Telomere dysfunction is associated with exacerbated intermittent hypoxia-induced cognitive deficits and nerve damage

PC12 cells were purchased from the China Infrastructure of Cell Line Resource (Beijing, China). Cells were cultured in DMEM high glucose mediumsupplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin at 37 °C in a chambercontaining 5% CO₂. PC12 cells in rapamycin group were treated with 20 nM rapamycin (Abcam, UK) before IH exposure. The IH cell model was established following our previously published protocol (Guo et al., 2025). Prior to IH exposure, the culture medium was replaced with fresh serum-free medium. PC12 cells were transferred to a Plexiglas chamber, which was cyclically flushed with hypoxic gas mixture (1.5% O₂, 5% CO₂, N₂; 600 s per cycle) or normoxic gas mixture (21% O₂, 5% CO₂, N₂; 300 s per cycle) for a total of 12 h. The chamber was regulated to maintain optimal conditions at 37 °C with 45% humidity under relatively germ-free conditions.

The Cas tel and Cas scramble plasmids were generously provided by Professor Yong Zhao from Sun Yat-sen University. A mixture containing 7.5 μg of pSPAX, 2.5 μg of pMD2. G, and 10 μg of Cas Tel plasmid was prepared and incubated for 20 min. Subsequently, the mixture was added to HEK293T cells cultured in Opti-MEM (Gibco, USA). After 6–8 h, the medium was replaced with fresh DMEM medium. The supernatant was collected after 48 or 72 h of culture and subjected to ultracentrifugation (4 °C, 25,000 rpm, 2 h) for viral concentration.

PC12 cells were seeded in 6-well plates at a density of 3 × 10⁵ per well. Fresh medium was replaced the next day, and 1 mL of virus was added. After 24 h, the medium was replaced with fresh medium again, followed by antibiotic selection with puromycin (2 μg/mL) after another 24 h.

Telomerase reverse transcriptase (Tert) maintains genomic stability by preserving telomere length through its activity. Building on our previous work, telomerase deficient mice was established (Zhao et al., 2023). Briefly, Tert^+/− mice (generously provided by Professor Yusheng Cong) were crossed to obtain Tert^−/− mice. We successfully established G1 Tert^−/− and G3 Tert^−/− mice, and selected female G3 Tert^−/− mice as a telomerase-deficient progeria mouse model for our experiment. Mice were kept under specific pathogen-free (SPF) conditions with free access to sterilized food and water. All animal experiments were approved by the Animal Ethics Committee of Tianjin Medical University General Hospital (IRB2025-DWFL-050).

IH was induced as previously described (Zhang et al., 2023; Sun et al., 2026). At 6–8 weeks of age, mice in each group were subjected to IH treatment for 9 weeks, 8 h per day (from 9:00 a.m. to 5:00 p.m.). The IH regimen was administered in specialized Plexiglas chambers, with each 90-s cycle consisting of a 55-s hypoxic phase (FiO₂ rapidly reduced to 5% via nitrogen infusion) followed by a 35-s re-oxygenation phase (FiO₂ restored to 21% using compressed air). This cycle was repeated 320 times per day. Throughout the exposure, chamber temperature and humidity were maintained at 22–24 °C and 45%, respectively. Normoxic mice were housed in identical chambers under a constant FiO₂ of 21% for the same duration. All mice were returned to standard cage housing with room air for the remaining 16 h each day.

To evaluate whether a telomere dysfunction-driven progeroid phenotype exacerbates neuronal susceptibility to IH, mice were allocated into five groups: wild-type (WT), G3 Tert^−/−, IH-exposed WT (IH), IH-exposed G3 Tert⁻/⁻ (IH-G3 Tert⁻/⁻), and IH-exposed G3 Tert⁻/⁻ treated with the senolytic agent fisetin (Fisetin). Beginning at 8 weeks of age, mice in fisetin group received gastric gavage with fisetin (20 mg/kg, Meilunbio) six times per week for a total of 9 weeks. Before conducting the gavage procedure, animals were fasted for 6–8 h. WT, G3 Tert^−/−, IH, and IH-G3 Tert^−/− groups received an equivalent volume of placebo. Fisetin was prepared according to the instructions by dissolving in 10% ethanol, 30% PEG 400, and 60% Phosal 50 pg. before gavage.

