb/ ) to identify known sequence information.tRForest (https:// trfor est.com/ ) was used to explore machine-learning predicted tRF gene targets and identify tRF associated gene ontologies (GO).For the GO output a tRForest generated dotplot was created showing the ten pathways with the highest gene ratios and to display data for the gene ratio, gene count, and adjusted p-value.Furthermore, a network plot was created using the cnetplot function, showing connections between genes and the highest-ranking pathways.Novel tRFs identified in the expression data that were not currently present in the tRFdb were reported with no predicted experimentally validated gene target interactions or associated ontologies.
2.5.4 | Combining different sncRNA population target genes to identify enriched pathways miRNA and tRF target genes identified using mirPath or tRForest were combined and investigated using Metascape. 36Briefly, for each gene target list, pathway process enrichment was carried out.Terms with a p-value <.01, a minimum of three genes, and an enrichment factor >1.5 were used. 37Protein-protein interaction (PPI) enrichment analysis was performed along with the molecular complex detection (MCODE) algorithm 38 in Metascape to identify densely connected network components.

| DISCUSSION
Here, we report differential expression of sncRNAs in human serum from older community dwelling women with respect to chronological age, sarcopenia status and HOMA2-IR.SncRNAs were strongly associated with age, with their gene targets enriched chromatin organization, WNT signaling and response to stress.Furthermore, the sarcopenia associated sncRNAs included gene targets involved in cell proliferation and differentiation, while the HOMA2-IR associated sncRNAs targeted lysine degradation, circadian rhythm, and fatty acid biosynthesis pathways.Such findings identify different classes of putative sncRNA biomarkers in serum associated with age and age associated pathologies; these sncRNA may function as epigenetic regulators to modify transcription of target genes involved in key molecular regulatory pathways.Collectively, our findings characterize for the first-time the global sncRNA landscape in serum from older individuals and provide novel targets for interventions aimed at improving ageing and health trajectories in older age.The analysis of sncRNA expression profiles identified four major sncRNA subtypes in serum from older individuals, including miRNAs, tRNAs, tRFs, and piRNAs.The expression of two miRNAs were found to be differentially expressed with age, miR-375/miR-375-3p and miR-769/miR-769-5p.miR-375/miR-375-3p, a 22-nucleotide mature miRNA located on the reverse strand of chromosome 2 was downregulated with age.miR-375 was first described in pancreatic islet cells where it functions as an important regulator of β cell development and function. 39,40However, more recently miR-375 has been shown function in a diverse number of cellular pathways, with changes in miR-375 expression reported in cancer, inflammation, autoimmune, cardiovascular diseases, and diabetes. 41For example, Xu et al., showed that miR-375-3p suppresses tumorigenesis and partially reverses chemoresistance by downregulating Hippo signaling through modulation of the Hippo-YAP1 pathway downstream genes CTGF, cyclin D1, and BIRC5. 423][44][45] Furthermore, miR-375 has been reported to directly target FOXO1, 46 which has been shown in both experimental models and humans studies to play an important role in longevity through the regulation of cellular processes such as insulin and insulin-like growth factor signaling, metabolism, autophagy, DNA damage repair, and oxidative stress resistance, 47,48 Thus, downregulation of miR-375/miR-375-3p may reduce repression of target genes within Hippo/FOXO signaling pathways with impacts on ageing and longevity.The second miRNA strongly associated with age was miR-769/ miR-769-5p which has been reported to be upregulated in many cancers; here expression of miR-769/miR-769-5p was increased with age.A number of studies have also reported changes to circulating miRNA expression with age, with for example miR-17, miR-19b, miR-20a, and miR-106a identified across a range of ageing models. 49he miRNAs associated with age identified in this study have not been reported in other ageing studies to date, although their target genes were enriched in ageing related pathways; here however we have assessed serum miRNAs using a non-targeted approach and in aged individuals over a relatively narrow age range, which may account for the differences observed.Interestingly, even over the narrow age range of individuals assessed in this study a strong linear relationship was observed between F I G U R E 9 Molecular complex detection (MCODE) algorithm run against PPI enrichment networks from combined age associated sncRNAs predicted target genes to identify densely connected network components.The MCODE networks identified for combined gene lists (8) are shown colored by cluster.
