The differential expression profiles of miRNA in serum-derived exosomes and its potential role in age-related hearing loss

Objectives: This study aimed to characterize serum exosomal miRNA profiles from patients with age-related hearing loss (ARHL) to identify key pathogenesis-related miRNAs for ARHL.

Methods: Peripheral venous blood samples were collected from patients with ARHL and elderly controls, and exosomes were isolated from serum of each subject. Then, the isolated exosomes were systematically identified by nanoparticle tracking analysis (NTA), transmission electron microscopy (TEM) and western blot. Subsequently, the isolated exosomes were submitted for miRNA sequencing and a series of bioinformatics analysis. Ultimately, four key DE-miRNAs, namely hsa-miR-100-5p, hsa-miR-23b-3p, hsa-miR-373-3p, and hsa-miR-27b-3p, were verified using quantitative real-time polymerase chain reaction (RT-qPCR).

Results: NTA, TEM and western blot confirmed exosomes were successfully isolated. After sequencing, 22 differential expressed miRNAs (6 up-regulation and 16 down-regulation) were identified between the exosomes from ARHL and controls, and then collectively identified 17,451 predicted target genes and 15,863 experimentally validated target genes. Gene Ontology enrichment analysis revealed that the target genes were significantly associated with “regulation of neuron projection development,” “sensory system development,” “proteasome-mediated ubiquitin-dependent protein catabolism,” and “ubiquitin-like protein ligase binding.” Kyoto Encyclopedia of Genes and Genomes (KEGG) showed the target genes were significantly enriched in “PI3K-Akt signaling pathway,” “MAPK signaling pathway,” “cellular senescence,” “autophagy,” “mTOR signaling pathway,” “ubiquitin-mediated proteolysis,” and “signaling pathways regulating stem cell pluripotency.” Additionally, the Reactome analysis highlighted the involvement of “MAPK family signaling cascades,” “negative regulation of the PI3K/AKT network,” and “antigen processing: ubiquitination and proteasome degradation.” Disease Ontology further demonstrated significant enrichment of target genes in neurological disorders. RT-qPCR showed hsa-miR-100-5p and hsa-miR-23b-3p exhibited markedly decreased levels, while hsa-miR-373-3p and hsa-miR-27b-3p were significantly up-regulated in ARHL, which were consistent with sequencing results, confirming a high relatively reliability of the sequencing results.

Conclusion: Ubiquitination modification, autophagy process, cellular senescence and nervous system regulation may jointly contribute to the core molecular mechanism of ARHL. The hsa-miR-100-5p, hsa-miR-23b-3p, hsa-miR-373-3p, and hsa-miR-27b-3p may preliminarily act as key regulatory factors to participate in the pathophysiological process of ARHL, providing exploratory evidence for their potential application value as molecular markers.

1 Introduction

Age-related hearing loss (ARHL), also known as presbycusis, is a disease associated with aging that concurrently impacts both the peripheral auditory system and the central auditory pathways (Mussoi and Brown, 2019). This condition exhibits a high prevalence globally. By 2024, the number of people aged 65 and above will exceed 240 million, accounting for approximately 16.5% of the total population, among which two-thirds have hearing loss (Ninoyu and Friedman, 2024). ARHL is not only characterized by high-frequency hearing loss and impaired speech recognition ability (Dubno et al., 2008), but is also closely related to cognitive decline and increased risk of dementia. Timely hearing aid intervention can effectively delay the deterioration of cognitive function (Lee, 2015). With the acceleration of the global aging population, hearing impairment characterized by both peripheral and central lesions has transcended a mere sensory function defect. It is progressively emerging as a significant public health concern that exacerbates the burden on an aging society. Consequently, a thorough investigation into its early warning mechanisms and effective intervention strategies is of considerable social value and practical importance.

As a degenerative disease with a complex pathogenesis and limited clinical treatment options, ARHL still lacks reliable biomarkers for early diagnosis and intervention at present. Although emerging evidence suggests that Apela/Toddler peptide may serve as a potential biomarker for age-related disorders (Monastero et al., 2023), and the auditory brainstem gap response has been proposed as a novel physiological indicator for ARHL (Williamson et al., 2015), these findings remain preliminary. With the accelerating global aging population, the prevalence of ARHL is projected to rise substantially, posing dual challenges to healthcare systems: On the one hand, there is an urgent need to develop functional and molecular biological markers capable of identifying preclinical reversible lesions. On the other hand, effective early warning intervention programs need to be established to address this increasingly severe public health issue.

MicroRNA (miRNA) is a type of endogenous small non-coding RNA approximately 20–24 nucleotides in length. It regulates gene expression at the post-transcriptional level by binding to the 3′ untranslated region (3′-UTR) of the target gene. In the cochlea of aged C57BL/6J mice, activated miR-34a impairs autophagic flux through ATG9A suppression, ultimately promoting cochlear cell death (Pang et al., 2017). Similarly, miR-204-5p regulates SIRT1 to promote endoplasmic reticulum stress-induced apoptosis of inner ear hair cells in aged mice (Hu et al., 2024). Our previous work demonstrated that miR-130b-3p delays cellular senescence by enhancing autophagy in sensory cells via targeting PPARγ (Zhang et al., 2024). These findings collectively underscore the pivotal regulatory functions of miRNAs in ARHL occurrence.

In recent years, exosome secretion and its role in diseases have received increasing attention. Exosomes are membrane vesicles with a diameter of 40–100 nm, encapsulated by a lipid bilayer. They can secrete circulating miRNA, and the miRNAs in exosomes are more stable during the circulation process, thus having greater potential value as biomarkers (Nik Mohamed Kamal and Shahidan, 2020; Redis et al., 2012). In 2018, Wong et al successfully extracted exosomes from the inner ear of Wistar rats for the first time (Wong et al., 2018). Research has found that exosomes derived from neural progenitor cells prevent hearing loss caused by ischemia-reperfusion injury in mice by inhibiting cochlear hair cell inflammation (Hao et al., 2022). Mesenchymal stem cell-derived exosomes increase the autophagic activity of hair cells, promote cell survival, reduce mitochondrial oxidative stress levels and the apoptosis rate of hair cells, thereby improving neomycin-induced ototoxicity (Liu et al., 2024b). However, there are relatively few studies on the relationship between exosomes and ARHL.

This study employed high-throughput sequencing technology to systematically analyze the expression profiles of miRNAs in serum exosomes of patients with ARHL, aiming to screen out key miRNAs closely related to the pathogenesis of ARHL and evaluate their potential value as biomarkers for disease diagnosis. The findings will provide novel insights into the molecular mechanisms underlying ARHL and may yield promising molecular targets for early clinical diagnosis and intervention.

2 Materials and methods

2.1 Ethical review

This study was approved by the Medical Ethics Committee of Chongqing General Hospital (approval no. KY S2023-044-01). Prior to sample collection, written informed consent was obtained from all participants, including 6 ARHL patients and 6 age-matched healthy controls recruited for this study. The clinical information of enrolled participants is summarized in Table 1 and Supplementary Figures 1, 2.

TABLE 1

Table 1. The general condition and hearing information of the subjects.

2.2 Collection of subjects

The study participants were divided into two groups: ARHL group and an elderly control group. According to the World Health Organization (WHO) diagnostic criteria for ARHL (World Health Organization, 2021), pure-tone audiometry was conducted by a certified audiologist in a professional soundproof booth to ensure the accuracy of the hearing thresholds. Participants in the ARHL group were required to have a mean hearing threshold > 35 dB HL at four frequencies (0.5, 1, 2, and 4 kHz), while those in the control group were required to maintain thresholds < 20 dB HL at the same frequencies.

Inclusion criteria: (1) over 60 years old, (2) be capable of completing basic daily living activities, such as taking a bath, dressing, grooming and using the toilet; And the ability to walk independently for 400 m, (3) right-handed, (4) there was no previous history of ear trauma or surgery, and no other cranial nerve injuries except for cranial nerve VIII, (5) have not used a hearing aid either currently or before, (6) sign the informed consent form and be willing to participate in the research.

Exclusion criteria: (1) Meniere’s disease, herpes zoster oticus, noise-induced deafness, exposure to ototoxic drugs, conductive hearing loss, and other ear diseases with known causes. (2) Meningitis, cognitive impairment, metabolic diseases, cardiovascular diseases, autoimmune diseases. (3) Previous or current mental or neurological disorders, addiction to psychoactive substances or sedatives. (4) History of head trauma.

