C57BL6NJcl male mice were purchased from CLEA Japan. They were acclimated for at least one week before initiating the experiments. The mice were provided a solid diet of CE2 (CLEA Japan) and water ad libitum and were housed in a conventional animal room with 12 h light/dark cycles at room temperature and 40-50% humidity. The mice were housed up to 5 mice per cage containing bedding, feed, and water, which were changed weekly. The mice were observed 2-3 times/day for monitoring. No abnormalities in mouse health or behavior were observed. Five 8-week-old mice were designated ‘Young’ mice, and five 82-week-old and three 102-week-old mice were designated ‘Aging’ mice. Small animal anesthesia machines (Muromachi Kikai) were used to anesthetize the mice. Isoflurane vaporized to a concentration of 4-5% was administered to the mice and maintained at 2-3% throughout the experiment. Following anesthesia, approximately 0.5-1.0 ml of blood was drawn from the heart, and the mice were promptly cervically dislocated to minimize distress. The time from the beginning of anesthesia to the end of blood collection was less than 10 min per animal. Death was confirmed by respiratory and cardiac arrest. Blood was placed in a Microtainer (cat. no. 365967, Becton Dickinson) for serum separation. After checking for coagulation, the blood was centrifuged at 6,000 x g for 3 min. The serum was separated and stored at -80˚C until use. The experiment was approved by the Hirosaki University Animal Experiment Ethics Committee and conducted based on the Hirosaki University Animal Experiment Guidelines (Approval No. AE01-2023-097-1).

The serum was filtered through a 0.20 µm filter and the RNAs derived from EVs were extracted from 200 µl of serum using the exoRNeasy midi kit (cat. no. 77144, Qiagen) and cel-miR-39 was added as a spike-in RNA. The concentration of the extracted EV-derived RNAs was measured using a Qubit™ microRNA Assay Kit (cat. no. Q32880, ThermoFisher Scientific) and a Qubit 4 Fluorometer (cat. no. Q33238, ThermoFisher Q33238, ThermoFisher Scientific) based on the manufacturer's protocol. To confirm the size of the RNAs in the serum EVs, an Agilent 2100 Bioanalyzer (Agilent Technologies, Inc.) and the Agilent RNA 6000 Pico kit (cat. no. 5067-1513, Agilent Technologies, Inc.) were used to perform RNA electrophoresis based on the manufacturer's protocol.

To examine miRNA expression in mouse serum, a miRNA microarray analysis was performed using 1.8 ng of RNAs extracted by the method described above. Microarray analysis was performed as previously described according to the manufacturer's instructions (25,26). The miRNA Complete Labeling Reagent and Hyb kit (cat. no. 5190-0456, Agilent Technologies) was used to label mouse serum EVs-derived RNAs with Cyanine-3 and hybridized to SurePrint G3 Mouse 8x60 K miRNA microarray slides (version 21.0) (cat. no. G4872A, design ID: 070155, Agilent Technologies) at 55˚C for 20 h. The fluorescence signals were detected using a SureScan Microarray Scanner (Agilent Technologies) and quantified using Feature Extraction software (Agilent Technologies). The microarray data were analyzed using R (version 4.3.1) and quantile normalization was done using limma (27). Microarray data were analyzed for genes with a fold-change cutoff of ≥1.5 (28,29), as the commonly used cutoff in various studies detecting genes in aging mice compared with young mice. A t-test was performed to identify genes with P-values <0.05. The resulting microarray data were registered to Gene Expression Omnibus (GSE274943).

RT-qPCR was used to validate the miRNAs in mouse serum. Total RNA and reverse transcriptase were incubated at 37˚C for 1 h for cDNA synthesis. Then, qPCR was performed using 10-fold diluted cDNA, TB Green Advantage qPCR Premix (cat. no. 639676, Takara Bio), and miRNA-specific primers (Table SI) based on the manufacturer's protocol. The conditions for qPCR were as follows: cDNA was denatured at 95˚C for 10 sec in a total volume of 25 µl, followed by 40 cycles of 95˚C for 5 sec and 60˚C for 20 sec. The cDNA was then incubated at 95˚C for 5 sec and 60˚C for 20 sec for 40 cycles. Cel-miR-39 was used as an external control. The quantitation of gene expression was done using the 2^-ΔΔCq method (30).

