TYPE Original Research

PUBLISHED 09 August 2022

DOI 10.3389/fcvm.2022.875434

Key ferroptosis-related genes in

OPEN ACCESS abdominal aortic aneurysm
EDITED BY
Georges Nemer,
Hamad bin Khalifa University, Qatar
formation and rupture as
REVIEWED BY
Ruijing Zhang,
determined by combining
Second Hospital of Shanxi Medical
University, China
Haochen Yu,
bioinformatics techniques
LMU Munich University Hospital,
Germany
Jinrui Ren1,2† , Yanze Lv1,2† , Lianglin Wu1,2 , Siliang Chen3 ,
*CORRESPONDENCE
Yuehong Zheng Chuxiang Lei3 , Dan Yang4 , Fangda Li1 , Changzheng Liu5 and
[email protected] Yuehong Zheng2,6*
† These authors have contributed 1
Department of Vascular Surgery, Peking Union Medical College Hospital, Chinese Academy
equally to this work
of Medical Sciences and Peking Union Medical College, Beijing, China, 2 State Key Laboratory
SPECIALTY SECTION of Complex Severe and Rare Diseases, Peking Union Medical College Hospital, Chinese Academy
This article was submitted to of Medical Sciences and Peking Union Medical College, Beijing, China, 3 Chinese Academy
Cardiovascular Genetics and Systems of Medical Sciences and Peking Union Medical College, Beijing, China, 4 Department
Medicine, of Computational Biology and Bioinformatics, Institute of Medicinal Plant Development, Chinese
a section of the journal Academy of Medical Sciences and Peking Union Medical College, Beijing, China, 5 NHC Key
Frontiers in Cardiovascular Medicine Laboratory of Systems Biology of Pathogens, Institute of Pathogen Biology (IPB), Chinese Academy
of Medical Sciences (CAMS) and Peking Union Medical College, Beijing, China, 6 State Key
RECEIVED 19February 2022
Laboratory of Complex Severe and Rare Disease, Department of Vascular Surgery, Peking Union
ACCEPTED 15 July 2022
Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical
PUBLISHED 09 August 2022
College, Beijing, China
CITATION
Ren J, Lv Y, Wu L, Chen S, Lei C,
Yang D, Li F, Liu C and Zheng Y (2022)
Key ferroptosis-related genes
in abdominal aortic aneurysm Objectives: Abdominal aortic aneurysm (AAA) is a cardiovascular disease with
formation and rupture as determined high mortality and pathogenesis closely related to various cell death types,
by combining bioinformatics
techniques. e.g., autophagy, apoptosis and pyroptosis. However, the association between
Front. Cardiovasc. Med. 9:875434. AAA and ferroptosis is unknown.
doi: 10.3389/fcvm.2022.875434

COPYRIGHT
Methods: GSE57691 and GSE98278 dataset were obtained from the Gene
© 2022 Ren, Lv, Wu, Chen, Lei, Yang, Expression Omnibus database, and a ferroptosis-related gene (FRG) set was
Li, Liu and Zheng. This is an
open-access article distributed under
downloaded from the FerrDb database. These data were normalized, and
the terms of the Creative Commons ferroptosis-related differentially expressed genes (FDEGs, AAA vs. normal
Attribution License (CC BY). The use,
samples) were identified using the limma package in R. FRGs expression was
distribution or reproduction in other
forums is permitted, provided the analyzed by Gene Set Expression Analysis (GSEA), and FDEGs were analyzed
original author(s) and the copyright by Gene Ontology (GO) and Kyoto Encyclopedia of Genes (KEGG) pathway
owner(s) are credited and that the
original publication in this journal is enrichment analyses using the clusterProfiler package in R and ClueGO
cited, in accordance with accepted in Cytoscape. Protein–protein interaction networks were assembled using
academic practice. No use, distribution
or reproduction is permitted which Cytoscape, and crucial FDEGs were identified using CytoHubba. Critical FDEG
does not comply with these terms. transcription factors (TFs) were predicted with iRegulon. FDEGs were verified
in GSE98278 set, and key FDEGs in AAA (compared with normal samples)
and ruptured AAA (RAAA; compared with AAA samples) were identified.
Ferroptosis-related immune cell infiltration and correlations with key genes
were analyzed by CIBERSORT. Key FEDGs were reverified in Ang II-induced
AAA models of ApoE−/− and CD57B/6J mice by immunofluorescence assay.