2.3 Serum sampling

The peripheral venous blood of six ARHL patients and six elderly controls were collected, and the serum was obtained by collecting supernatant after centrifugation at 1,900 × g for 10 min, followed by 13,000 × g for 2 min at 4 °C. Among them, three of them were used for sequencing, and the remaining three samples were employed for expression validation using real-time quantification PCR (RT-qPCR) (Table 1).

2.4 Isolation of serum-derived exosomes

Exosomes were isolated from serum of ARHL patients and healthy controls at 4°C using the ultra-centrifugation method. Briefly, the collected serum samples were thawed on ice, and then subjected to sequential centrifugation: initial clarification at 500 × g for 5 min, followed by transfer of the supernatant to fresh 50 mL tubes for centrifugation at 2,000 × g for 30 min. The resulting supernatant was further processed by centrifugation at 10,000 × g for 60 min. The clarified supernatant was then filtered through 0.22 μm sterile filters and ultracentrifuged at 120,000 × g for 70 min. After careful removal of the supernatant, the exosome-containing pellet was resuspended in sterile PBS for downstream applications.

2.5 Characterization of the isolated serum-derived exosomes

The morphology, particle size and expression of surface markers of the isolated serum-derived exosomes were characterized by transmission electron microscope (TEM), NanoSight NS 300 system (NTA), and western blot. For TEM, the specific operation was as follows: add 20 μL of the prepared exosome suspension to a copper mesh for natural adsorption for 5–10 min. Then, use filter paper to absorb the excess liquid and let it dry. Subsequently, 20 μL of 2% phosphotungstic acid solution was added for staining for 3–5 min. The excess staining solution was then aspirated and dried under an incandescent lamp. Finally, the sample images were observed and captured using the JEM1400 transmission electron microscope from JEOL Company. For NTA, the sample was diluted with sterile PBS and filtered through a 0.22 μm filter to adjust the particle concentration to 10⁸–10⁹ particles /mL. According to the instructions, use a syringe to draw 1 mL of the sample and slowly inject it into the sample cell. After installing the laser module and inserting the thermometer probe, start the camera in the software, adjust the focal length and parameters, select the measurement mode and then run the test. Finally, the quantity and size of exosomes were detected using the NTA (NanoSight Technology, Malvern, United Kingdom), which is equipped with a green laser and a CMOS camera. For western blot, the exosomes were mixed with the sample loading buffer and then placed in a water bath at 100°C for 5–6 min. Then, the proteins were incubated with anti-CD9 antibody (1:1,000), anti-TSG101 antibody (1:1,000), anti-HSP70 antibody (1:1,000), and ani-GAPDH antibody (1:10,000) at 4°C overnight, followed by goat anti-rabbit IgG (H+L)-HRP (1:5,000) and goat anti-mouse IgG (H+L) -HRP (1:10,000) at 37°C for 1 h. After rinsing with 1 × PBST for 5 min for three times, protein bands were visualized using a chemiluminescence development.

2.6 Exosomal miRNA sequencing

Total RNA was extracted from the isolated serum-derived exosomes using mirVANA miRNA Isolation kit (Thermo Fisher Scientific, United States) following the manufacturer’s instructions. The concentrations and quality of the isolated RNA were determined using a NanoDrop 2000 (Thermo Fisher Scientific, MA, United States). RNA samples were commissioned to Xiamen Life Interconnection Technology Co., Ltd (Hangzhou, China) for miRNA sequencing (n = 3). After constructing the cDNA library using the TruSeq small RNA kit (Illumina, San Diego, United States), PE150 sequencing was performed on the Illumina NovaSeq 6000 platform. The original data was evaluated for quality by the fastqc software (version 0.11.2), which was used to remove the N base, filter out the low-quality sequence (q20), and cut off the connector. Subsequently, taking the GRCh38 genome (Ensembl release 102) as a reference, non-coding RNAs such as rRNA/tRNA were removed by bowtie alignment with the Rfam library. Finally, miRDeep2 (version 0.1.3) was used to identify the expression levels of miRNAs in all the samples.

2.7 Identification of DE-miRNAs and functional analyses

The expression levels were homogenized and differentially expressed using DESeq2 (version 3.21) in R 3.4.1. This project adopted the standards of false discovery rate (FDR) (the Benjamini and Hochberg method) < 0.05 and | log₂ fold change (FC)| = 1 to screen differentially expressed miRNAs (DE-miRNAs). After that, the screened DE-miRNAs between the two groups were used to search their target genes using online tools, including predicted target genes, and experimentally verified target genes. Afterward, the predicted target genes, and experimentally verified target genes of DE-miRNAs were both submitted for functional analyses, containing gene ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), Reactome Pathway Databas (Reactome), and Disease Ontology (DO). Finally, the DAVID 6.8 Bioinformatics Resources was used to visualized these pathways, and FDR < 0.05 was considered as the statistical significance.

2.8 RT-qPCR

The expression levels of the selected DE-miRNAs were determined using the stem-loop method. Briefly, the other three isolated total RNA were reverse transcribed into cDNA using the PrimeScript™ II first-strand cDNA synthesis kit (Taraka, Japan) in line with the manufacturer’s instructions. The cDNA synthesis reaction was performed in a 10 μL system containing 1 μL RT-Primer, 1 μL dNTP mixture, 500 ng RNA, and RNase-free H₂O. The mixture was incubated at 65°C for 5 min, followed by the addition of 4 μL 5 × PrimeScript II Buffer, 0.5 μL RNase Inhibitor, 1 μL PrimeScript II RTase, and 4.5 μL RNase-free H₂O. The obtained mixture was incubated at 42° for 60 min and then at 95° for 5 min in sequence. Finally, qPCR amplification was carried out using Power SYBR Green PCR Master Mix (Thermo Fisher Scientific) following the supplier’s protocols. The sequences of all primers are shown in Table 2, and U6 was used as the internal reference gene. The relative expression levels of hsa-miR-100-5p, hsa-miR-23b-3p, hsa-miR-373-3p, and hsa-miR-27b-3p were calculated by the 2^{–Δ Δ Ct} method.

TABLE 2

Table 2. Details of PCR primers utilized in this investigation.

2.9 Statistical analysis

Data were described as mean ± standard deviation (SD) with triple repetition, and statistical analysis was conducted using GraphPad Prism 5 (Graphpad Software, San Diego, CA). For comparison for two groups, the student’s t-test was utilized to evaluate the differences. Statistical significance was defined as P < 0.05.

3 Results

3.1 Exosomes successfully isolated from serum of ARHL patients and healthy controls

In order to explore the exosomal mechanisms of ARHL, exosomes were isolated from serum samples of ARHL patients and healthy controls, and then were characterized by TEM, NTA and western blot. NTA results revealed that the major peaks in particle size of the control serum-derived exosomes and ARHL serum-derived exosomes were respectively about 111.4 and 103.7 nm; and the mean peaks of the control serum-derived exosomes and ARHL serum-derived exosomes were approximately 142 and 158.4 nm (Figure 1A), which was consistent with the previously reported exosomes size distributions (Hade et al., 2021; Isaac et al., 2021). Then, TEM showed that exosomes isolated from control and ARHL serum were both spherical or disc-shaped with a diameter of about 150 nm, and had a bilayer membrane structure (Figure 1B). Furthermore, western blot analysis confirmed the presence of exosomal marker proteins TSG101, HSP70, and CD9 in the isolated substances, while these markers were absent in the supernatant (Figure 1C). These results indicated that exosomes were successfully isolated from the serum of the ARHL patients and healthy controls with this ultracentrifugation method.

FIGURE 1

Figure 1. Identification of exosomes isolated from serum of control individuals and ARHL patients. (A) Particle distribution and an average diameter of exosomes measured by NTA. (B) TEM revealed spherical or disc-shaped of exosomes with a double-membrane structure. (C) Western blot showed TSG101, HSP70 and CD9 were all expressed. Control, exosomes isolated from control individuals; ARHL, exosomes isolated from ARHL patients; Control-supernatant, supernatant from control individuals; ARHL-supernatant, supernatant from ARHL patients.

