To profile the genetic alterations in aortic tissues collected from patients with AAA, two datasets with similar grouping characteristics were retrieved (Figure 1A). As shown in Figure 1B, differentially expressed genes (DEGs) were identified in patients with AAA based on the GSE47472 dataset (Supplementary Figure 1). Among these DEGs, 169 were significantly upregulated, while 155 were significantly downregulated. The upregulated DEGs were highly enriched in biological processes, such as the perinuclear region of the cytoplasm and regulation of transcription (Figure 1C). Based on the GSE7084 dataset (Supplementary Figure 2), 1,653 AAA-specific DEGs were identified (Figure 1D), which consisted of 698 significantly upregulated and 955 significantly downregulated genes. Different from the GSE47472 dataset, these upregulated DEGs were highly enriched in the inflammatory response and the immune response (Figure 1E). To obtain more reliable candidates associated with AAA, we cross-compared the upregulated genes from both datasets. Four genes, including PTPN22, CPVL, ARHGDIB, and ANGPTL6, were concurrently upregulated (Figure 1F). The receiver operating characteristic (ROC) curve results based on the GSE47472 (Figure 1G) and GSE7084 (Figure 1H) datasets further revealed that these four candidates were highly predictive of the AAA diagnosis. These results suggest that PTPN22, CPVL, ARHGDIB, and ANGPTL6 are potential diagnostic biomarkers for patients with AAA.

Considering that the aortic diameter in AAA is highly associated with the expression of genetic biomarkers, the GSE57691 dataset was analyzed to precisely identify reliable biomarkers (Figure 2A and Supplementary Figure 3). Among the four identified genes (PTPN22, CPVL, ARHGDIB, and ANGPTL6), we found that PTPN22 expression was consistently upregulated in both small and large AAA samples (Figure 2B). To explore whether the increase in PTPN22 expression is conserved among species, the GSE109039 dataset was retrieved (Supplementary Figures 4A,B). An exclusive increase in the expression of PTPN22 was observed in the experimental AAA model (Supplementary Figure 4C). To further validate the expression and diagnostic value of PTPN22, an independent cohort of peripheral blood mononuclear cells (PBMCs) from patients with AAA was derived (17). As expected, PTPN22 expression was significantly increased in PBMCs from AAA patients (Figure 2C). The diagnostic value (p = 0.0067) of PTPN22 in patients with AAA was also confirmed by the ROC curve analysis (Figure 2D). Taken together, these results reveal that PTPN22 is a genetic biomarker for patients with AAA.

Single-cell RNA sequencing is a powerful tool that can be used to dissect cellular heterogeneity within tissues (18). To explore the potential cellular origin of upregulated PTPN22 within aortic tissues from patients with AAA, the GSE166676 dataset, which is consisted of two healthy volunteers and four patients with AAA (18), was retrieved (Figure 3A). First, the clustering analysis revealed 10 distinct cell populations within aortic tissues (Figure 3B), which is consistent with a previous report (18). As shown in Figure 3C, PTPN22 expression was consistently upregulated in aortic tissues from patients with AAA. To determine the potential origin, aortic cells were re-clustered using lineage-specific biomarkers, including EPCAM (epithelial cells), CLDN5 (endothelial cells), COL1A2 (mesenchymal cells), and PTPRC (immune cells) (Figure 3D). As shown in Figure 3E, the distribution of PTPN22-expressing cells was highly co-localized with clusters assigned as immune cells, suggesting that AAA-upregulated PTPN22 may be derived from aortic immune cells. Collectively, these findings suggest that PTPN22 expression in aortic immune cells is remarkably upregulated in patients with AAA.

To precisely determine the cellular origin of AAA-upregulated PTPN22, immune cells were extracted (Figure 4A) and re-clustered with their identities annotated using reported marker genes (Figure 4B). As expected, PTPN22 expression in immune cells was also upregulated in AAA samples (Figure 4C) when compared with control samples. As shown in Figure 4D, PTPN22 expression was uniquely distributed in aortic immune cell subpopulations, including T cells, natural killer (NK) cells, B cells, neutrophils, and dendritic cells (DCs), but not in basophils and monocytes. Together, these results reveal that upregulated PTPN22 expression in immune cells from AAA patients is mainly distributed in the subpopulations, including T cells, NK cells, and B cells.

To evaluate the expression and diagnostic value of PTPN22 in patients with TAAD, thoracic aortic tissues from 11 patients with TAAD and 8 control participants were profiled by RNA sequencing (Figure 5A). We identified 909 DEGs that were significantly upregulated in TAAD samples, and the expression patterns of these genes among various samples are shown using heatmap (Figure 5B). The significantly upregulated DEGs were highly enriched in certain Gene Ontology (GO) terms, including programmed cell death and metabolic processes (Figure 5C), revealing the prominent role of cell death within aortic tissues during TAAD progression. However, unexpectedly, the expression of PTPN22 in TAAD samples was unaltered when compared with control samples (Figure 5D). As shown in Figure 5E, the ROC curve showed that the expression of PTPN22 lacked diagnostic value for TAAD in our aortic tissue-based analysis (p = 0.9388). Although TAAD and AAA are anatomically inextricable, they share several diversities (19, 20). The GSE9106 dataset, which included peripheral blood cell-based RNA profiling of TAA, was retrieved to evaluate the association of PTPN22 with TAA (Supplementary Figures 5A,B). However, PTPN22 expression was unaltered between controls and patients with TAA (Supplementary Figure 5C). The ROC curve analysis further showed that the area under the curve (AUC) for PTPN22 (p = 0.6081) was 0.531 (Supplementary Figure 5D), indicating the limited diagnostic value of PTPN22 for TAA. Collectively, these clinical TAAD sample-based findings confirm that the expression and diagnostic value of PTPN22 is restricted to AAA, rather than TAAD.