The Barnes maze test was conducted during the 9th week of IH exposure to assess spatial learning and memory. Prior to formal testing, mice were habituated to the experimental environment. This was followed by four consecutive days of training, a two-day rest period, and finally a probe trial on day seven with the escape tunnel removed. Mice were allowed to freely explore the arena for 2 min while their behavior was videotaped for analysis. The primary outcomes were escape distance and latency, as well as the number of successful crossings over the target hole. These parameters were automatically acquired and analyzed using Digbehv Animal Behavior Analysis System, with manual verification of tracking accuracy to ensure data fidelity.

We performed mice brain MRI acquisition on a 9.4 T MRI scanner (Bruker BioSpec 94/30 USR, Germany). Isoflurane (5%) was used to anesthetize mice and 1–2% isoflurane was used for maintenance. For structural MRI of the brain, we first acquired 2D T2-weighted TurboRARE images using the following parameters: repetition time (TR) = 3,205 ms, echo time (TE) = 33 ms, 20 contiguous slices with 1 mm thickness, field of view (FOV) = 15 mm × 15 mm. The 3D T2-weighted TurboRARE sequence was acquired using the following parameters: TR = 1800 ms, TE = 37 ms, RARE factor = 12, FOV = 18 mm × 18 mm × 9 mm, 60 contiguous slices with 0.15 mm thickness (Cheng et al., 2023).

All subsequent analyses were performed under strict blinding conditions. For hippocampal volume analysis, the original DICOM files were first converted to NIFTI format using dcmniix, followed by 3-fold voxel size augmentation through DPABI. DICOM files were anonymized by an independent researcher not involved in morphometry, using random numerical IDs. Hippocampus was then automatic segmented using SPM12 software with the Turone Mouse Brain Atlas and Template (TMBTA) as the reference space. The TMBTA delineated seven distinct hippocampal subregions: hippocampal formation, CA1, CA2, and CA3 regions, along with three dentate gyrus (DG) subdivisions - polymorph layer, molecular layer, and granule cell layer of DG. The volumes of all subregions were summed to represent total hippocampus. To exclude differences in intracranial size, total intracranial volume (TIV) was calculated as the concomitant variable for normalization of hippocampal volume.

Finally, for brain T2 map imaging, the multiple gradient-echo sequence was acquired with the following parameters: TR = 850 ms, TE = 4 ms, flip angle = 25°, FOV = 18 mm × 18 mm, 18 contiguous slices with 0.141 mm thickness. A researcher blinded manually defined the hippocampus as a region of interest (ROI) in all MR imaging slices of each mouse. The mean T2 relaxation time of hippocampal slices was calculated for each mouse and used as the final quantitativemeasurement.

PC12 cells were cultured on coverslips, fixed with 4% paraformaldehyde (PFA) for 15 min at room temperature, and permeabilized with 0.2% Triton X-100 for 20 min. After 2 × 5 min washes with phosphate-buffered saline (PBS), The cells were dehydrated through a graded ethanol series before being air-dried at room temperature. For telomere fluorescence in situ hybridization (FISH), 0.6 μL of TelC-488-labeled telomere probe (Panagene, Korea) was combined with 50 μL of hybridization buffer containing 35 μL of 70% formamide, 0.5 μL Tris–HCl pH 7.2, 5 μL of 1 × blocking solution, and 9 μL distilled water. The cells were co-incubated with the hybridization buffer. Slides were then denatured at 80 °C for 3 min using a heating block and subsequently hybridized overnight at room temperature in a humidity box. Slides were sequentially washed 2 × 15 min in solutioncontaining 70% formamide, 0.1 M Tris–HCl pH 7.2, 10% BSA and distilled water, followed by 3 × 5 min washes in TBST (1 × TBS, 1% TWEEN-20) and blocked with PBG buffer (1 × PBS, 0.5% BSA, 1% gelatin) for 1 h. Cells were co-incubated with rabbit anti-53BP1 antibody (1:1000 dilution, abcam, ab175933, GR3310237-5) diluted with PBG for 1 h at room temperature. Following 4 × 5 min washes with PBST, cells were co-incubatedwith a fluorophore-conjugated secondary antibody (1:2000 dilution, Thermo Fisher Scientific, A11035, 2,701,068) for 30 min at room temperature. Then cells were dehydrated with a graded ethanol series after 4 × 5 min PBST washes. Nuclei were stained with DAPI (Abcam, UK). Fluorescent images were obtained using a Fluorescence microscope (Zeiss, Germany) and Zen software. For each experimental condition, three independent cell culture and treatment experiments were performed. In each experiment, images were acquired from three randomly selected, non-overlapping fields of view using a 100x oil immersion objective. TIFs were quantified using ImageJ software by counting all foci within each cell that displayed an intact nucleus.