3.2 Quality control of sequence reads and annotation of miRNAs

Exosomal RNA sequencing requires a minimum input of over 1 ng RNA for library construction, with a total library quantity exceeding 10 ng being adequate for subsequent sequencing. The RNA concentrations isolated from each sample ranged from 0.19 to 0.28 ng/μL, as well as the RNA amount were 5.13–7.56 ng (Table 3). The library concentrations of each sample were 4.99–9.7 ng/μL, as well as the library amount ranged from 69.86 to 135.8 ng (Table 3). These results indicated that the quantities of extracted RNA satisfied these criteria for both library preparation and sequencing. Additionally, Qsep100 analysis verified the integrity of the isolated exosomal RNA and ruled out potential cellular contamination (Figure 2A).

TABLE 3

Table 3. The RNA concentration of exosomes constructed by the library, and the quality of the constructed library.

FIGURE 2

Figure 2. Quality control of sequence reads and identification of microRNAs (miRNAs). (A) The results of exosomal RNA Qsep100. The horizontal axis represents the length of RNA, and the vertical axis represents the fluorescence intensity of the RNA sample. (B) Distribution of sequencing quality. (C) Statistics of miRNA length expressed.

Quality control metrics confirmed that each sample generated more than 10 million clean reads (Table 4). After aligning with the Rfam database using Bowtie, we identified and filtered out rRNA and other non-coding RNAs (ncRNAs), with each sample containing fewer than 500,000 ncRNA reads (Table 5). Subsequent alignment to the reference genome showed that all samples had mapping rates above 25% (Table 6), confirming the high quality of the sequencing data, which is suitable for further analysis.

TABLE 4

Table 4. Quality assessment of sample sequencing output data.

TABLE 5

Table 5. Statistics of Rfam database comparison.

TABLE 6

Table 6. Statistics of read alignment rates after Rfam library filtration.

In addition, the overall base quality of the sequencing data met the Q30 standard (Figure 2B). Read length distribution analysis indicated that most reads were 20–24 nt in length, with a prominent peak at 22 nt (Figure 2C).

3.3 16 DE-miRNAs identified between the exosomes from controls and ALHR samples

According to the thresholds of log₂FC = 1 and FDR < 0.05, a total of 22 DE-miRNAs were identified in the ALHR and elderly control samples, among which 6 were upregulated (hsa-miR-323a-3p, hsa-miR-6877-5p, hsa-miR-224-5p, hsa-miR-27b-3p, hsa-miR-181a-3p, and hsa-miR-373-3p) and 16 were downregulated (hsa-miR-23b-3p, hsa-miR-378a-3p, hsa-miR-100-5p, hsa-miR-194-5p (precursors are hsa-mir-194-1 and hsa-mir-194-2, respectively), hsa-miR-19b-3p, hsa-miR-1273h-5p, hsa-miR-1304-3p, hsa-mir-146b-5p, hsa-miR-145-5p, hsa-miR-590-3p, hsa-miR-339-5p, hsa-miR-431-5p, hsa-miR-196b-5p, hsa-miR-204-3p, hsa-miR-942-5p) (Figure 3A). The heatmap of the identified DE-miRNAs revealed that the identified DE-miRNAs could clearly distinguish the ALHR samples from the controls (Figure 3B). The bar chart provided a more intuitive visualization of the differential expression changes for each DE-miRNA (Figure 3C). Online searches for the target genes of DE mature miRNAs revealed that these DE-miRNAs collectively identified 17,451 predicted target genes and 15,863 experimentally validated target genes (Figure 3D).

FIGURE 3

Figure 3. Screening of differentially expressed miRNAs (DE-miRNAs). (A) The expression volcano diagram of DE-miRNAs. (B) The expression heatmap of DE-miRNAs. (C) The expression bar chart of DE-miRNAs (To the right is an upward adjustment, and to the left is a downward adjustment). (D) Venn diagram of the intersection of the predicted target genes and the experimentally verified target genes.

3.4 Enrichment analyses of the target genes

The predicted and experimentally verified target genes of the identified DE-miRNAs were respectively subjected to enrichment analyses, including GO, KEGG, Reactome and DO. In addition to enriching analyses of cytological functions and signaling pathway functions, we performed DO enrichment analysis on the target genes. This strategy sought to explore the potential involvement of DE-miRNAs in diverse diseases via the regulation of their target genes, and to examine whether common molecular mechanisms or pathways exist across these diseases.

GO enrichment analysis of the predicted target genes showed significant enrichment in several critical biological processes and cellular functions, including “synapse organization,” “small GTPase-mediated signal transduction,” “modulation of chemical synaptic transmission,” “axonogenesis,” “cell junction assembly,” “regulation of neuron projection development,” “forebrain,” “glutamatergic synapse,” “sensory system development,” “neuron to neuron synapse,” and “dendrite development” (Figure 4A). Furthermore, the verified target genes of the identified DE-miRNAs were significantly enriched in “DNA-binding transcription factor binding,” “nuclear envelope,” “mitotic cell cycle phase transition,” “regulation of protein kinase activity,” “cell-substrate junction,” “viral process,” “proteasome-mediated ubiquitin-dependent protein catabolic,” “chromosomal region,” “regulation of protein catabolic process,” and “ubiquitin-like protein ligase binding” (Figure 4B).

FIGURE 4

Figure 4. GO and KEGG functional analyses of the screened target genes of DE-miRNAs. (A) Top 20 significantly GO terms enriched by the predicted target genes of DE-miRNAs. (B) Top 20 significantly GO terms enriched by the verified target genes of DE-miRNAs. (C) Top 20 significantly KEGG terms enriched by the predicted target genes of DE-miRNAs. (D) Top 20 significantly KEGG terms enriched by the verified target genes of DE-miRNAs.

The KEGG functional enrichment analysis of the target genes showed that the predicted target genes were significantly enriched in several pathways, such as “PI3K-Akt signaling pathway,” “human papillomavirus infection,” “MAPK signaling” “pathway,” “endocytosis,” “regulation of actin cytoskeleton,” “Rap1 signaling pathway,” “axon guidance,” “cellular senescence,” “Hippo signaling pathway,” “signaling pathways regulating pluripotency of stem cells,” “ErbB signaling pathway,” and “phosphatidylinositol signaling pathway” (Figure 4C). In addition, the verified target genes exhibited significant enrichment in pathways related to “pathways of neurodegeneration—multiple disease,” “endocytosis,” “protein processing in endoplasmic reticulum,” “autophagy-animal,” “cellular senescence,” “mTOR signaling pathway,” “ubiquitin mediated proteolysis,” “signaling pathway regulating pluripotency of stem cells,” “cell cycle,” and “adherens junction” (Figure 4D).

Reactome pathway enrichment analysis demonstrated significant enrichment of the predicted target genes in key biomolecular pathways including “diseases of signal transduction by growth factor receptor and second messenger,” “RHO GTPase cycle,” “MAPK family signaling cascade,” “neuronal system,” “MAPK1/MAPK3 signaling,” “RAF/MAP kinase cascade,” “PIP3 activates AKT signaling,” “signaling by NTRKs,” “negative regulation of the PI3K/AKT network,” “signaling by TGF-β receptor complex,” and “MITF-M-dependent gene expression”(Figure 5A). The verified target genes were significantly enriched in “diseases of signal transduction by growth factor receptor and second messengers,” “M Phase,” “Class I MHC mediated antigen processing and presentation,” “transcriptional regulation by TP53,” “RHO GTPase effectors,” “antigen processing: ubiquitination and proteasome degradation,” “signaling by TGFB family members,” “intracellular signaling by second messengers,” “signaling by TGFB family members,” “TGF-beta receptor signaling activates SMADs,” and “SUMOylation” (Figure 5B).

FIGURE 5

Figure 5. Reactome and disease ontology (DO) functional analyses of the screened target genes of DE-miRNAs. (A) Top 20 significantly Reactome terms enriched by the predicted target genes of DE-miRNAs. (B) Top 20 significantly Reactome terms enriched by the verified target genes of DE-miRNAs. (C) Top 20 significantly DO terms enriched by the predicted target genes of DE-miRNAs. (D) Top 20 significantly DO terms enriched by the verified target genes of DE-miRNAs.

In addition, the DO functional enrichment analysis of the target genes displayed that the predicted target genes were significantly associated with “intellectual disability,” “peripheral nervous system neoplasm,” “neuroblastoma,” “autonomic nervous system neoplasm,” “pervasive developmental disorder,” “autosomal dominant intellectual developmental disorder,” “physical disorder,” “autistic disorder,” “congenital nervous system abnormality,” and “developmental and epileptic encephalopathy” (Figure 5C). The verified target genes demonstrated significant enrichment in diseases or pathological states such as “organ system benign neoplasm,” “peripheral nervous system neoplasm,” “neuroblastoma,” “autonomic nervous system neoplasm,” and “malignant ovarian surface epithelial-stromal neoplasm” (Figure 5D).

