In the present study, 8 key ferroptosis-related genes were identified from GSE153434, GSE147021 and FerrDb datasets, and ARTNL, HIF1A, SAT1 and PLIN4 were differentially expressed between normal and TAA/TAD samples. In the erastin-induced ferroptosis model of VSMCs, the above genes showed similar expression changes. In addition, the effect of macrophage infiltration on ferroptosis of VSMCs was detected by the in vitro co-culture model. This study may contribute to a better understanding of the unique role of ferroptosis in TAA/TAD and identify the key ferroptosis-related genes as potential therapeutic targets.
Ferroptosis is characterized by the iron accumulation and the massive lipid oxidation, and leads to oxidative stress and cell death [6]. Ferroptosis participates in various pathological conditions, including infections, neurodegeneration, ischemia/reperfusion injury and cancer [14]. Besides, ferroptosis is closely associated with the inflammatory response in vascular endothelial cells, VSMCs, and macrophages [15–17]. In this study, the most significant GO-enriched term was cellular divalent inorganic cation homeostasis, which showed the association between ferroptosis and TAD/TAA. Besides, we found that the common DEGs were enriched in pathways related to the inflammatory response, such as regulation of inflammatory response and Th17 cell differentiation. Recent studies have demonstrated that a lot of inflammatory cells infiltrate the aortic wall, including macrophages, neutrophils, monocytes, NK cells, T cells, and B cells [18]. In our study, we found ferroptosis-related genes differently expression in VSMCs after co-cultured with M0/M1/M2 macrophages, but other immune cells, such as Tregs, Th cells, monocytes and NK cells, remain further explored.
HIF-1 signaling pathway is a critical pathway for atherosclerosis, ischemic stroke and tumorigenesis. On a cellular level, the ability to adapt to hypoxia is largely controlled by the HIF signaling pathway [19]. Hypoxia-induced activation of HIF induces transcription of genes including erythropoietin and its receptor transferrin, which lead to iron transport and absorption [19, 20]. Consistent with our findings, some studies revealed that tissues from TAA/TAD patients express high levels of HIF1A. However, the relationship between HIF1A and ferroptosis differ from cell to cell. For example, HIF1A was reported as a key factor in increasing the ferroptosis resistance of tumor cells under hypoxia, while HIF1A stabilizer significantly promoted ferroptosis of lung epithelial cells [21]. Our present study pointed out that HIF1A expression in VSMCs was upregulated in the erastin-induced ferroptosis model and macrophage co-culture model. Although there are some differences between the two kinds of models, both models experience ferroptosis and the elevation of the HIF1A level. These results demonstrate that macrophage infiltration promotes ferroptosis of VSMCs via upregulation of HIF1A, which plays a key role in the development of TAA/TAD.
SAT1 was initially identified in a glioblastoma cell as a p53-inducible gene [22]. It has been reported that decreased expression of SAT1 significantly alleviates p53-induced ferroptosis and ROS stress [23]. In our study, the mRNA levels of SAT1 in TAA and TAD were upregulated and similar changes in SAT1 expression were found in VSMCs treated with erastin. According to a bioinformatics-based study, the expression level of SAT1 is correlated with infiltration levels of M0 macrophages, M1 macrophages and CD8 + T cells [24]. Notably, we found that co-culture of VSMCs and macrophages resulted in SAT1 expression being upregulated. These data suggest that macrophage infiltration may promote ferroptosis in VSMCs through overexpression of SAT1. Therefore, SAT1 may play a key function in ferroptosis of VSMCs after immune infiltration and can be exploited as a potential therapeutic target to inhibit TAA/TAD formation.
ARNTL (also called BMAL1) is associated with bipolar disorder, type 2 diabetes and hypertension [25–27]. In a study of ferroptosis, ARNTL destabilized HIF1A and diminished ferroptosis-related tumor cell death [28]. Our discovery of the downregulation of ARNTL in TAA/TAD suggests that ARNTL is a crucial factor in alleviating ferroptosis and negatively regulates TAA/TAD formation. Notably, Minghua and his colleague reported the expression of ARNTL could not be inhibited by erastin (10uM) in Calu-1, HT1080, and HL-60 cell lines [29]. However, we found the expression of ARNTL was downregulated in VSMCs treated with 20uM erastin. When lower concentrations of erastin were used, the downregulation of ATNTL was not statistically significant. One of the reasons for the different results may be the difference in erastin concentration, and the different kinds of cell lines may be another reason. In contrast to erastin induction, ARNTL was significantly upregulated in VSMCs co-cultured with M0/M1/M2 macrophages, suggesting these two types of stimuli induce different unclear downstream signaling pathways of ferroptosis.
DNA damage inducible transcript 4 (DDIT4) was reported to be upregulated in LUAD cell lines treated with erastin, which was consistent with our results in VSMCs [30]. However, no significant change in DDIT4 expression was observed between normal and TAA/TAD tissue samples. The relatively small sample sizes and significant heterogeneity may explain the fact that no difference was found in aortic tissue samples. LPCAT3, SLC7A5, PLIN4 and MUC1 are reported to participate in the ferroptosis pathway. However, the differential expression of these genes in tissue samples and VSMCs ferroptosis models is not as significant as in the database. Besides, the exact regulatory mechanism of these genes is unclear, especially in VSMCs, and further research is required.
This study has several limitations needed to be considered. First, the key ferroptosis-related genes were identified and validated, but the specific pathways were not confirmed by target gene knockdown/overexpression experiments, which would be of importance in future studies. Second, the expression of these key genes was found to be significantly different in TAA/TAD samples compared with control, but these ferroptosis-related genes were not verified in TAA/TAD animal models. Future studies should examine the specific mechanisms explaining what we documented in this study.