To evaluate gene expression in senescence-induced HUVECs, RNA-sequence data of HUVECs exposed to 4 Gy of radiation for 10 days were downloaded from the Sequence Read Archive (SRA) (GSE130727, SRR9016151-SRR 9016156). FASTQ files downloaded from SRA were quality-checked and trimmed using Fastp (31). Gene expression levels were calculated from FASTQ files using salmon (32) and tximport (33). R (version 4.3.1) was used for data analysis and quasi-likelihood F-tests were used with the expression analysis packages, limma (27) and edgeR (34). The p-values for multiple correction by the Benjamini-Hochberg method were less than 0.05. In addition, genes with a fold-change ≥1.5 are selected. Hierarchical clustering and principal component analysis (PCA) were done to examine differences in gene expression between the 4 Gy-irradiated and unirradiated HUVECs. Gene Set Enrichment Analysis (GSEA) and Gene Ontology (GO) analysis were performed using ClusterProfiler (35), fgsea, and the AnnotationDbi package. Kyoto Encyclopedia of Genes and Genomes (KEGG) and pathway analysis using ReactomePA were also conducted (36).

HUVECs, single donor P1 (cat. no. C-12200, Promo Cell) were purchased from Takara Bio, and cultured in Basic Cell Growth Medium 2 Kit (cat. no. C-22011, Promo Cell) with 5% Fetal Calf Serum (FCS), 5.0 ng/ml Epidermal Growth Factor (recombinant human), 10 ng/ml Basic Fibroblast Growth Factor (recombinant human), 10 ng/ml Insulin-like Growth Factor (Long-type), 20 ng/ml Insulin-like Growth Factor (Long R3 IGF, recombinant human), 1.0 µg/ml ascorbic acid, 22.5 µg/ml heparin, and 0.2 µg/ml hydrocortisone. For the recovery of EVs, exosome-depleted fetal bovine serum (cat. no. 558-39501, Fujifilm Wako) was used instead of FCS in the above medium. 41220, Promo Cell) to pass the HUVECs.

HUVECs from P2 to P6 were seeded at 1x10⁵ cells/60 mm dish and incubated overnight at 37˚C in a 5% CO₂ atmosphere. HUVECs were irradiated with 4 Gy X-rays (MBR-1520R-3 X-ray machine, Hitachi Ltd.) at a dose rate of 1.0 Gy/min (150 kVp, 20 mA, 0.5 mm aluminum, and 0.3 mm copper filters).

Total RNA was extracted from HUVECs that were cultured for 10 days after 4 Gy irradiation. The control group consisted of HUVECs that were passaged on day 5 and cultured for 10 days to avoid becoming 100% confluent. Intracellular RNA was extracted from cell pellets after centrifuging at 300 g for 3 min using the miRNeasy mini kit (cat. no. 217004, Qiagen) based on the manufacturer's protocol. The concentration of the extracted RNA was measured using a Nanodrop instrument (ThermoFisher Scientific) based on the manufacturer's protocol.

The HUVEC culture supernatant was filtered through a 0.20 µm filter and RNAs from the EVs were extracted from 2 ml of culture supernatant using the exoRNeasy midi kit (cat. no. 77144, Qiagen). The concentration of RNAs in the extracted EVs was determined using the Qubit™ microRNA Assay Kit (cat. no. Q32880, ThermoFisher Scientific) and Qubit 4 Fluorometer (cat. no. Q33238, ThermoFisher Scientific) based on the manufacturer's protocol.