3.5 Construction of two miRNA-target regulatory networks

Using stringent criteria, we systematically constructed two miRNA-target regulatory networks. First, we ranked the prediction scores from each database in descending order, filtered for high-confidence predictions, and built a DE-miRNA-predicted target gene regulatory network comprising 141 pairs of significant regulatory interactions (Figure 6A). In this DE-miRNA-predicted target gene regulatory network, hsa-miR-19b-3p and hsa-miR-500-3p were the hub nodes. Second, by strictly selecting interactions supported by experimental validation, support_type, and pubmed_id literature evidence, we constructed a highly reliable DE-miRNA-verified target gene regulatory network containing 189 interaction pairs (Figure 6B). In this DE-miRNA-verified target gene regulatory network, hsa-miR-1273h-5p, hsa-miR-500-3p, hsa-miR-224-5p, and hsa-miR-373-3p were the hub nodes.

FIGURE 6

Figure 6. The co-expression networks of DE-miRNAs and screened target genes, as well as experimental verification. (A) The co-expression networks of DE-miRNAs and predicted target genes. The rectangle and oval represent miRNAs and mRNAs, respectively. (B) The co-expression networks of DE-miRNAs and verified target genes. The rectangle and oval represent miRNAs and mRNAs, respectively. The color indicates the significant difference of each transcript. (C) Expression levels of exosomal hsa-miR-100-5p, hsa-miR-23b-3p, hsa-miR-373-3p, and hsa-miR-27b-3p in control and ARHL-derived exosomes. N = 3. Data were expressed mean ± standard deviation (SD).

3.6 Down-regulated hsa-miR-100-5p, hsa-miR-23b-3p, and up-regulated hsa-miR-373-3p, hsa-miR-27b-3p in ALHR exosomes using RT-qPCR

Four DE-miRNAs (hsa-miR-100-5p, hsa-miR-23b-3p, hsa-miR-373-3p, and hsa-miR-27b-3p) were selected for RT-qPCR analysis. It was found that compared with the serum-derived exosomes from the controls, the expression levels of hsa-miR-100-5p (P = 0.003) and hsa-miR-23b-3p (P = 0.0029) were significantly reduced, while the expression levels of hsa-miR-373-3p (P = 0.0112) and hsa-miR-27b-3p (P = 0.0048) were evidently elevated in the serum-derived exosomes from the ALHR patients (P < 0.05) (Figure 6C). The differential expression patterns verified by RT-qPCR were consistent with the sequencing result, which implied a high relatively reliability of the sequencing results.

4 Discussion

The pathogenesis of ARHL is complex and highly heterogeneous. Current clinical diagnosis and therapeutic evaluation primarily rely on a comprehensive assessment approach combining clinical symptoms with audiometric findings. There remains an urgent clinical need to develop functional and molecular biomarkers for detecting reversible preclinical pathological changes, as well as to establish effective intervention strategies to address these challenges in clinical management. ARHL is influenced by a series of external and internal factors, including long-term exposure to external noise (Otte et al., 1978), ototoxic medication use (Gratton et al., 1996), cochlear vascular pathologies and hemodynamic alterations (Gratton et al., 1997), neurotransmitter activity modifications (Sha et al., 2008), dietary habits (Gratton et al., 1997), oxidative stress (Ohlemiller et al., 2010), and mitochondrial gene deletions (Engle et al., 2013), methylation epigenetic modifications (Schuknecht and Gacek, 1993), etc. Despite these identified risk factors, the precise molecular mechanisms underlying ARHL pathogenesis remain poorly understood.

As an important regulatory factor for the occurrence and development of deafness, the role of miR-204 and its subtypes in the auditory system has been confirmed by multiple studies. Studies have shown that miR-204-5p is significantly upregulated in the cochlear tissues of aged mice, exerting a protective effect by inhibiting H₂O₂-induced apoptosis of HEI-OC1 cells and promoting cell proliferation (Hu et al., 2024). In the acute hearing loss model, miR-204-5p can interact with the transcription factor Sp1 to alleviate hypoxia or LPS-induced damage to HEI-OC1 cells (Xie et al., 2021). Furthermore, miR-204-5p modulates spiral ganglion neuron viability in a concentration-dependent manner through targeted regulation of TMPRSS3 expression, with its inhibition promoting neuronal functional recovery and ameliorating sensorineural hearing loss (Jiang et al., 2020; Li et al., 2014). Intriguingly, both miR-204-5p and miR-204-3p show significant downregulation in hippocampal tissues following noise-induced hearing loss (Ha et al., 2021). Our study reveals for the first time a marked reduction of exosomal miR-204-3p in serum samples from presbycusis patients, providing novel clues into the molecular mechanisms of ARHL, although the precise underlying mechanisms require further investigation.

Accumulating evidence highlights the crucial role of miR-431 in mammalian inner ear development and auditory function maintenance. During cochlear development, miR-431 exhibits high expression levels in spiral ganglion neurons of neonatal mice, with progressive downregulation throughout maturation. Notably, miR-431 overexpression leads to significant reductions in spiral ganglion neuron density and subsequent hearing impairment (Fan et al., 2016). In noise-induced hearing loss models, animals with miR-431 overexpression demonstrate more severe auditory dysfunction accompanied by greater loss of cochlear inner hair cell ribbon synapses compared to controls, suggesting miR-431 may enhance cochlear vulnerability to acoustic trauma (Fujita et al., 2023). Multivariate analysis further confirmed that miR-431 was independently associated with hearing loss at 6 months (Huang et al., 2025). Our current study reveals significantly decreased miR-431 expression in serum exosomes from elderly subjects, implying its potential involvement in the pathogenesis of ARHL.

In addition, our RT-qPCR validation confirmed that hsa-miR-100-5p and hsa-miR-23b-3p were significantly downregulated, while hsa-miR-373-3p and hsa-miR-27b-3p were upregulated in serum exosomes of ARHL patients, consistent with sequencing data. Hsa-miR-100-5p, down-regulated in ARHL-associated exosomes, is a known modulator of mTOR signaling (Liu et al., 2024a), a pathway previously implicated in ARHL via its regulation of cochlear hair cell autophagy and senescence (Ege et al., 2024). A previous study of Wang et al. (Wang and Liu, 2025) demonstrated that miRNA-634 was highly expressed in exosomes extracted from the peripheral blood samples of ARHL patients, and could promote apoptosis of HEI-OC1 cells by regulating the AKT/mTOR signaling pathways, thereby influencing the progression of ARHL. It has been discovered that the hsa-miR-23b-3p can prevent neuronal cell apoptosis and the production of reactive oxygen species (Zhan et al., 2016). Itokazu et al. demonstrated that the expression of miR-23b-3p in exosomes decreased with the increase of donor age, as well as TGFBR3 was identified as the key target of miR-23b-3p, and its inhibition led to the upregulation of multiple genes related to the stemness characteristics of mesenchymal stem cells (MSCs), suggesting exosomes delivered miR-23b-3p could serve as a biomarker for predicting the therapeutic effect of MSC treatment (Itokazu et al., 2025). The roles of hsa-miR-373-3p have been widely reported in many cancers, including cervical cancer, pancreatic carcinoma, and choriocarcinoma (Hu et al., 2018; Lu et al., 2021; Yu et al., 2021). It has been reported that high level of miR-373-3p could enhance lactate production by targeting MFN2, thus promoting aerobic glycolysis and proliferation of colon cancer cells (Wang et al., 2024a). In addition, PPAR signaling pathway is crucial for ARHL, and miR-27b-3p is predicted to regulate PPARG, which indicated miR-27b-3p and PPARG may be potential strategies for ARHL treatment (Zhang et al., 2019). Taken together, it can be inferred that hsa-miR-100-5p, hsa-miR-23b-3p, hsa-miR-373-3p, and hsa-miR-27b-3p may be potential crucial regulators of ARHL, due to related pathways critical for auditory homeostasis, neuroprotection, and senescence. Research has found that ubiquitinated proteins accumulate in the cochlea of aged mice, and endoplasmic reticulum stress impairs the function of clearing abnormal proteins through the ubiquitin-proteasome system (Wang et al., 2015). The deubiquitinating enzyme UCHL1 is down-regulated in H2O2-induced HEI-OC1, while its overexpression can enhance cell viability, alleviate oxidative damage, apoptosis and senescence (Li et al., 2023). Fbx2 is a ubiquitin ligase F-box protein that is highly expressed in Corti organs. Mice with targeted deletion of the Fbxo2 gene began to show age-related hearing loss at 2 months (Nelson et al., 2007). Our current findings further support the pivotal role of ubiquitination in ARHL. Bioinformatics analysis demonstrated that predicted target genes of DE-miRNAs were significantly enriched in key ubiquitin-related processes, including “proteasome-mediated ubiquitin-dependent protein catabolism” and “ubiquitin-like protein ligase binding” (GO terms), as well as the “ubiquitin mediated proteolysis” pathway (KEGG). Additionally, validated target genes showed enrichment in the Reactome pathway of “antigen processing: ubiquitination and proteasome degradation.” These consistent findings across multiple analytical approaches strongly implicate ubiquitination dysfunction as a critical mechanism underlying ARHL.