Frontiers in Cardiovascular Medicine 01 frontiersin.org

Ren et al. 10.3389/fcvm.2022.875434

Results: In AAA and normal samples, 40 FDEGs were identified, and the
expression of suppressive FRGs was significantly downregulated with GSEA.
For FDEGs, the GO terms were response to oxidative stress and cellular
response to external stimulus, and the KEGG pathways were the TNF and
NOD-like receptor signaling pathways. IL6, ALB, CAV1, PTGS2, NOX4, PRDX6,
GPX4, HSPA5, HSPB1, and NCF2 were the most enriched genes in the
crucial gene cluster. CEBPG, NFAT5, SOX10, GTF2IRD1, STAT1, and RELA were
potential TFs affecting these crucial genes. Ferroptosis-related immune cells
involved in AAA formation were CD8+ T, naive CD4+ T, and regulatory T
cells (Tregs); M0 and M2 macrophages; and eosinophils. Tregs were also
involved in RAAA. GPX4, SLC2A1, and PEBP1 expression was downregulated
in both the RAAA and AAA samples. GPX4 and PEBP1 were more important in
AAA because they influenced ferroptosis-related immune cell infiltration, and
SLC2A1 was more important in RAAA.
Conclusions: This is the first study to show that ferroptosis is crucial
to AAA/RAAA formation. The TNF and NOD-like signaling pathways and
ferroptosis-related immune cell infiltration play key roles in AAA/RAAA. GPX4
is a key ferroptosis-related gene in AAA. Ferroptosis and related genes might
be promising targets in the treatment of AAA/RAAA.

KEYWORDS

AAA (abdominal aortic aneurysm), RAAA, ferroptosis, infiltration of immune cells,

GPX4

Introduction stress, and VSMC autophagy is also involved in AAA formation

(10). In our previous studies, macrophage pyroptosis was found
Abdominal aortic aneurysms (AAAs) occur more frequently to be involved in AAA (11). Therefore, in addition to the
in men than women (1), with the prevalence of AAAs in 65- inflammatory response, oxidative stress, extracellular matrix
year-old men at 1–2%, and approximately 50% of patients degradation and other processes, many cell death modalities
with a ruptured AAA (RAAA) die before receiving medical play important roles in AAA progression.
treatment (2). Histopathological analyses have revealed that Ferroptosis is a regulated form of cell death (12) that
AAA is associated with inflammation, smooth muscle cell depends on iron and causes unrestricted lipid peroxidation
(SMC) apoptosis and matrix degradation, and AAA risk factors and subsequent membrane damage (13). Iron metabolism
include age, male smoking, and low levels of high-density and lipid peroxidation are increasingly considered to be
lipoprotein (HDL) cholesterol (3). The role of inflammation the central mediators of ferroptosis, which means that
in AAA pathogenesis has been established. Notably, oxidative ferroptosis plays an important role in regulating oxidative stress
stress increases in inflammation, playing an important role in and the inflammatory response (14). Accumulating evidence
the pathophysiology of inflammation (4). In addition, reactive suggests that ferroptosis is involved in diverse cardiovascular
oxygen species (ROS) play key roles in regulating matrix diseases, such as atherosclerosis, stroke, ischemia–reperfusion
metalloproteinases (MMPs) and inducing SMC apoptosis, and injury and heart failure (15–17). Several epidemiological
ROS may also contribute to the pathogenesis of hypertension, studies and animal experiments suggest that ferroptosis plays
which is a risk factor for AAA (5). The most widely used an essential role in the pathophysiology of cardiovascular
classification refers to three types of cell death in animals: diseases, such as atherosclerosis (18). Meanwhile, several
apoptosis, necroptosis, and autophagy (6). The importance of studies demonstrated the mechanism of ferroptosis in aortic
cell death in all stages of atherosclerosis is obvious (7). For dissection (19, 20). Ferroptosis is a novel and critical
example, the apoptosis of vascular SMCs (VSMCs) promotes pathological mechanism involved in SMC loss and AD
the development of AAAs (8); SMC necrotic apoptosis has been development, and Brd4770 is a novel ferroptosis inhibitor
proven to contribute to the AAA pathogenesis (9); autophagy is with the same protective effect against ferrostatin-1 at optimal
related to cellular protection against inflammation and oxidative concentrations (21).