Cellular senescence in PC12 cells was assessed using a Senescence-associated β-Galactosidase Staining Kit (Beyotime, C0602, China), following the manufacturer’s protocol. Cells were fixed with 4% PFA for 10 min. After 3 × 5 min PBS washes, cells were incubated with the SA-β-gal staining working solution (pH 6.0) in a dry, CO₂-free incubator at 37 °C for 12 h in the dark. Following incubation, cells were observed under an optical microscope. For each sample, at least three random, non-overlapping fields were captured. SA-β-gal-positive cells (displaying blue staining) and the total cell number were manually counted using ImageJ software. Each experiment was independently repeated three times.

PC12 cells were cultured on coverslips, fixed with 4% PFA for 15 min at room temperature. After washing with PBS, cells were permeabilized and blocked simultaneously by incubating in PBS containing 0.3% Triton X-100 and 3% bovine serum albumin (BSA) for 2 h at room temperature. Cells were incubated with rabbit anti-p21 antibody (1:200 dilution, Proteintech, 28,248-1-AP, 00137197) diluted with PBS overnight at 4 °C. After washing, cells were incubated with a fluorophore-conjugated secondary antibody (1:200 dilution, Thermo Fisher Scientific, A21206, 2,156,521) for 1 h at room temperature in the dark. DAPI (Abcam, UK) was used for staining the nucleus. After incubation, cells were observed under a fluorescence microscope. For each sample, at least three random, non-overlapping fields were captured. P21-positive cells (displaying green staining) and the total cell number were manually counted using ImageJ software. Each experiment was independently repeated at least three times.

DNA damage in PC12 cells was assessed using the DNA Damage Assay Kit (Beyotime, C2035S, China). Briefly, samples were first incubated with the kit’s primary antibody against γ-H2AX, followed by incubation with a fluorescence-labeled secondary antibody. After incubation, cells were observed under a fluorescence microscope. For each sample, at least three random, non-overlapping fields were captured. γ-H2AX-positive cells (displaying green staining) and the total cell number were manually counted using ImageJ software. Each experiment was independently repeated three times.

PC12 cells were cultured on glass slides, washed three times withPBS, and fixed with 4% PFA for 15 min at room temperature for subsequent staining. Mice were transcardially perfused with ice-cold PBS followed by 4% PFA for tissue fixation. Whole brains were carefully extracted and fixed with 4% PFA at 4 °C for 24 h. The brain samples were then embedded in optimal cutting temperature (OCT, Sakura, USA) after graded sucrose dehydration. The brain samples were sectioned at appropriate thickness using a frozen slicer maintained at −20 °C. The cells or tissue sections were placed at room temperature for 15 min following removal from −20 °Crefrigerator before staining. Following 3 × 5 min PBS washes, cells or brain tissue sections were permeabilized with 0.3% Triton X-100 for 30 min. A TdTdUTP nick-end labeling (TUNEL) apoptosis assay kit (Beyotime Biotechnology Co., Ltd., China) was used to assess cell apoptosis. Cells or brain tissue sections were incubated with TUNEL reaction mixture for 1 h at 37 °C in the dark, followed by nuclear staining with DAPI (Abcam, UK). Fluorescent images were obtained using a Fluorescence microscope (Zeiss, Germany) and Zen software. For in vitro experiments, three independent biological replicates were performed per condition. From each replicate, three random, non-overlapping fields of view were analyzed. For in vivo experiments, each group comprised n = 5 mice. From each mouse, three consecutive brain tissue sections were collected, and three non-overlapping fields of view were analyzed per section.

Total RNA was isolated from PC12 cells in four groups using Trizol (Thermo Fisher Scientific, US), including NC group, Cas tel group, IH 12 h group and IH 12 h-Cas tel group. Three samples are contained in each group. RNA integrity and quantity were assessed using Agilent 2,100 bioanalyzer. After library preparation and quality assessment, RNA sequencing was conducted on the Illumina platform. Following quality control processing, high-quality clean reads were mapped to the reference genome. StringTie was then performed to assemble the mapped reads of each sample.

Quantification of gene expression level was calculated using featureCounts. Differential gene expression analysis was conducted using DESeq2, which implements a negative binomial distribution-based statistical model for RNA-seq data analysis. Statistical significance was adjusted using the false discovery rate (FDR) method for multiple testing, with an adjusted significance threshold of Q-value≤0.05.