VSMCs are the only cell type within the aortic media, and loss of VSMCs resulting from apoptotic cell death contributes etiologically to both human and experimental AAA (21, 22). To explore the potential role of PTPN22 in aortic VSMCs, cells expressing ACTA2, a marker gene for VSMC, were further extracted from scRNA-seq data analyzed above (Supplementary Figure 6). As expected, the expression of PTPN22 was consistently increased in VSMCs from AAA patients (Figure 6A). Although a murine AAA model induced by calcium chloride (CaCl₂) has been well-established (23, 24), there is still a lack of ideal cellular AAA models for in vitro studies (25). To confirm that PTPN22 expression was increased in the in vitro AAA model, a CaCl₂-based apoptotic VSMC model was adopted (Figure 6B) (24, 26). As shown in Figure 6C, VSMCs were morphologically changed in response to CaCl₂ treatment in a concentration-dependent manner. Consistent with a previous report (26), VSMCs treated with CaCl₂ at a concentration of 100 mmol/L showed remarkably diminished α-smooth muscle actin (α-SMA) expression that did not extend entirely across the cell body (Figures 6D,E), suggesting that VSMCs treated with CaCl₂ (100 mmol/L) can reproduce the aortic aneurysmal phenotype in vitro. Moreover, PTPN22 expression was significantly increased in VSMCs treated with CaCl₂, as assessed by quantitative real-time polymerase chain reaction (qRT-PCR) (Figure 6F). Taken together, these results further highlight the important role of PTPN22 in AAA.

The datasets were retrieved from the Gene Expression Omnibus database¹ and were analyzed using RStudio. An adjusted p-value of < 0.05 and a | Log2FC| of ≥ 1 were considered statistically significant. The GSE47472 dataset contained 8 control aortic tissues and 14 AAA neck tissues (41), and the GSE7084 dataset contained 8 control abdominal aortic tissues and 7 AAA tissues. The GPL2507 platform was adopted (42). The GSE57691 dataset consisted of 10 control aortic tissues from organ donors, 20 aortic tissues from patients with small AAA (mean maximum aortic diameter of 54.3 ± 2.3 mm), and 29 aortic tissues from patients with large AAA (mean maximum aortic diameter of 68.4 ± 14.3 mm) (43). The GSE109639 dataset contained tissues from 6 sham mice and 6 mice with experimental AAA induced by CaCl₂ (44). The GSE16676 dataset contained tissues from two healthy volunteers and four patients with AAA (18). A count matrix consisting of transcriptomics data of PBMCs from patients with AAA and healthy volunteers was deposited previously (17).

To explore the potential biological functions of the identified DEGs, the Database for Annotation, Visualization and Integrated Discovery² was employed to conduct the GO analysis (45).

The ROC curve was generated using MedCalc software (version 19.0.7), and the AUC value with 95% confidence interval was calculated.

VSMCs (American Type Culture Collection, Cat#PCS-100-012) were kindly provided by Dr. Tao Zhuang (Shanghai East Hospital, Tongji University School of Medicine) (46). VSMCs were maintained in Dulbecco’s Modified Eagle Medium (Gibco, US) supplemented with 10% fetal bovine serum (Gibco), 100 U/mL penicillin (Gibco), and 100 μg/mL streptomycin (Gibco). VSMCs were incubated at 37°C in a humidified atmosphere supplemented with 5% carbon dioxide. VSMCs at passages 12–15 were used in the present study. To build a reliable experimental AAA model in vitro, CaCl₂ at different concentrations (0, 1, 10, and 100 mmol/L) was used to treat VSMCs for 24 h, as described previously (24, 26). Treatment with 100 mmol/L CaCl₂ for 24 h was ultimately selected to induce AAA in vitro.

For α-SMA immunofluorescence staining, the cells were fixed with 4% paraformaldehyde and blocked with phosphate-buffered saline containing 10% normal goat serum, as previously described (47). The cells were then incubated with primary antibody against α-SMA (Cat#ab7817, Abcam) followed by the corresponding secondary antibody. After, the cells were incubated with DAPI and images were captured using a microscope.

Total RNA was purified using TRIzol (Invitrogen, US), and cDNA libraries were synthesized with the Reverse Transcription Reagent kit (Takara, Japan). cDNA was submitted for qRT-PCR using SYBR Green mix (Takara) with GAPDH normalization. Primers (Supplementary Table 1) were purchased from Sangon Biological Engineering (Shanghai, China). The qRT-PCR data were analyzed using the comparative Ct (^{Δ Δ} Ct) method (48).

The study was performed in accordance with the Declaration of Helsinki and was approved by the ethics committee of our institution, and all patients gave written informed consent before operation (approval number: 2021-KY-079). Thoracic aortic tissues were collected from 11 patients with TAAD and eight control participants. Written informed consent was obtained from all of the study participants.

The data are shown as the mean ± standard deviation. Paired groups were compared using the two-tailed Student’s t-test, while multiple groups were compared using the one-way analysis of variance with Tukey’s multiple comparisons test. A p-value of < 0.05 was considered statistically significant. Statistical analyses were performed using GraphPad Prism 7.