The transcriptomic data GSE58137, GSE65219, GSE129917, and GSE62420 were obtained from the publicly available Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/) maintained by the National Center for Biotechnology Information (NCBI). GSE58137 and GSE65219 include healthy volunteers from different age groups, including both young and elderly individuals, with transcriptomic sequencing performed on their peripheral venous blood samples (25). A summary of demographic characteristics for all included samples is provided in Supplementary Table S1 . GSE129917 contains expression data from Jurkat cells transfected with LEF1 siRNA to investigate the function of LEF1 (26). GSE62420 provides microglia-specific transcriptomic profiles from mouse brain tissues (27).

Differential analysis of the datasets was performed using the “limma” package (28), with a significance threshold of P-value < 0.05 and |log₂Foldchange| > 1 to identify differentially expressed genes (DEGs) associated with aging or LEF1. The overlapping DEGs associated with aging across different datasets were identified, and a Venn diagram was generated using the “VennDiagram” package (29).

To investigate the immune characteristics across different age groups, we used CIBERSORT (30) to analyze the proportions of 22 immune cell subtypes in the largest sample dataset, GSE65219. The infiltration patterns of various immune cell types across samples were visualized using a histogram and a heatmap, generated with the “ggplot2” and “pheatmap” packages.

We analyzed the relationships between LEF1 and other genes using the R package “Hmisc” to calculate correlation coefficients and P-values, applying a significance threshold of r > 0.8, P-value < 0.05. Visualizations were generated via the “ggplot2” package.

Enrichr (https://maayanlab.cloud/Enrichr/) (31) was used for functional and pathway enrichment analyses. The data were visualized using “ggplot2”.

PPI network analysis was performed using the online database STRING (version 12.0) (https://string-db.org/) (32). To ensure network reliability, an interaction composite score > 0.4 was set as the minimum threshold for interactions. The PPI network was visualized using Cytoscape software (version 3.9.0) (33). The CytoHubba plugin was employed to calculate the degree and MCC values for each intersection, identifying hub genes. The top 10 ranked genes were selected for further analysis.

All methylation probes associated with aging traits and located near CpG islands (including CpG islands and their surrounding regions: N_Shore, S_Shore, N_Shelf, and S_Shelf) were downloaded from the Epigenome-Wide Association Study (EWAS) database (https://ngdc.cncb.ac.cn/ewas/atlas) (34). Epigenetic data, including ChIP-seq and ATAC-seq data, were obtained from the ENCODE database (https://www.encodeproject.org/) (35) and visualized using IGV (version 2.11.2) (36).

To construct the shRNA expression vector, the plasmid pKLV-U6gRNA (BbsI)-PGKpuro2ABFP (50946, Addgene) was linearized using BbsI (R3059L, NEB, Ipswich, USA) and BamHI (R0136S, NEB), followed by gel purification. Two sets of shLEF1 oligonucleotides were synthesized by Tsingke Biotechnology (Shanghai, China). The sense and antisense strands of shLEF1 were annealed to form complementary double-stranded DNA fragments, which were then ligated into the linearized 50946 plasmid. The ligated plasmid was subsequently transformed into Escherichia coli Stbl3 receptor cells. Colonies containing the shLEF1 plasmid were cultured in LB medium supplemented with 100 μg/mL ampicillin. All recombinant plasmids were verified by sequencing at Tsingke. The shRNA oligonucleotide sequences used are listed in Table 1 .

All cell lines were obtained from our laboratory. Jurkat cells were cultured in RPMI-1640 medium supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin. HEK-293T and HMC3 cells were cultured with 10% fetal bovine serum, 90% Dulbecco’s modified Eagle’s medium, and 1% penicillin/streptomycin. All cells were maintained at 37°C in a 5% CO₂ atmosphere.

Lentivirus carrying the target plasmid was produced by co-transfecting HEK293T cells with the packaging plasmid psPAX2 (12260, Addgene) and the envelope plasmid pMD2.G (12259, Addgene) using Lipofectamine 2000 (Invitrogen). Fresh media were exchanged 6 h after transfection. After 72 hours, the viral supernatant was collected, centrifuged at 1,000g for 10 minutes at 4°C to remove debris, aliquoted, and stored at -80°C.