Data are presented as the mean ± standard deviation (SD). The Shapiro–Wilk test was used to assess the normality of the data. Statistical analysis was performed using GraphPad Prism v7.0. Comparisons between two independent groups were performed using the unpaired, two-tailed Student’s t-test. For comparisons across multiple groups, a one-way analysis of variance (ANOVA) was performed. When the overall ANOVA was significant (p < 0.05), Tukey’s honestly significant difference (HSD) post hoc test was used for all pairwise comparisons between groups. A p-value of < 0.05 was considered statistically significant for all tests.

To investigate the interaction between telomere dysfunction and IH in vivo, we utilized third-generation telomerase reverse transcriptase knockout mice (G3 Tert⁻/⁻), a well-established progeria model characterized by critically short telomeres and accelerated aging phenotypes. An OSA model was then established by exposing mice to IH, followed by cognitive assessment using the Barnes maze test conducted 9 weeks after IH exposure.

During the training phase (days 1–4) of the Barnes maze, escape latency and escape distance progressively decreased in all groups—including WT, G3 Tert⁻/⁻, IH, and IH–G3 Tert⁻/⁻ mice—indicating learning acquisition. However, on day 4, both escape latency and distance in IHmice increased approximately fourfold compared to those in the WT group. Moreover, IH-G3 Tert^−/− mice exhibited further impairments, with approximately 70% longer escape latency and distance than IH mice. Notably, treatment with the senolytic agent fisetin significantly improved the performance of IH–G3 Tert⁻/⁻ mice in these parameters (Figures 1A–C).

On day 7, during the probe trial in which the escape tunnel was removed, there were no significant differences among the groups in total distance traveled or average movement speed (Figures 1D,E), suggesting that IH treatment did not impair general motor function. However, IH mice crossed the former location of the escape tunnel significantly fewer times than WT mice, indicating impaired spatial memory (Figure 1F). These results suggest that the prolonged escape latency, increased escape distance, and reduced tunnel crossings in IH mice stem from deficits in spatial learning and memory rather than motor impairment. Notably, IH–G3 Tert⁻/⁻ mice displayed more severe cognitive deficits compared to IH mice, and was partially rescued by fisetin treatment (Figure 1F). Together, these results demonstrate that G3 Tert⁻/⁻ mice display exacerbated cognitive deficits following IH exposure, and that these deficits are ameliorated by fisetin treatment.

Magnetic resonance imaging (MRI) is a noninvasive and highly informative modality that enables multidimensional structural and functional assessment of the brain. Clinically, it plays a critical role in evaluating cerebral integrity and diagnosing neurological disorders. Given the well-established link between hippocampal atrophy and mild cognitive impairment (MCI), structural MRI (sMRI) was performed immediately after the Barnes maze test. Three-dimensional T2-weighted TurboRARE sequences were acquired to assess brain structure, followed by volumetric quantification of the hippocampus and its subregions.

Total intracranial volume (TIV) did not significantly differ among groups, excluding intracranial size as a confounding factor (Figures 2A,B). A significant reduction in total hippocampal volume was observed in IH–G3 Tert⁻/⁻ mice compared to IH mice. This volume loss was particularly pronounced in specific subregions, including CA1, CA2, and the dentate gyrus (DG). Notably, treatment with fisetin significantly attenuated hippocampal volume loss in IH–G3 Tert⁻/⁻ mice (Figures 2C–G).

Importantly, alterations in brain microstructure often precede overt volumetric atrophy and can be detected via intrinsic MRI parameters. Among these, T2 relaxation time is a sensitive marker for microstructural changes due to pathological processes. Previous studies have demonstrated that reduced T2 values are correlated with the progression of AD, but so far no research has detected a significant association between OSA models and reduced brain T2 values. To further assess microstructural integrity, T2 mapping MRI was performed using a multi-gradient-echo (MGE) sequence. Axial T2 map images revealed reduced hippocampal T2 signal intensity in IH-G3 Tert^−/− mice compared to IH mice, an effect that was mitigated by fisetin treatment (Figure 2I). Quantitative analysis confirmed that fisetin treatment significantly attenuated the reduction in hippocampal T2 values in IH-G3 Tert^−/− mice (p = 0.577 vs. untreated IH-G3 Tert^−/− mice; p = 0.456 vs. IH mice) (Figure 2H).