RT-qPCR was performed to confirm the expression change of miR-21-5p in HUVECs cultured for 10 days after 4 Gy irradiation. Total RNA was reverse-transcribed into cDNA at 37˚C for 1 h. qPCR was performed with 10-fold diluted cDNA using the TB Green Advantage qPCR Premix (cat. no. 639676, Takara Bio) and miRNA-specific primers (Table SI) based on the manufacturer's protocol. U6 small nuclear RNA was used as an internal control. In the Mir-X miRNA First-Strand Synthesis kit (cat. no. 638313, Takara Bio) containing the U6 primer, the U6 primer sequence is a trade secret and not disclosed.

RT-qPCR was used to validate the messenger RNAs (mRNAs) of the candidate miR-21-5p target genes and genes that were downregulated by RNA-sequencing reanalysis in HUVECs cultured for 10 days following 4 Gy irradiation. Candidate target genes were downloaded from TargetScan (https://www.targetscan.org/vert_72/). cDNA synthesis was done using 200 ng of total RNA extracted from HUVECs cultured for 10 days after 4 Gy irradiation with the High Capacity cDNA Reverse Transcriptase Kit (cat. no. 4368814, ThermoFisher Scientific) based on the manufacturer's protocol. Total RNA (200 ng) and 2x Reverse Transcriptase Master Mix solution were mixed in a total volume of 20 µl. The reverse transcription reaction was performed at 25˚C for 10 min, 37˚C for 120 min, and 85˚C for 5 min. Next, qPCR was done using 5-fold diluted cDNA, Power SYBR Green PCR Master Mix (2x) (cat. no. 4367659, ThermoFisher Scientific), and gene-specific primer pairs (Table I) based on the manufacturer's protocol. Of the primers used, primer 3 (version 4.1.0) was designed for RT-qPCR of mRNA. Actin beta was used as an internal control.

RT-qPCR was performed for changes in the expression of miR-21-5p in EVs in culture supernatants secreted from HUVECs cultured for 10 days after 4 Gy irradiation. Total RNA (100 µg) was extracted from EVs in the culture supernatants secreted from HUVECs 10 days after 4 Gy irradiation. The Mir-X miRNA First-Strand Synthesis kit (cat. no. 638313, Takara Bio) was used for cDNA synthesis based on the manufacturer's protocol. Total RNA and reverse transcriptase were incubated at 37˚C for 1 h. Then, qPCR was performed using 10-fold diluted cDNA using TB Green Advantage qPCR Premix (cat. no. 639676, Takara Bio) and miRNA-specific primers (Table SI) based on the manufacturer's protocol. The respective miRNA sequences published on miRbase were used as primer sequences. Cel-miR-39 was used as an external control. The quantitation of gene expression was done using the 2^-ΔΔCq method (30).

To detect cell senescence in HUVECs cultured for 10 days following 4 Gy irradiation, the senescence β-galactosidase staining kit (cat. no. 9860, Cell Signaling Technology) was used. HUVECs (1x10⁵ cells) were treated with 1x Fixative Solution. After washing three times with phosphate-buffered saline (PBS), 1.0 ml of β-galactosidase stain was added and incubated at 37˚C overnight. HUVECs were observed under an optical microscope with the β-galactosidase staining solution remaining.

To quantify cellular senescence in HUVECs, the cellular senescence detection kit SPiDER-βGal (cat. no. SG02, Dojindo Molecular Technologies, Inc.) was used. HUVECs (1x10⁵ cells) were collected and incubated with Hank's Balanced Salt Solution (HBSS). After washing twice with HBSS, 1.0 ml of the prepared SPiDER-βGal working solution was added and incubated in a 5% CO₂ incubator at 37˚C for 1 h. After washing twice with HBSS, 1.0 ml of the prepared SPiDER-βGal working solution was added and incubated at 37˚C in a 5% CO₂ incubator for 30 min. The supernatant was removed by aspiration, the cells were washed twice with HBSS, and detached with the DetachKit (cat. No. C-41220, Promo Cell). The fluorescence of the detached HUVECs was analyzed using CytoFLEX flow cytometer and CytExpert software version 2.4 (both Beckman Coulter) and kaluza2.2 software (Beckman Coulter).