Cellular senescence has been implicated in the pathogenesis of various age-related degenerative disorders, including ARHL (Chen et al., 2025). In ARHL, cellular senescence manifests not only through morphological alterations such as progressive loss of afferent and efferent synapses, but also via functional declines characterized by reduced K⁺ currents and diminished nonlinear capacitance (Jeng et al., 2020). Oxidative stress-induced premature senescence of auditory cells represents a key contributor to ARHL development (Rivas-Chacón et al., 2021). Experimental studies using H₂O₂-induced senescence models in HEI-OC1 cells demonstrate that mitigation of mitochondrial oxidative stress significantly attenuates senescence-associated phenotypes and delays auditory cell aging (Rivas-Chacón et al., 2023). Our current findings further corroborate the pivotal role of cellular senescence signaling pathways in the occurrence of ARHL.

The changes in autophagy are closely related to the occurrence of presbycusis (Fujimoto et al., 2017). Oxidative stress impairs autophagic flux in auditory hair cells, leading to premature senescence characteristics including nuclear deformation and reduced proliferation capacity (Tsuchihashi et al., 2015). Neurotransmitters significantly inhibit D-galactose-induced hair cell loss by promoting autophagic influx (Wang et al., 2025). The dual leucine zipper kinase pathway may contribute to cochlear hair cell senescence and ARHL progression through aberrant activation of autophagy in hair cells (Ding et al., 2025). Overexpression of RONIN increased autophagy levels and weakened hair cell senescence induced by D-galactose (Wei et al., 2025). Our previous animal studies have substantiated the critical involvement of autophagy in ARHL development (Zhang et al., 2024). The current study identified significant enrichment of validated target genes from DE-miRNAs in autophagy-related KEGG pathways. Currently, studies related to human samples and the autophagy pathway are still relatively scarce.

Previous studies have found that PIN1 mediates the senescence of cochlear hair cells through the PI3K/Akt/mTOR signaling pathway (Zhang et al., 2021). Notably, 17β-estradiol exerts protective effects against ARHL by promoting angiogenesis via PI3K/AKT activation (Feng et al., 2021). Furthermore, rapamycin ameliorates presbycusis by enhancing autophagy in spiral ganglion neurons through mTOR inhibition (Liu et al., 2022). Based on our current research, it is suggested that PI3K/AKT and mTOR signals also play significant roles in ARHL development.

Our GO analysis identified “regulation of neuron projection development,” “sensory system development,” and “dendrite development” as top-enriched terms for DE-miRNA target genes, pathways critical for maintaining central auditory pathway integrity. ARHL is not merely a peripheral cochlear disorder; central auditory processing deficits (e.g., impaired speech recognition, delayed auditory reaction time) are increasingly recognized as key clinical features, driven by structural and functional disruptions in the auditory cortex, cochlear nucleus, and hippocampus (Jayakody et al., 2018; Sardone et al., 2019). In addition, the DO analysis also revealed significant enrichment of both predicted and validated target genes in nervous system-related categories, strongly suggesting an association between ARHL and neurological dysfunction. Existing research indicates that the pathological mechanism of presbycusis involves multi-level neurological functional abnormalities: in the peripheral auditory system, the loss of synapses between inner hair cells and spiral ganglion neurons is an early feature of cochlear aging and preceded the loss of neurons and hair cells (Wang et al., 2024b). In the auditory central system, the expressions of complement components (C1q, C3) and pro-inflammatory cytokines (TNF-α, IL-1β) in the cochlear nucleus were significantly upregulated, indicating that neuroinflammatory responses and complement system activation may be involved in the disease process (Cassinotti et al., 2022). At the neuroelectrophysiological level, the discharge frequency of central neurons and the intensity of excitatory postsynaptic currents show an age-dependent decrease under continuous acoustic stimulation (Wang et al., 2019). In higher auditory centers, ARHL patients exhibit not only an inverse correlation between hippocampal gray matter volume and hearing thresholds (Shim et al., 2023), but also altered spontaneous neural activity characterized by decreased activity in the superior temporal gyrus, parahippocampal gyrus, precuneus, and inferior parietal lobule, coupled with increased activity in the middle frontal gyrus, cuneus, and postcentral gyrus (Chen et al., 2018). Collectively, these data indicated that DE-miRNA-mediated disruptions in nervous system development neurological functions may contribute to both peripheral and central ARHL pathology. The clinical management of ARHL currently lacks non-invasive early diagnostic tools, relying primarily on pure-tone audiometry that only detects hearing loss post-irreversible cochlear damage, but our identification of serum exosomal miRNAs (hsa-miR-100-5p, hsa-miR-23b-3p, hsa-miR-373-3p, hsa-miR-27b-3p) offers a promising liquid biopsy solution with unique advantages for ARHL. Exosomal miRNAs are inherently stable in serum [protected by the lipid bilayer from RNase degradation (Katayama et al., 2025)], robust to pre-analytical variability (e.g., sample storage, processing time), a critical clinical biomarker requirement, and serum exosomes are easily obtainable via peripheral venous blood (unlike inaccessible cochlear tissue biopsies in living patients), enabling repeated sampling for longitudinal monitoring. This is particularly valuable for gradually progressive ARHL, as serial measurement of the four key miRNAs could detect early pathological changes (e.g., hsa-miR-100-5p downregulation preceding threshold shifts) before clinical symptoms emerge. Additionally, the four miRNAs show strong discriminatory power: our RT-qPCR data revealed significant expression differences between ARHL and control groups (all P < 0.01), and their combined signature may improve diagnostic accuracy. Finally, exosomal miRNAs may enable personalized therapeutic response monitoring. For example, if future studies validate mTOR inhibitors (e.g., rapamycin) as ARHL treatments (Li et al., 2025; Liu et al., 2022), increased hsa-miR-100-5p expression may act as a pharmacodynamic marker of efficacy, indicating restored mTOR regulation and autophagy. Therefore, these features position serum exosomal miRNAs as transformative tools for ARHL’s early diagnosis, prognosis, and therapeutic monitoring, addressing an unmet clinical need.

However, this study has several limitations that need to be acknowledged. First, the sample size for the core miRNA sequencing experiment was relatively small (n = 3 per group), which is a common constraint in initial high-throughput discovery studies but may increase the risk of false positive or false negative results. The small sample size limits the statistical power to detect subtle differences in miRNA expression and reduces the generalizability of our findings to the broader ARHL population. Although we validated four key DE-miRNAs using RT-qPCR in an additional set of samples (n = 3 per group), the overall sample size remains modest, and the results should be interpreted as preliminary. To address the limitation of small sample size, future studies should validate our findings in a larger, independent cohort (e.g., n ≥ 50 per group) with diverse demographic backgrounds (including different ages, genders, and regional origins) to improve the generalizability of the results. Additionally, functional experiments (such as in vitro cell models of auditory cell senescence and in vivo ARHL animal models) are needed to verify the regulatory roles of the four key DE-miRNAs (hsa-miR-100-5p, hsa-miR-23b-3p, hsa-miR-373-3p, and hsa-miR-27b-3p) in ARHL-related pathways (e.g., ubiquitination, autophagy, and cellular senescence). Longitudinal follow-up studies could further explore the prognostic value of these miRNAs in ARHL progression, providing more robust evidence for their clinical application as diagnostic or prognostic biomarkers.