For stable expression, Jurkat and HMC3 cells were transduced with the lentivirus. After 24 hours, the medium was replaced with fresh culture medium. Transduced cells were then selected with 1 μg/mL puromycin for 72 hours.

Healthy individuals were recruited after obtaining informed consent for this study. The inclusion criteria were as follows (1): age-matched participants; (2) availability of blood samples for PBMC isolation; and (3) no history of diabetes or other malignant diseases. The study was approved by the Institutional Review Board and Ethics Committee of Huashan Hospital, Fudan University. Participant demographic information is provided in Supplementary Table S1 .

PBMCs were freshly isolated from human peripheral blood using Ficoll-Paque (GE Healthcare). After density gradient centrifugation, the PBMC layer was carefully collected and washed twice with PBS. The cell pellet was then resuspended for further experiments.

Total RNA was extracted from human PBMC using Trizol reagent (Thermo Fisher Scientific). Once the RNA was obtained, it was reverse transcribed into cDNA using the cDNA Synthesis Kit (RR036A, TaKaRa) according to the manufacturer’s instructions. Subsequently, amplification and quantification were performed using RT-qPCR on the QuantStudio™ 7 Pro Real-Time PCR System (Applied Biosystems) with Taq Pro Universal SYBR qPCR Master Mix (Vazyme). The relative expression levels were calculated using the 2^–ΔΔCT method. Housekeeping gene GAPDH expression was used as an internal reference. The primers used are listed in Table 2 .

The culture supernatants from LEF1-knockdown Jurkat and HMC3 cells were collected, and ELISA was performed according to the instructions of the kit. The absorbance value at 450 nm was detected by microplate reader to calculate the expression levels of TNF-α and IL-6. Cytokine concentrations were calculated based on a standard curve. ELISA kits were obtained from Jianglai Biotechnology (Shanghai, China).

The intracellular ROS generation was evaluated using the EVOS M5000 imaging system (Invitrogen). According to the manufacturer’s instructions, intracellular ROS levels were analyzed using Reactive Oxygen Species Assay Kit (S0033, Beyotime). The cells were treated and rinsed with PBS, then incubated at 37 °C for 20 min in 5% CO₂ with DCFH-DA (10 µmol/L, 1 mL) in PBS. Finally, after three washes with PBS, fluorescence images of cells under different treatments were captured using a fluorescence microscope.

Genomic DNA was extracted from PBMCs in blood samples using the DNA Isolation Kit (DP304, TIANGEN). The genomic DNA was then treated with a DNA bisulfite conversion kit (DP215, TIANGEN) to convert unmethylated cytosines into uracils. MethPrimer (https://www.methprimer.com/) (37) was used to predict CpG islands in the LEF1 promoter region and design primer sequences. This study used two primers, one specific for methylated DNA and the other for unmethylated DNA. The primer sequences are listed in Table 3 . The expected PCR products for methylated and unmethylated DNA were 143 bp and 147 bp, respectively. MSP was performed using a methylation-specific PCR kit (EM101, TIANGEN). The PCR products were analyzed by 2% agarose gel electrophoresis.

All experiments were repeated at least three times. Statistical analyses were performed using R (version 4.4.1) and GraphPad Prism 10. Data are presented as mean ± SEM. Detailed statistical analyses for each experiment can be found in the corresponding figure legends. Unless otherwise specified, statistical comparisons were conducted using an unpaired two-tailed Student’s t-test, as indicated in the figure legends. A P-value < 0.05 was considered statistically significant.