Frontiers in Cardiovascular Medicine 02 frontiersin.org

Ren et al. 10.3389/fcvm.2022.875434

Most important, ferroptosis plays an important role with FDEGs was performed with the clusterProfiler package
in cardiovascular disease by participating in inflammatory (24) in R (4.1.1). Hypergeometric distribution was evaluated
reactions and oxidative stress, regulating cell death and to calculate the significance differences in FDEG expression in
promoting atherosclerosis. each signaling pathway. Using this method, we identified the
The relationship between ferroptosis and AAA has not been signaling pathways significantly affected by FDEGs (P < 0.05).
reported. Recent research suggests that ferroptosis might be
involved in AAA formation, which is a potential research target.
In this study, we focused on analyzing ferroptosis-related genes Gene set enrichment analysis
(FRGs) that play a role in the AAA formation.
Gene set enrichment analysis (GSEA) of the FRGs was
performed with the clusterProfiler package R software. The
Materials and methods GSEA reference sets were built on the basis of the FRG set.
To perform the GSEA, the AAA samples (GSE57691) were
Data acquisition and processing divided into two groups (the AAA and normal sample groups).
In contrast with conventional GSEA, the grouping of these
For this study, gene expression sets were downloaded samples was based on the expression level of each gene in a
from the Genes Expression Omnibus database1 . Specifically, single-gene analysis.
the GSE57691 dataset [including 49 AAA samples 9 arterial
occlusive disease (AOD) and 10 donated normal aortic vessel
samples] (22) and GSE98278 dataset (including 31 unruptured Construction and analysis of a
AAA and 17 RAAA samples). A FRG dataset that included protein–protein interaction network,
information on 259 genes, was downloaded from the FerrDb hub genes and transcription factor
database2 . We used the limma package (23) in R (4.0.5) to prediction
correct the sample data.
The correlations between FDEGs were evaluated by
Pearson’s test and mapped using the Corrplot package in R
Principal component analysis, volcano software. Enriched GO and KEGG of FDEGs were mapped with
plots, and heatmaps the Cytoscape (3.6.5) ClueGO plug-in. The STRING database
[protein–protein interaction (PPI) Networks3 ] was used to
After the data were standardized using the limma construct a PPI network with the identified FDEGs. The PPI file
software package, we identified FRGs and ferroptosis-related was imported into Cytoscape (3.6.5), and the CytoHubba plug-
differentially expressed genes (FDEGs) by comparing AAA in was used to map the PPIs. The degree, closeness, intermediate
samples with normal samples in the GSE57691 dataset. Genes degree of each node in the network, and the average value of
for which the difference in expression was greater than 1.5- the nodal degree of each protein were defined to generate the
fold (| fold change| ≥ 1.5) and the P value (false discovery thresholds for inclusion of PPI network nodes, and proteins for
rate) ≤ 0.05 were considered to be differentially expressed. which the node degree was greater than the threshold value were
Volcano plots and heatmaps showing FDEGs were generated selected for inclusion. The transcription factors identified as hub
using the ggplot2 package in R software. The expression of genes were predicted with the iRegulon plug-in (25).
FDEGs in AAA samples and normal samples was also verified
by determining their expression levels in unruptured AAA and
RAAA samples in the other dataset (GSE98278). Infiltration of immune cells in the
samples
Gene ontology and Kyoto To investigate the infiltration of immune cells in AAA
encyclopedia of genes and genomes and normal aortic vessel samples for which FRG expression
enrichment analyses had been determined, we extracted subsets from the GSE57691
and GSE98278 databases and evaluated and predicted the
Analysis of gene ontology (GO) terms and Kyoto enrichment of immune cells in the samples. We used the
encyclopedia of genes and genomes (KEGG) pathways enriched CIBERSORT package (26) in R software. CIBERSORT is a
tool used to deconvolute the expression matrix of immune

1 https://www.ncbi.nlm.nih.gov/geo/
2 http://www.zhounan.org/ferrdb/ 3 http://string-db.org

Frontiers in Cardiovascular Medicine 03 frontiersin.org

Ren et al. 10.3389/fcvm.2022.875434

cell subtypes based on the principle of linear support vector a confocal microscope (3DHISTECH, Pannoramic MIDI).
regression. The correlations between key FDEGs and immune At least five fields in each coverslip were captured. Image
cells were calculated and mapped using R software. Single genes analysis was performed with ImageJ software (NIH, Bethesda,
from among the key genes that correlated with infiltrating MD, United States).
immune cells were assessed, and the results are reported.