5 Conclusion

The pathogenesis of ARHL is multifaceted. Our results suggest that ubiquitination modification, autophagy process, cellular senescence and nervous system regulation may jointly contribute to the core molecular mechanism of ARHL. The hsa-miR-100-5p, hsa-miR-23b-3p, hsa-miR-373-3p, and hsa-miR-27b-3p may preliminarily act as key regulatory factors to participate in the pathophysiological process of ARHL, providing exploratory evidence for their potential application value as molecular markers. Future validation in larger cohorts and functional studies is required to confirm their clinical utility.

Data availability statement

The RNA-seq data generated in this study have been deposited in the NCBI Sequence Read Archive (SRA) database under the BioProject accession number PRJNA1382517. The data are publicly accessible via the following URL: https://www.ncbi.nlm.nih.gov/sra/PRJNA1382517.

Ethics statement

The studies involving humans were approved by the Chongqing General Hospital, Chongqing University. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

Author contributions

JZ: Writing – original draft, Writing – review & editing. HM: Writing – original draft, Writing – review & editing. JK: Investigation, Software, Writing – review & editing. ZT: Conceptualization, Data curation, Methodology, Writing – review & editing. ZC: Methodology, Supervision, Writing – review & editing. GY: Formal analysis, Project administration, Validation, Writing – review & editing. LY: Funding acquisition, Resources, Writing – review & editing. JG: Visualization, Writing – review & editing. XY: Investigation, Writing – review & editing. CP: Methodology, Writing – review & editing. KW: Formal analysis, Writing – review & editing. XC: Resources, Writing – review & editing. SK: Writing – original draft, Writing – review & editing. WY: Writing – original draft, Writing – review & editing.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This study was supported by the Senior Medical Talents Program of Chongqing for Young and Middle-aged (7th, Organization Department of Chongqing Municipal Committee of CPC, Chongqing Municipal Health Commission), Academy’s Hundred Disciplinary Talents Program (No. 2025022106), Chongqing Technology Innovation and Application Development Special Project (No. CSTB2023TIAD-KPX0059), National Clinical Research Center for Otolaryngologic Diseases, Beijing, China (No. 2024KF008), National Natural Science Foundation of China (No. 81873702), Major Programs of Chongqing Science and Health Union (No. 2022DBXM006), and Hearing, Speech, and Cognition Laboratory.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fnagi.2025.1694514/full#supplementary-material

References

Cassinotti, L. R., Ji, L., Borges, B. C., Cass, N. D., Desai, A. S., Kohrman, D. C., et al. (2022). Cochlear Neurotrophin-3 overexpression at mid-life prevents age-related inner hair cell synaptopathy and slows age-related hearing loss. Aging Cell 21:e13708. doi: 10.1111/acel.13708

PubMed Abstract | Crossref Full Text | Google Scholar

Chen, Y. C., Chen, H., Jiang, L., Bo, F., Xu, J. J., Mao, C. N., et al. (2018). Presbycusis disrupts spontaneous activity revealed by resting-state functional MRI. Front. Behav. Neurosci. 12:44. doi: 10.3389/fnbeh.2018.00044

PubMed Abstract | Crossref Full Text | Google Scholar

Chen, Y., Huang, H., Luo, Y., Wu, H., Deng, W., Min, X., et al. (2025). Senolytic treatment alleviates cochlear senescence and delays age-related hearing loss in C57BL/6J mice. Phytomedicine 142:156772. doi: 10.1016/j.phymed.2025.156772

PubMed Abstract | Crossref Full Text | Google Scholar

Ding, R., Huang, W., Shen, C., Pan, Y., Zhong, Y., Kong, B., et al. (2025). DLK/JNK3 upregulation aggravates hair cell senescence in mice cochleae via excessive autophagy. Aging Cell 24:e70099. doi: 10.1111/acel.70099

PubMed Abstract | Crossref Full Text | Google Scholar

Dubno, J. R., Lee, F. S., Matthews, L. J., Ahlstrom, J. B., Horwitz, A. R., and Mills, J. H. (2008). Longitudinal changes in speech recognition in older persons. J. Acoust. Soc. Am. 123, 462–475. doi: 10.1121/1.2817362

PubMed Abstract | Crossref Full Text | Google Scholar

Ege, T., Tao, L., and North, B. J. (2024). The role of molecular and cellular aging pathways on age-related hearing loss. Int. J. Mol. Sci. 25:9705. doi: 10.3390/ijms25179705

PubMed Abstract | Crossref Full Text | Google Scholar

Engle, J. R., Tinling, S., and Recanzone, G. H. (2013). Age-related hearing loss in rhesus monkeys is correlated with cochlear histopathologies. PLoS One 8:e55092. doi: 10.1371/journal.pone.0055092

PubMed Abstract | Crossref Full Text | Google Scholar

Fan, Y., Zhang, Y., Wu, R., Chen, X., Zhang, Y., Chen, X., et al. (2016). miR-431 is involved in regulating cochlear function by targeting Eya4. Biochim. Biophys. Acta 1862, 2119–2126. doi: 10.1016/j.bbadis.2016.08.015

PubMed Abstract | Crossref Full Text | Google Scholar

Feng, Z. Y., Huang, T. L., Li, X. R., Chen, L., Deng, S., Xu, S. R., et al. (2021). 17β-Estradiol promotes angiogenesis of stria vascular in cochlea of C57BL/6J mice. Eur. J. Pharmacol. 913:174642. doi: 10.1016/j.ejphar.2021.174642

PubMed Abstract | Crossref Full Text | Google Scholar

Fujimoto, C., Iwasaki, S., Urata, S., Morishita, H., Sakamaki, Y., Fujioka, M., et al. (2017). Autophagy is essential for hearing in mice. Cell Death Dis. 8:e2780. doi: 10.1038/cddis.2017.194

PubMed Abstract | Crossref Full Text | Google Scholar

Fujita, T., Seist, R., Kao, S. Y., Soares, V., Panano, L., Khetani, R. S., et al. (2023). miR-431 secreted by human vestibular schwannomas increases the mammalian inner ear’s vulnerability to noise trauma. Front. Neurol. 14:1268359. doi: 10.3389/fneur.2023.1268359

PubMed Abstract | Crossref Full Text | Google Scholar

Gratton, M. A., Schmiedt, R. A., and Schulte, B. A. (1996). Age-related decreases in endocochlear potential are associated with vascular abnormalities in the stria vascularis. Hear. Res. 102, 181–190. doi: 10.1016/s0378-5955(96)90017-9

PubMed Abstract | Crossref Full Text | Google Scholar

Gratton, M. A., Smyth, B. J., Lam, C. F., Boettcher, F. A., and Schmiedt, R. A. (1997). Decline in the endocochlear potential corresponds to decreased Na,K-ATPase activity in the lateral wall of quiet-aged gerbils. Hear. Res. 108, 9–16. doi: 10.1016/s0378-5955(97)00034-8

PubMed Abstract | Crossref Full Text | Google Scholar

Ha, S., Kim, K. W., Lee, S. M., Lee, C. H., and Kim, S. Y. (2021). MicroRNA dysregulation in the hippocampus of rats with noise-induced hearing loss. Oxid. Med. Cell Longev. 2021:1377195. doi: 10.1155/2021/1377195

PubMed Abstract | Crossref Full Text | Google Scholar

Hade, M. D., Suire, C. N., and Suo, Z. (2021). Mesenchymal stem cell-derived exosomes: Applications in regenerative medicine. Cells 10:1959. doi: 10.3390/cells10081959

PubMed Abstract | Crossref Full Text | Google Scholar

Hao, F., Shan, C., Zhang, Y., Zhang, Y., and Jia, Z. (2022). Exosomes derived from microRNA-21 overexpressing neural progenitor cells prevent hearing loss from ischemia-reperfusion injury in mice via inhibiting the inflammatory process in the cochlea. ACS Chem. Neurosci. 13, 2464–2472. doi: 10.1021/acschemneuro.2c00234

PubMed Abstract | Crossref Full Text | Google Scholar

Hu, W., Liu, Q., Pan, J., and Sui, Z. (2018). MiR-373-3p enhances the chemosensitivity of gemcitabine through cell cycle pathway by targeting CCND2 in pancreatic carcinoma cells. Biomed. Pharmacother. 105, 887–898. doi: 10.1016/j.biopha.2018.05.091