the expression of these and chronological age, whether this reflects large differences in the ageing processes between older individuals during this period of the life-course will need to be determined using a larger number of individuals across a wider age range.
Pathway analysis showed that lysine degradation and central carbon metabolism were enriched amongst the gene targets of miR-769/miR-769-5p, suggesting altered expression of these miRNAs maybe associated with changes in metabolism during ageing.Metabolic alterations have been shown to contribute to aging, [50][51][52][53] with ageing clocks recently developed based upon some of the characteristic metabolic changes observed with age. 54For example, during ageing glucose metabolism shifts from aerobic to anaerobic where oxygen consumption and ATP synthesis are not tightly coupled, leading to a reduction in ATP availability, and increased ROS production.Abnormally high levels of ROS can directly induce genomic instability and increase HIF-1α levels, promoting metabolic programming towards the Warburg effect. 55As the miR-NAs identified in this study associated with age regulate mRNAs with key regulatory roles in both longevity and metabolic pathways, the altered expression of miRNAs in the present study could either reflect ageing dependent changes in metabolism and/or be part of the casual pathway by which the changes in metabolism occur.
Along with differential expression of miRNAs, 58 tRNAs and 14 tRFs were also associated with age, demonstrating an age-associated epigenetic profile that exists across multiple sncRNA subtypes.Of the 14 tRFs differentially associated with age, tRF-16-RPM830D (tRF-5019a, tRF-5019b) and tRF-31-87R8WP9N1EWJ0 (tRF-5030c) have previously been characterized within tRFdb.tRF-16-RPM830D (tRF-5019a, tRF-5019b) a 5′-tRF, a class of tRF produced from mature tRNAs by cleavage of the 5′ end in the D-loop, 56 was upregulated with age and has been associated with a number of gene targets involved in pathways linked with the regulation of small GTPase mediated signal transduction.Small GTPases have been strongly linked to aging pathways, for example the small GTPase Ras is a signaling intermediary of the mammalian insulin/IGF-1-signaling (IIS) pathway known to play an evolutionary conserved role in lifespan, through the activation of FOXO transcription factors via inhibition of the lipid kinase PI3K and its downstream target AKT. 57Thus, modulation of FOXO signaling pathways by tRFs, together with the miRNAs previously reported could indicate a synergistic effect of different sncRNA populations acting the modulate similar pathways.
Age-associated changes within the different populations of ncRNAs may suggest crosstalk between sncRNA subtypes to modify cellular communication and cell specific signaling pathways associated with ageing, akin to those observed between long non-coding RNAs (ln-cRNAs) and miRNAs of which lncRNAs act as miRNA sponges, modulating its availability to endogenous mRNA targets. 58,59Combining available gene targets for the age associated sncRNAs revealed the top pathways enriched amongst the combined age associated sncRNAs included chromatin organization, WNT signaling, cellular response to stress, and oncogene induced senescence, pathways implicated in ageing across a range of cell types, [60][61][62][63][64] suggesting that the circulating sncRNA signature may reflect the ageing process observed within cells.
Associations were observed between circulating tRNAs, tRFs, and piRNAs and sarcopenia status in older individuals.There were three tRFs significantly associated with sarcopenia, however, only tRF-5003c was currently characterized in the tRF database.Gene targets of tRF-5003c were enriched in pathways involved in neural precursor cell proliferation, this enrichment was driven by the tRF-5003c targets NOTCH1, DISC1, and GLI2, which play key role in cell proliferation and differentiation in a number of different cell types including muscle satellite cells.For example, NOTCH signaling is a key determinant of muscle regenerative potential, with reduced NOTCH activation in satellite cells (SC) associated with a decrease in SC function and impaired muscle regeneration, 65 while DISC1 is essential for oxidative phosphorylation and 66 GLI2 is a regulator of both MYF5 and MYOD, key myogenic regulators required for SC differentiation and muscle regeneration. 67,68In our study there was a positive association between tRF-5003c and sarcopenia status suggesting that increased tRF-5003c expression may contribute to the reduced expression of NOTCH, DISC1, and GLI2 and impaired muscle function.