To identify aging-related blood biomarkers, we included peripheral blood data from three cohorts of healthy volunteers (GSE58137_GPL6947/GPL10558 and GSE65219). After grouping the samples by age, we performed differential expression analysis for each dataset and identified four common aging-related blood biomarkers: CCR7, FCGBP, IGJ, and LEF1 ( Figure 1A ). These biomarkers exhibited significant differential expression between elderly and young individuals. Furthermore, in the dataset with the largest sample size (GSE65219), we examined the expression levels of these four biomarkers and found that their expression markedly declined with increasing age ( Figure 1B ). Additionally, we performed immune infiltration analysis to examine the dynamic changes in peripheral blood cell subpopulation proportions across different age groups. We observed that during aging, the proportions of monocytes, activated CD4+ memory T cells, and resting NK cells increased, whereas the proportions of B cells, naïve CD4+ T cells, and resting CD4+ memory T cells decreased. These findings suggest that the immune system may be in a state of chronic inflammation or immunosenescence ( Figures 1C, D ).

In recent years, aging-related research has made unprecedented progress, with epigenetics, particularly DNA methylation, playing a mechanistic role in the aging process as one of the nine hallmarks of aging (7). Therefore, we downloaded all aging-associated methylation probes located near CpG islands (including CpG islands and their surrounding regions: N_Shore, S_Shore, N_Shelf, and S_Shelf) from the EWAS database ( Supplementary Table S2 ) (34). Upon annotation, we found that only the promoter region of LEF1-related transcripts contained aging-associated methylation probes. Moreover, previous studies have suggested that LEF1 regulates cellular senescence and aging, and its expression consistently declines with age. Therefore, among the four identified aging-related blood biomarkers, we focused on LEF1 for further investigation.

To further understand the role of LEF1 in the aging process, we first performed a correlation analysis of LEF1 expression in the GSE65219 dataset. We identified co-expressed genes with a correlation coefficient greater than 0.8, as listed in Supplementary Table S3 . Notably, another aging-related blood biomarker, CCR7, was positively correlated with LEF1 expression (R = 0.875, P < 0.05). Furthermore, gene enrichment analysis revealed that LEF1 co-expressed genes were primarily associated with the interleukin-2 signaling pathway, which plays a critical role in regulating T-cell proliferation, differentiation, survival, and function within the immune system ( Figure 2A ).

Subsequently, we utilized RNA sequencing data from GEO (GSE129917), which includes Jurkat cells subjected to siRNA targeting LEF1, to further explore the gene network regulated by LEF1 and its potential involvement in aging-related pathways. Through differential gene enrichment analysis (| logFC | > 1, P < 0.05), our results suggest that LEF1 may influence immune response pathways, including systemic lupus erythematosus, the interleukin-2 signaling pathway, and type II interferon signaling, by regulating associated genes. Additionally, LEF1 may also affect the BDNF signaling pathway and the amyloid pathway, with the BDNF signaling pathway playing a crucial role in neuronal growth, survival, and function, suggesting that LEF1 may be involved in cognitive regulation and neurodegenerative processes ( Figure 2B ). Furthermore, we conducted a PPI analysis of differentially expressed genes using the STRING database and visualized the results in Cytoscape ( Figure 2C ). Using the MCC algorithm in the cytoHubba plugin, we identified the top 10 hub genes, which included histone genes (H4C6, H2AC13, H2AC18) as well as immune and signaling-related genes (CD24, PTPRC, RELB) ( Figure 2D ). These findings suggest that LEF1 may not only influence gene expression and chromatin regulation but also participate in immune signaling modulation, tumor-related pathways, and the NF-κB pathway, thereby affecting cellular states and functions. The detailed information on the 10 hub DEGs is presented in Supplementary Table S4 .

As a nuclear transcription factor, LEF1 specifically binds to cis-regulatory DNA elements (e.g., promoters or enhancers) to regulate target gene expression. Using ChIP-seq data from the ENCODE database, we identified potential LEF1 target genes. Enrichment analysis revealed that LEF1 is associated with genes linked to neurodegenerative diseases, suggesting its regulatory role in neurodegenerative pathways ( Figure 2E ). LEF1-associated target genes related to neurodegenerative diseases, such as EGR1, RHOB, and ID2, exhibit binding peaks ( Supplementary Figures S1A-C ). Following siRNA-mediated LEF1 knockdown, their expression levels were reduced ( Supplementary Figure S1D ). Downregulation of these genes affects key processes, including synaptic plasticity, neuroinflammation, axon regeneration, and neuronal survival, which are implicated in neurodegenerative diseases and brain injury repair (38–40).