PubMed Abstract | Crossref Full Text | Google Scholar

Hu, Y., Luo, X., Chen, H., Ke, J., Feng, M., and Yuan, W. (2024). MiR-204-5p regulates SIRT1 to promote the endoplasmic reticulum stress-induced apoptosis of inner ear cells in C57BL/6 mice with hearing loss. PLoS One 19:e0309892. doi: 10.1371/journal.pone.0309892

PubMed Abstract | Crossref Full Text | Google Scholar

Huang, M., Jing, Y., Peng, X., Xie, J., Zhao, Q., Yin, J. L., et al. (2025). Neonatal hearing loss risk factors and miRNA biomarkers identified through AABR screening. J. Multidiscip. Healthc. 18, 2541–2550. doi: 10.2147/JMDH.S506757

PubMed Abstract | Crossref Full Text | Google Scholar

Isaac, R., Reis, F. C. G., Ying, W., and Olefsky, J. M. (2021). Exosomes as mediators of intercellular crosstalk in metabolism. Cell Metab. 33, 1744–1762. doi: 10.1016/j.cmet.2021.08.006

PubMed Abstract | Crossref Full Text | Google Scholar

Itokazu, M., Takehara, T., Toriumi, K., Iwawaki, N., Mori, T., Ondoera, Y., et al. (2025). Aging-related changes of miR-23b-3p expression in extracellular vesicles from mesenchymal stromal cells affect TGFBR3 signaling. Sci. Rep. 15:22186. doi: 10.1038/s41598-025-06982-y

PubMed Abstract | Crossref Full Text | Google Scholar

Jayakody, D. M. P., Friedland, P. L., Martins, R. N., and Sohrabi, H. R. (2018). Impact of aging on the auditory system and related cognitive functions: A narrative review. Front. Neurosci. 12:125. doi: 10.3389/fnins.2018.00125

PubMed Abstract | Crossref Full Text | Google Scholar

Jeng, J. Y., Johnson, S. L., Carlton, A. J., De Tomasi, L., Goodyear, R. J., De Faveri, F., et al. (2020). Age-related changes in the biophysical and morphological characteristics of mouse cochlear outer hair cells. J. Physiol. 598, 3891–3910. doi: 10.1113/JP279795

PubMed Abstract | Crossref Full Text | Google Scholar

Jiang, W., Peng, A., Chen, Y., Pang, B., and Zhang, Z. (2020). Long non-coding RNA EBLN3P promotes the recovery of the function of impaired spiral ganglion neurons by competitively binding to miR-204-5p and regulating TMPRSS3 expression. Int. J. Mol. Med. 45, 1851–1863. doi: 10.3892/ijmm.2020.4545

PubMed Abstract | Crossref Full Text | Google Scholar

Katayama, E. S., Hue, J. J., Loftus, A. W., Ali, S. A., Graor, H. J., Rothermel, L. D., et al. (2025). Stability of microRNAs in serum and plasma reveal promise as a circulating biomarker. Noncoding. RNA Res. 15, 132–141. doi: 10.1016/j.ncrna.2025.08.001

PubMed Abstract | Crossref Full Text | Google Scholar

Lee, J. Y. (2015). Aging and speech understanding. J. Audiol. Otol. 19, 7–13. doi: 10.7874/jao.2015.19.1.7

PubMed Abstract | Crossref Full Text | Google Scholar

Li, L., Xu, K., Bai, X., Wang, Z., Tian, X., and Chen, X. (2023). UCHL1 regulated by Sp1 ameliorates cochlear hair cell senescence and oxidative damage. Exp. Ther. Med. 25:94. ume>:94. doi: 10.3892/etm.2023.11793

PubMed Abstract | Crossref Full Text | Google Scholar

Li, W., Li, Y., Wang, M., Liu, H., Hong, G., Jiang, L., et al. (2025). TNFAIP8L2 maintains hair cell function and regulates age-related hearing loss via mTORC1 signaling. Mol. Ther. 33, 3036–3055. doi: 10.1016/j.ymthe.2025.03.046

PubMed Abstract | Crossref Full Text | Google Scholar

Li, Y., Peng, A., Ge, S., Wang, Q., and Liu, J. (2014). miR-204 suppresses cochlear spiral ganglion neuron survival in vitro by targeting TMPRSS3. Hear. Res. 314, 60–64. doi: 10.1016/j.heares.2014.05.002

PubMed Abstract | Crossref Full Text | Google Scholar

Liu, H., Chen, Y., Huang, Y., Wei, L., Ran, J., Li, Q., et al. (2024a). Macrophage-derived mir-100-5p orchestrates synovial proliferation and inflammation in rheumatoid arthritis through mTOR signaling. J. Nanobiotechnol. 22:197. doi: 10.1186/s12951-024-02444-1

PubMed Abstract | Crossref Full Text | Google Scholar

Liu, H., Kuang, H., Wang, Y., Bao, L., Cao, W., Yu, L., et al. (2024b). MSC-derived exosomes protect auditory hair cells from neomycin-induced damage via autophagy regulation. Biol. Res. 57:3. doi: 10.1186/s40659-023-00475-w

PubMed Abstract | Crossref Full Text | Google Scholar

Liu, H., Li, F., Li, X., Wu, Q., and Dai, C. (2022). Rapamycin ameliorates age-related hearing loss in C57BL/6J mice by enhancing autophagy in the SGNs. Neurosci. Lett. 772:136493. doi: 10.1016/j.neulet.2022.136493

PubMed Abstract | Crossref Full Text | Google Scholar

Lu, Y., Li, X., Zuo, Y., Xu, Q., Liu, L., Wu, H., et al. (2021). miR-373-3p inhibits epithelial-mesenchymal transition via regulation of TGFβR2 in choriocarcinoma. J. Obstet Gynaecol. Res. 47, 2417–2432. doi: 10.1111/jog.14809

PubMed Abstract | Crossref Full Text | Google Scholar

Monastero, R., Magro, D., Venezia, M., Pisano, C., and Balistreri, C. R. A. (2023). promising therapeutic peptide and preventive/diagnostic biomarker for age-related diseases: The Elabela/Apela/Toddler peptide. Ageing Res. Rev. 91:102076. doi: 10.1016/j.arr.2023.102076

PubMed Abstract | Crossref Full Text | Google Scholar

Mussoi, B. S. S., and Brown, C. J. (2019). Age-Related changes in temporal resolution revisited: Electrophysiological and behavioral findings from cochlear implant users. Ear Hear. 40, 1328–1344. doi: 10.1097/AUD.0000000000000732

PubMed Abstract | Crossref Full Text | Google Scholar

Nelson, R. F., Glenn, K. A., Zhang, Y., Wen, H., Knutson, T., Gouvion, C. M., et al. (2007). Selective cochlear degeneration in mice lacking the F-box protein, Fbx2, a glycoprotein-specific ubiquitin ligase subunit. J. Neurosci. 27, 5163–5171. doi: 10.1523/JNEUROSCI.0206-07.2007

PubMed Abstract | Crossref Full Text | Google Scholar

Nik Mohamed Kamal, N. N. S. B., and Shahidan, W. N. S. (2020). Non-Exosomal and exosomal circulatory micrornas: Which are more valid as biomarkers? Front. Pharmacol. 10:1500. doi: 10.3389/fphar.2019.01500

PubMed Abstract | Crossref Full Text | Google Scholar

Ninoyu, Y., and Friedman, R. A. (2024). The genetic landscape of age-related hearing loss. Trends Genet. 40, 228–237. doi: 10.1016/j.tig.2023.12.001

PubMed Abstract | Crossref Full Text | Google Scholar

Ohlemiller, K. K., Dahl, A. R., and Gagnon, P. M. (2010). Divergent aging characteristics in CBA/J and CBA/CaJ mouse cochleae. J. Assoc. Res. Otolaryngol. 11, 605–623. doi: 10.1007/s10162-010-0228-1

PubMed Abstract | Crossref Full Text | Google Scholar

Otte, J., Schunknecht, H. F., and Kerr, A. G. (1978). Ganglion cell populations in normal and pathological human cochleae. Implications for cochlear implantation. Laryngoscope 88(8 Pt 1), 1231–1246. doi: 10.1288/00005537-197808000-00004

PubMed Abstract | Crossref Full Text | Google Scholar

Pang, J., Xiong, H., Lin, P., Lai, L., Yang, H., Liu, Y., et al. (2017). Activation of miR-34a impairs autophagic flux and promotes cochlear cell death via repressing ATG9A: Implications for age-related hearing loss. Cell Death Dis. 8:e3079. doi: 10.1038/cddis.2017.462