The incidence of insulin resistance and type 2 diabetes (T2D) have also been shown to increase with age. 69In this study, HOMA2-IR levels were negatively associated with miR-10b levels.miR-10b has previously been reported to be reduced in muscle from hyperglycemic compared to normoglycemia rats, 70 and in muscle tissue from twins discordant for T2D. 71The top pathways enriched amongst the miRNAs associated with HOMA2-IR levels were lysine degradation, circadian rhythm, and fatty acid metabolism.3][74][75] Furthermore, lysine degradation was one of the pathways enriched amongst the age-associated sncRNAs, suggesting that both age and HOMA2-IR may act on the same pathways through sncRNA-mediated mechanisms, identifying a potentially novel mechanism by which age modulates insulin sensitivity.Alterations in the circadian clock and fatty acid metabolism have also been implicated in mediating severity of insulin resistance.patients with T2D, circadian changes in insulin sensitivity were abolished compared to healthy individuals.Furthermore, specific disruption of the muscle clock resulted in diminished insulin sensitivity in the muscle, causing hyperglycemia in the non-fasting condition and glucose intolerance. 76In skeletal muscle, free fatty acids (FFAs) inhibit insulin-stimulated glucose uptake at the level of glucose transport and/or phosphorylation through mechanisms that involve intramyocellular accumulation of diacylglycerol (DAG) and long-chain acyl-CoA, activation of protein kinase C (PKC), and decreased tyrosine phosphorylation of insulin receptor substrate 1/2 (IRS-1/2).Given the enrichment of fatty acid metabolism amongst the gene targets of the HOMA2-IR associated sncRNAs, a combined sncRNA and lipidomic signature may have utility as predictive model of IR, as fatty acids are known not only to be important for membrane fluidity but also act as signaling molecules and epigenetic regulators. 77There was no overlap between the IR and sarcopenia enriched pathways, although a bi-directional link between IR and sarcopenia has been suggested, 78 however the gene targets of the majority of the sarcopenia associated sncRNAs are uncharacterized at present.
The main strengths of this study are that we have characterized, for the first time, the circulating sncRNA landscape of serum from older individuals, identifying potential biomarkers of chronological age, sarcopenia and HOMA2-IR.Furthermore, changes in expression of sn-cRNA populations have been linked to regulatory target genes and gene pathways implicated in key biological processes associated with ageing, skeletal muscle regulation, and insulin resistance.However, there remain several limitations to this study.Firstly, although this was an exploratory study using small RNA-seq to assess the expression of the sncRNAs within serum of older individuals, the sample size was relatively small and this maybe a reason why we only observed nominally significant results in some of the analyses.Secondly, the age range of our participants was limited and confined to older women, due to the availability of samples; further studies using larger numbers of participants over a wider age range, and including both women and men, together with replication in a second independent cohort would provide further insight into the viability of using the sncRNAs identified in this study as biomarkers of ageing across the lifecourse.Although, the target genes of the age associated sncRNAs identified in this study were enriched in processes such as chromatin organization, and senescence which are well established markers of the ageing process observed across many cell types, suggesting circulating sncRNA profiles may reflect tissue ageing and have utility as valuable predictors of chronological and biological age. 79Thirdly, due to the high degree of novelty of a proportion of the differentially expressed sncRNAs identified, especially tRFs and piRNAs, which have only recently been identified, it was not possible to identify predicted gene targets, ontologies or pathways and therefore assign specific biological function for all differentially expressed sncRNAs, requiring further investigation to identify the functional role of these sncRNAs.However, the identification of novel sncRNAs within this study associated with age, sarcopenia, and HOMA2-IR reveals promising biomarkers and provides an opportunity for future biological characterization of predicted gene targets and pathways involved in the regulation of age-associated muscle dysregulation and insulin resistance.