Using the DICE and ImmuNexUT databases, we found that LEF1 exhibits the highest expression levels in T cells ( Supplementary Figures S2A, B ). Therefore, we further investigated the function of LEF1 by establishing an LEF1-knockdown human T cell line (Jurkat cells). RT-qPCR experiments confirmed LEF1 knockdown at the RNA level and demonstrated that reduced LEF1 expression affected the mRNA levels of SOCS1 and SOCS3, which are negative regulators of immune responses ( Supplementary Figure S2C , Figures 3A, B ). As suppressors of cytokine signaling, SOCS1 and SOCS3 modulate immune responses by inhibiting cytokine signaling pathways (41). To further validate the pro-inflammatory effects of LEF1 downregulation, we performed ELISA assays. The results showed a significant increase in IL-6 and TNF-α levels in the supernatants of LEF1-knockdown cells ( Figures 3C, D ).

In addition to its specific expression in immune cells, particularly T cells, we found that LEF1 is also specifically expressed in microglia ( Supplementary Figure S3A ). Further analysis revealed that LEF1 expression progressively declined with age in both human brain tissues (based on the BrainSpan Atlas, which profiles gene expression across human brain development) and mouse brain microglia (GSE62420) ( Supplementary Figures S3B, C ), suggesting a potential role of LEF1 in age-related central nervous system regulation (27, 42). Given our previous analysis indicating a strong association between LEF1 and neurodegenerative diseases, as well as its loss promoting inflammatory responses, we hypothesized that LEF1 might mediate neurodegenerative disease progression by regulating neuroinflammation. Therefore, we further investigated the function of LEF1 in human microglial HMC3 cells. Consistent with our findings in Jurkat cells, LEF1 knockdown in microglial cells resulted in a reduction in SOCS1 and SOCS3 expression, accompanied by a significant increase in IL-6 and TNF-α levels in the culture supernatant ( Figures 4A-D , Supplementary Figure S3D ). Additionally, we evaluated whether LEF1 knockdown affects ROS levels in HMC3 cells. ROS was detected using fluorescence staining, where green fluorescence intensity positively correlates with ROS levels. The results showed that LEF1 knockdown significantly increased ROS levels in microglial cells ( Figures 4E , Supplementary Figure S3E ). Notably, oxidative damage is a key mechanism underlying the pathogenesis of neurodegenerative diseases (43, 44).

Previously, we identified methylation probes in the LEF1 promoter region that are associated with aging characteristics ( Supplementary Table S2 ). We hypothesized that the decreased expression of LEF1 during aging might be related to changes in DNA methylation patterns. To test this hypothesis, we recruited 4 elderly individuals over the age of 65 and 4 young individuals between the ages of 18-25, collected their peripheral blood, and isolated PBMCs. RT-qPCR results showed a significant decrease in LEF1 expression in the elderly group ( Supplementary Figure S4 ). To further investigate the DNA methylation status of the LEF1 promoter region during aging, we performed methylation-specific PCR (MSP). First, using the MethPrimer software, we identified a CpG island in the LEF1 promoter region ( Figure 5A ). The MSP results revealed a significant increase in the methylation level of CpG sites in the promoter region of LEF1 in healthy elderly individuals compared to young individuals (P < 0.01) ( Figures 5B, C ).

Additionally, methylation probes in regions near LEF1 have been reported to show increased methylation levels during aging ( Table 4 ). ChIP-seq data from the ENCODE database further confirmed the binding signals of DNA methyltransferases DNMT1 and DNMT3B in the LEF1 promoter region, and this region includes the CpG_225 site ( Figure 5D ), suggesting that LEF1 expression may be regulated by DNA methylation. Beyond DNA methylation, we also analyzed ATAC-seq data from males of different ages. The results revealed that chromatin accessibility in the LEF1 promoter region decreased with age ( Figure 5E ). These findings collectively suggest that LEF1 expression during aging may be regulated by DNA methylation-mediated epigenetic modifications.