PubMed Abstract | Crossref Full Text | Google Scholar

Redis, R. S., Calin, S., Yang, Y., You, M. J., and Calin, G. A. (2012). Cell-to-cell miRNA transfer: From body homeostasis to therapy. Pharmacol. Ther. 136, 169–174. doi: 10.1016/j.pharmthera.2012.08.003

PubMed Abstract | Crossref Full Text | Google Scholar

Rivas-Chacón, L. D. M., Martínez-Rodríguez, S., Madrid-García, R., Yanes-Díaz, J., Riestra-Ayora, J. I., Sanz-Fernández, R., et al. (2021). Role of oxidative stress in the senescence pattern of auditory cells in age-related hearing loss. Antioxidants 10:1497. doi: 10.3390/antiox10091497

PubMed Abstract | Crossref Full Text | Google Scholar

Rivas-Chacón, L. D. M., Yanes-Díaz, J., de Lucas, B., Riestra-Ayora, J. I., Madrid-García, R., Sanz-Fernández, R., et al. (2023). Cocoa polyphenol extract inhibits cellular senescence via modulation of SIRT1 and SIRT3 in auditory cells. Nutrients 15:544. doi: 10.3390/nu15030544

PubMed Abstract | Crossref Full Text | Google Scholar

Sardone, R., Battista, P., Panza, F., Lozupone, M., Griseta, C., Castellana, F., et al. (2019). The age-related central auditory processing disorder: Silent impairment of the cognitive ear. Front. Neurosci. 13:619. doi: 10.3389/fnins.2019.00619

PubMed Abstract | Crossref Full Text | Google Scholar

Schuknecht, H. F., and Gacek, M. R. (1993). Cochlear pathology in presbycusis. Ann. Otol. Rhinol. Laryngol. 102(1 Pt 2), 1–16. doi: 10.1177/00034894931020S101

PubMed Abstract | Crossref Full Text | Google Scholar

Sha, S. H., Kanicki, A., Dootz, G., Talaska, A. E., Halsey, K., Dolan, D., et al. (2008). Age-related auditory pathology in the CBA/J mouse. Hear. Res. 243, 87–94. doi: 10.1016/j.heares.2008.06.001

PubMed Abstract | Crossref Full Text | Google Scholar

Shim, Y. J., Jung, W. H., Billig, A. J., Sedley, W., and Song, J. J. (2023). Hippocampal atrophy is associated with hearing loss in cognitively normal adults. Front. Neurosci. 17:1276883. doi: 10.3389/fnins.2023.1276883

PubMed Abstract | Crossref Full Text | Google Scholar

Tsuchihashi, N. A., Hayashi, K., Dan, K., Goto, F., Nomura, Y., Fujioka, M., et al. (2015). Autophagy through 4EBP1 and AMPK regulates oxidative stress-induced premature senescence in auditory cells. Oncotarget 6, 3644–3655. doi: 10.18632/oncotarget.2874

PubMed Abstract | Crossref Full Text | Google Scholar

Wang, S., Lv, S., Hu, J., Shi, Y., Li, Y., Zhang, J., et al. (2025). Conditional overexpression of neuritin in supporting cell protects cochlear hair cell and delays age-related hearing loss by enhancing autophagy. Int. J. Mol. Sci. 26:3709. doi: 10.3390/ijms26083709

PubMed Abstract | Crossref Full Text | Google Scholar

Wang, W., Sun, Y., Chen, S., Zhou, X., Wu, X., Kong, W., et al. (2015). Impaired unfolded protein response in the degeneration of cochlea cells in a mouse model of age-related hearing loss. Exp. Gerontol. 70, 61–70. doi: 10.1016/j.exger.2015.07.003

PubMed Abstract | Crossref Full Text | Google Scholar

Wang, Y., and Liu, Y. (2025). MiRNA-634 promotes apoptosis in HEI-OC1 cells by regulating AK4 and AKT/mTOR signaling pathways. Hear. Res. 466:109386. doi: 10.1016/j.heares.2025.109386

PubMed Abstract | Crossref Full Text | Google Scholar

Wang, Y., Lun, J., Zhang, Y., Yu, M., Liu, X., Guo, J., et al. (2024a). miR-373-3p promotes aerobic glycolysis in colon cancer cells by targeting MFN2. Acta Biochim Biophys. Sin 56, 1498–1508. doi: 10.3724/abbs.2024090

PubMed Abstract | Crossref Full Text | Google Scholar

Wang, Y., Wang, M., and Xie, R. (2019). D-Stellate neurons of the ventral cochlear nucleus decrease in auditory nerve-evoked activity during age-related hearing loss. Brain Sci. 9:302. doi: 10.3390/brainsci9110302

PubMed Abstract | Crossref Full Text | Google Scholar

Wang, Y., Zhao, H., Zhao, K., He, H., Li, X., Qiu, J., et al. (2024b). Chronic inflammation and age-related hearing: Based on mendelian randomization. J. Inflamm. Res. 17, 8921–8934. doi: 10.2147/JIR.S486301

PubMed Abstract | Crossref Full Text | Google Scholar

Wei, Y., Zhang, Y., Cao, W., Cheng, N., Xiao, Y., Zhu, Y., et al. (2025). RONIN/HCF1-TFEB axis protects against d-galactose-induced cochlear hair cell senescence through autophagy activation. Adv. Sci. 12:e2407880. doi: 10.1002/advs.202407880

PubMed Abstract | Crossref Full Text | Google Scholar

Williamson, T. T., Zhu, X., Walton, J. P., and Frisina, R. D. (2015). Auditory brainstem gap responses start to decline in mice in middle age: A novel physiological biomarker for age-related hearing loss. Cell Tissue Res. 361, 359–369. doi: 10.1007/s00441-014-2003-9

PubMed Abstract | Crossref Full Text | Google Scholar

Wong, E. H. C., Dong, Y. Y., Coray, M., Cortada, M., Levano, S., Schmidt, A., et al. (2018). Inner ear exosomes and their potential use as biomarkers. PLoS One 13:e0198029. doi: 10.1371/journal.pone.0198029

PubMed Abstract | Crossref Full Text | Google Scholar

World Health Organization (2021). World report on hearing. Geneva: World Health Organization.

Google Scholar

Xie, L., Zhou, Q., Chen, X., Du, X., Liu, Z., Fei, B., et al. (2021). Elucidation of the Hdac2/Sp1/ miR-204-5p/ Bcl-2 axis as a modulator of cochlear apoptosis via in vivo/ in vitro models of acute hearing loss. Mol. Ther. Nucleic Acids 23, 1093–1109. doi: 10.1016/j.omtn.2021.01.017

PubMed Abstract | Crossref Full Text | Google Scholar

Yu, M. M., Wang, G. J., Wu, K. H., Xue, S. L., Ju, L. L., Li, Q. R., et al. (2021). MicroRNA-373-3p inhibits the growth of cervical cancer by targeting AKT1 both in vitro and in vivo. Acta Biochim. Pol. 68, 611–617. doi: 10.18388/abp.2020_5446

PubMed Abstract | Crossref Full Text | Google Scholar

Zhan, L., Yao, Y., Fu, H., Li, Z., Wang, F., Zhang, X., et al. (2016). Protective role of miR-23b-3p in kainic acid-induced seizure. Neuroreport 27, 764–768.

Google Scholar

Zhang, J., Sun, W., Kuang, S., Gan, Q., Li, H., Ma, H., et al. (2024). miR-130b-3p involved in the pathogenesis of age-related hearing loss via targeting PPARγ and autophagy. Hear. Res. 449:109029. doi: 10.1016/j.heares.2024.109029

PubMed Abstract | Crossref Full Text | Google Scholar

Zhang, J., Wang, N., and Xu, A. (2019). Cmah deficiency may lead to age-related hearing loss by influencing miRNA-PPAR mediated signaling pathway. PeerJ 7:e6856. doi: 10.7717/peerj.6856

PubMed Abstract | Crossref Full Text | Google Scholar

Zhang, Y., Lv, Z., Liu, Y., Cao, H., Yang, J., and Wang, B. (2021). PIN1 protects hair cells and auditory HEI-OC1 cells against senescence by inhibiting the PI3K/Akt/mTOR pathway. Oxid. Med. Cell Longev. 2021:9980444. doi: 10.1155/2021/9980444

PubMed Abstract | Crossref Full Text | Google Scholar