To evaluate neuronal death, TUNEL staining was conducted. The number of apoptotic cells in the hippocampus was significantly elevated in IH mice, and even more pronounced in IH–G3 Tert⁻/⁻ mice. This increase in apoptosis was attenuated by fisetin treatment (Figures 2J,K). Collectively, these results demonstrate that G3 Tert^−/− mice showed exacerbated hippocampal atrophy, microstructural disruption, and cellular apoptosis following IH exposure, all of which were alleviated by fisetin treatment.

To model telomere-damaged neuronal cells, PC12 cells were infected with the Cas-Tel virus. Telomeric DNA damage was assessed by quantifying telomere dysfunction-induced foci (TIFs), defined as the colocalization of the DNA damage marker 53BP1 with telomeric regions. TIFs were identified by immunofluorescence, where 53BP1 (red fluorescence) overlapped with TelC-488-labeled telomere probes (green fluorescence) (Figure 3A). Immunofluorescence (IF) analysis revealed that the Cas-Tel group exhibited approximately a twofold increase in TIFs per cell compared to the negative control (NC) group (Figure 3B), confirming that Cas-Tel viral infection effectively induced telomere damage in PC12 cells.

To examine the response of these cells to IH, each group of cells was exposed to IH for 4, 8, or 12 h. IF results demonstrated that IH exposure for 8 and 12 h significantly increased TIFs in a time-dependent manner (Figures 3C,D). Correspondingly, TUNEL staining revealed that IH exposure for 8 or 12 h significantly increased the rate of apoptosis, with 12 h inducing the most pronounced effect (Figures 4A,B). Based on these findings, and to establish a more robust experimental model, 12-h IH exposure was selected for subsequent experiments.

To assess cellular stress responses, we analyzed DNA damage and senescence-associated markers. The percentage of γ-H2AX-positive was significantly higher in Cas-Tel group and was further increased upon IH exposure (Figures 4C,D). Consistently, both the percentage of SA-β-gal-positive cells and p21-positive cells showed a concordant upregulation (Figures 5A–D). In summary, telomere-damaged PC12 cells displayed heightened stress responses to IH, manifested by increased TIFs, apoptosis, and the upregulation of senescence- and DNA damage-associated markers.

To explore the molecular alterations associated with increased IH stress in telomere-damaged cells, we performed RNA sequencing on PC12 cells from four groups: NC, CasTel, IH 12 h, and IH 12 h–CasTel. Comparative transcriptomic analysis identified 847 differentially expressed genes (DEGs) in the IH 12 h–CasTel group compared to the IH 12 h group, including 754 upregulated and 93 downregulated genes (Figure 6A). Gene Ontology (GO) enrichment analysis of biological processes revealed that these DEGs were significantly associated with key functional categories, particularly extracellular matrix (ECM) organization, extracellular structure organization, cofactor transport, and cell–substrate adhesion (Figures 6B,C).

Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis further showed that these DEGs were enriched in several biologically relevant pathways, including neutrophil extracellular trap formation, cell adhesion molecules, cytokine–cytokine receptor interactions, and ECM–receptor interactions. Additionally, key neural signaling pathways—such as GABAergic and glutamatergic synapses—and inflammatory pathways, including TNF and IL-17 signaling, were also significantly represented (Figures 7A,B).

Protein–protein interaction (PPI) network analysis of the DEGs identified the top 10 hub genes (Figure 7C). With the exception of Structural Maintenance of Chromosomes 4 (SMC4), the remaining nine genes clustered within a single functional module and were upregulated. Most of these hub genes were involved in immune-inflammatory responses, notably Interleukin-6 (IL-6), C-X-C motif chemokine ligand 10 (CXCL10), and Intercellular Cell Adhesion Molecule 1 (ICAM1). In summary, RNA sequencing of telomere-damaged PC12 cells exposed to IH revealed a distinct gene expression profile, characterized by the upregulation of pathways and hub genes central to immune and inflammatory responses.

To explore whether pharmacological intervention could mitigate the stress response in our cellular model, we treated telomere-damaged PC12 cells exposed to IH with rapamycin, an mTOR inhibitor with reported anti-aging properties. The TUNEL assay showed that the significant increase in apoptotic cells in the IH 12 h-Cas Tel group was markedly attenuated by rapamycin treatment (Figures 8A,B). Meanwhile, rapamycin treatment also effectively decreased the percentage of cells positive for the senescence markers SA-β-gal and p21 in the IH-Cas Tel group (Figures 8C–F). In summary, rapamycin treatment reversed the IH-induced increase in senescence-associated markers and apoptosis in telomere-damaged PC12 cells.

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.