| CONCLUSION
We identify changes in individual sncRNA populations within human serum that are associated with chronological age, sarcopenia and HOMA2-IR in older community dwelling women.Furthermore, we identify predicted gene targets of these sncRNAs and pathway enrichment of the predicted genes in important biological processes linked with molecular regulation of each phenotype.These findings support the premise that epigenetic regulatory mechanisms may contribute to ageing, the age-related decline in glycemic control, and may be important regulators of muscle health in older age.Moreover, they provide a highly novel set of serum-based biomarkers in humans for the development of intervention strategies which could function to modulate the epigenetic landscape of ageing and associated pathologies.

T A B L E 1
Participant characteristics.

3. 4 . 1 |
Predicted gene target and pathway analysis of sarcopenia associated sncRNAs are associated with oxidoreductase activity and neural precursor cell proliferation/regulation Of the three nominally significant (p < .05)tRFs associated with sarcopenia, tRF-31-P4R8YP9LON4VD was F I G U R E 2 Volcano plots of age associated (FDR < 0.05) miRNAs (A), tRFs, (B) and (p < .05)piRNAs (C).Regression graphs of top two differentially expressed miRNAs (D), tRFs (E), piRNAs (F), respectively.For volcano plots differentially expressed FDR (FDR < 0.05) associated sncRNAs are in blue, nominally associated (p < .05)sncRNAs are in green and all others are in black.Dashed black line represents an FDR < 0.05 and the dashed grey line represents p-value < .05.

F
I G U R E 8 (A) Heat map of age associated sncRNA enriched pathways (top 20) across input predicted target gene lists colored by pvalues with darker orange representing greater significance.(B) Circos plots showing the overlaps between sncRNA target gene lists (i) at the gene level (purple links identical genes), (ii) including the shared term level, with linked genes (blue) belonging to the same enriched ontology term.The inner circle represents gene lists, with hits arranged along the arc.Genes that hit multiple lists are colored in dark orange, and genes unique to a list are shown in light orange.

2.4 | Small RNAseq data processing and analysis
further analysis.Outliers were determined visually from PCA plots.The sncRNAseq data can be accessed on the gene expression omnibus (https:// www.ncbi.nlm.nih.gov/ geo/ ), under accession number GSE231785.

Baseline sncRNA abundance in human serum from older individuals
Participant characteristics are summarized in Table1.RNA was extracted from serum samples from 42 women from the HSSe; this comprised 21 individuals diagnosed with sarcopenia, with a mean age of 79.15 (± SD 2.77); these were the only individuals with sarcopenia in the HSSe cohort who had sufficient serum available for RNA extraction.21individualswithout sarcopenia were also selected from the HSSe cohort, these were matched with the sarcopenic participants for age (mean age of 78.57.1 (± SD 2.49)) and HOMA2-IR levels (0.88 ± SD 0.50).Sarcopenic participants had a mean grip strength (kg) of 17.95 (± SD 4.84), gait speed (m/s) of 0.88 (±SD 0.15), and ALMi (kg/m 2 ) of 5.64 (±SD 0.40) compared to controls with a mean grip strength (kg) of 22.19 (±SD 4.70), gait speed (m/s) of 1.01 (±SD 0.19), and ALMi (kg/ m 2 ) of 6.10 (±SD 0.65).3.2 |sncRNA profiles of control serum samples (n = 21) were used to determine the abundance of the different classes of sncRNA based on absolute read counts.piRNAs were the most abundant (85.3%) sncRNA species detected in serum, followed by tRNAs (4.06%), miRNAs (2.74%), and tRFs (0.54%) (Figure