To confirm the cell cycle of 4 Gy-irradiated and nonirradiated HUVECs, each exfoliated cell was fixed by adding 70% ethanol and incubated at -20˚C overnight or longer. Then, each exfoliated cell was washed three times with PBS, and 200 µl of 250 µg/ml RNase A (cat. no. 318-06391, Fujifilm Wako) was added and incubated at room temperature for 30 min. Propidium iodide was then added at 50 µg/ml and fluorescence was detected with a CytoFLEX flow cytometer and CytExpert software version 2.4 (both Beckman Coulter). All events were measured up to 10,000 events per sample. For data analysis, histograms were generated using kaluza2.2 software (Beckman Coulter).

HUVEC culture supernatant (2 ml) was centrifuged at 300 g for 3 min. The collected supernatant was passed through a 0.20 µm filter and enriched for EVs by ultrafiltration using an Amicon Ultra 0.5 ml centrifugal filter (cat. No. UFC510096, Merke Millipore). EVs were then collected using the PS capture™ Exosome Flow Cytometry Kit (cat. no. 297-79701, Fujifilm Wako) based on the manufacturer's recommended protocol. Fluorescein isothiocyanate (FITC) anti-human CD63 Antibody (cat. no. 353005 BioLegend), FITC anti-human CD9 Antibody (cat. no. 312103, BioLegend), FITC anti-human CD81 Antibody (cat. no. 349503, BioLegend), and FITC Mouse IgG1, κ isotype Ctrl antibody (cat. no. 400110, BioLegend) were used. The data were analyzed using a CytoFLEX flow cytometer and CytExpert software version 2.4 (both Beckman Coulter). For data analysis, histograms were generated using kaluza2.2 software (Beckman Coulter).

All statistical analyses were performed using R (version 4.3.1). Two-group tests were subject to the Shapiro-Wilk test to confirm normality, followed by the F test, and Welch's T-test for data without equal variance. Mann-Whitney's U test was performed if normality was rejected. The significance level was set at P<0.05. For all experiments using HUVECs, the number of samples was n=4.

In our previous study, we found that cognitive function was impaired in 58-week-old mice in the Morris water maze (28). To further confirm changes in the expression of miRNAs in blood EVs in aged mice, we compared EVs RNAs from ‘Aging’ (82 and 102 weeks old) and ‘Young’ (8 weeks old) mice. The size of the EV RNA in the serum of the Aging and Young mice was found to be in the range of 25-200 nucleotides (nt) (Fig. 1A). We performed a mouse miRNA microarray using the extracted serum RNAs to identify miRNAs whose expression was significantly altered in Aging mice compared with Young mice. We found that mmu-miR-21a-5p, mmu-miR-3473b, and mmu-miR-7047-5p were significantly increased in the serum of the Aging mice (Fig. 1B). RT-qPCR was used to confirm the expression of the three miRNAs that were significantly increased. We found that mmu-miR-21a-5p and mmu-miR-7047-5p were significantly increased in Aging mice (Fig. 1C). This suggests that these two blood miRNAs are molecules associated with aging. One possible reason for the discrepancy between PCR results and microarray results is that microarrays and RT-qPCR have different sensitivities, so a sample that is statistically significant by microarray may not be statistically significant by RT-qPCR.

The altered miRNAs described above may be derived from the senescence of vascular endothelial cells (22,23). Therefore, we considered the possibility that the senescence of vascular endothelial cells affects the expression of miRNAs in blood EVs. To confirm the gene expression changes in senescence-induced HUVECs, we downloaded and reanalyzed the RNA-sequence data (GSE130727, SRR9016151-SRR9016156) of 4 Gy-exposed HUVECs. Cluster analysis revealed that gene expression was different between 4 Gy-irradiated and nonirradiated HUVECs (Fig. 2A). Similarly, PCA showed a significant change in gene expression between the 4 Gy-irradiated and nonirradiated HUVECs (Fig. 2B). Furthermore, differentially expressed gene (DEG) analysis using edgeR identified 1881 genes that were altered by ≥1.5-fold (Fig. 2C). This reanalysis indicated that 4 Gy irradiation alters the expression of many HUVEC genes.

Enrichment analysis was performed on 1881 genes that were altered by DEG analysis of the RNA sequences. A GSEA analysis was performed on the GO terms to determine whether senescence induction of HUVECs by 4 Gy irradiation was affected by a decrease or increase of the altered genes. The results indicated that most of the variable genes were enriched in the downregulated gene group because the enrichment score was less than zero (Fig. 3A). Based on this result, GO enrichment analysis was performed on the downregulated genes, and terms associated with DNA replication and cell division were identified (Fig. 3B). In addition, KEGG pathway and Reactome pathway enrichment analyses were also performed and pathways related to the cell cycle and DNA repair were annotated (Fig. 3C and D). These results suggest that there are changes in DNA replication and cell cycle-related genes in 4 Gy-irradiated senescence-induced HUVECs.

We identified mmu-miR-21a-5p and mmu-miR-7047-5p as miRNAs that are significantly up-regulated in Aging mice (Fig. 1); however, because no miRNA corresponding to mmu-miR-7047-5p was found in humans, we focused only on miR-21-5p in this study. The results of an RNA-sequence enrichment analysis showed that the expression of 435 genes was decreased by 4 Gy irradiation. The common genes among the 378 candidate target genes of miR-21-5p were examined by target scan, and cell division cycle 25A (CDC25A), mutS homolog 2 (MSH2), methylthioadenosine phosphorylase (MTAP), methylenetetrahydrofolate dehydrogenase, cyclohydrolase and formyltetrahydrofolate synthetase 1 (MTHFD1), zinc finger protein 367 (ZNF367), and zinc finger protein 704 (ZNF704) were identified (Fig. 4A). Next, to induce senescence in HUVECs, the cells were cultured for 10 days after 4 Gy irradiation. Total RNA was extracted from the HUVECs for RT-qPCR of intracellular miR-21-5p and the candidate target genes. The results indicated that 4 Gy irradiation increased miR-21-5p expression in HUVECs and decreased the expression of five of the six target candidate mRNAs (Fig. 4B and C). Next, SA-β-GAL staining of 4 Gy-irradiated HUVECs revealed that the cytoplasm of 4 Gy-irradiated HUVECs was stained darker and bluer compared with that of nonirradiated HUVECs (Fig. 4D). To quantitate the staining results, HUVECs were fluorescently stained using the cell senescence detection kit SPiDER-βGal, and the fluorescence intensity was measured by flow cytometry. The average SA-β-GAL fluorescence intensity of nonirradiated HUVECs was approximately 59,000, whereas the average SA-β-GAL fluorescence intensity of 4 Gy-irradiated HUVECs was approximately 150,000, which is a statistically substantial increase (Figs. 4E and S1). Because the expression of CDC25A, which is involved in the cell cycle, was decreased, we analyzed the cell cycle of HUVECs cultured for 10 days after 4 Gy irradiation. There were fewer S-phase cells in the 4 Gy-irradiated group compared with that in the nonirradiated group (Figs. 4F and S2). This suggests that 4 Gy irradiation induces senescence in HUVECs and affects the S phase of the cell cycle through increased expression of miR-21-5p.

To identify tetraspanins, which are surface antigens of EVs secreted from HUVECs cultured for 10 days after 4 Gy irradiation, EVs were collected from 2 ml of culture supernatant using PS capture and incubated with specific antibodies for each FITC-labeled tetraspanin family member and fluorescence was detected by flow cytometry. The results indicated that CD9 was strongly expressed in EVs secreted from HUVECs, confirming that EVs were collected (Fig. 5A). Next, we measured the expression of miR-21-5p in the EVs. The miR-21-5p expression in the supernatant of HUVECs cultured for 10 days after 4 Gy irradiation considerably increased by 2.7-fold compared with the nonirradiated HUVECs (Fig. 5B). This indicates that miR-21-5p is increased not only present intracellularly, but also in EVs of HUVECs irradiated with 4 Gy.