Statistical analysis
Ang II-Induced AAA animal model and
aortic tissue collection Data were analyzed using R software (4.1.1). All values are
presented as the mean ± SEM of three experimental repeats.
ApoE-knockout (ApoE−/− ) male mice and C57BL/6 male Statistical differences between two groups were determined
mice were obtained and housed in a pathogen-free barrier by Student’s t test, and for comparisons between three
facility. AAA was induced in male mice (10–12 weeks old) or more groups, ANOVA followed by Bonferroni multiple
by the infusion of Ang II (1,000 ng/kg per minute; Sigma– comparisons was performed. p < 0.05 was considered to be
Aldrich, St. Louis, MO, United States) or saline for 28 d through statistically significant.
implanted ALZET mini-osmotic pumps (model 2007; DURECT,
Cupertino, CA, United States) as described in our previous
study (27). The mice were euthanized when 16 weeks old, and Results
further histological and molecular analyses were performed.
Abdominal aortic tissues were obtained from AAA and normal
The workflow of this study and the
mice who had undergone the surgical procedures as previously
FDEGs in the AAA aortic vessel and
described. Abdominal aortic tissues were preserved at −80◦ C
or soaked in formalin and embedded in wax blocks. All animal
normal samples
studies were approved by the Institutional Animal Care and
The workflow of the present study is shown in Figure 1. FRG
Use Committee of Peking Union Medical College Hospital, and
sets were extracted from the GSE57691 dataset (49 AAA and
experiments conformed to the Guide for the Care and Use
10 normal samples), and 240 of the 259 FRGs were identified
of Laboratory Animals [National Institutes of Health (NIH)
in both GEO sets. Among genes that met the threshold p
publication no. 85–23, 1996]. The study protocol was approved
value < 0.05 and | fold-change| > 1.5, in the AAA samples, the
by the Ethics Committee of Peking Union Medical College
expression of 278 FRGs was upregulated, and that of 1,307 FRGs
Hospital. All experiments were performed in accordance with
was downregulated (Supplementary Figure 1). We screened
the relevant guidelines and regulations.
40 FDEGs in the AAA samples, among which the expression
of 10 FDEGs was upregulated and that of 30 FDEGs was
downregulated compared with their expression in the normal
Hematoxylin–eosin staining and
samples (Figures 2A–C and Supplementary Table 1).
immunofluorescence

Sections of 5 µm thickness were stained with hematoxylin–

eosin (HE) (Gene Pool, GPB1826, China) according to
The GSEA of each gene
standard pathology-based procedures. For immunofluorescence
A GSEA was performed for each crucial gene
analysis, cells grown on fibronectin-coated glass coverslips or
(Supplementary Figure 2). Gene sets associated with
frozen aortic sections were fixed with 4% paraformaldehyde
suppressive FRGs were significantly downregulated in the
(PFA) for 15 min, permeabilized with 0.1% Triton X-100,
AAA group, which suggested that ferroptosis plays important
and blocked with 10% goat serum at room temperature
roles in AAA progression.
for 1 h. The samples were incubated with anti-glutathione
peroxidase 4 (GPX4) antibody (1:50, sc-166570, Santa Cruz
Biotechnology, United States), anti-solute carrier family 2
member 1 (SLC2A1) antibody (1:200, ab115730, Abcam, GO and KEGG analyses of the FDEGs in
United States), and anti-SLC2A1 antibody (1:200, ab76582, the AAA aortic vessel and normal
Abcam, United States) at 4◦ C overnight. The next day, the slides samples
were incubated with goat anti-rabbit/rat tetramethylrhodamine
(TRITC)/fluorescein isothiocyanate (FITC) secondary antibody The biological functions of the FDEGs in AAA and
for 1 h at room temperature and mounted with DAPI normal samples were identified through GO and KEGG
(Beyotime, C1002, Beijing, China). Images were obtained using pathway analyses (Figure 3 and Supplementary Tables 2, 3).

Frontiers in Cardiovascular Medicine 04 frontiersin.org

Ren et al. 10.3389/fcvm.2022.875434

FIGURE 1
The workflow of this study.

FIGURE 2
FDEGs in AAAs aortic vessels samples and normal samples of GSE57691. The Principal Component Analysis of AAAs aortic vessels samples and
normal samples in GSE57691 set (A). The volcano plots of FDEGs in FRGs subsets of GSE57691 (B). The heatmap of FDEGs (C).

Frontiers in Cardiovascular Medicine 05 frontiersin.org

Ren et al. 10.3389/fcvm.2022.875434

FIGURE 3
The GO and KEGG analysis of FDEGs in AAAs aortic vessels samples and normal samples of GSE57691. The GO-BP, CC, and MF analysis of
FDEGs (A,B). the KEGG analysis of FDEGs (C). The top 5 enriched GO (D) and KEGG (E) pathway of FDEGs.

Frontiers in Cardiovascular Medicine 06 frontiersin.org

Ren et al. 10.3389/fcvm.2022.875434

The crucial FDEG-enriched GO biological processes (BPs) samples, and their expression was downregulated in the AAA
were response to oxidative stress and cellular response to samples compared with that in the normal aortic wall samples.
external stimulus, and the GO cell components (CCs) were
pigment granules and NADPH oxidase complexes, while the
GO molecular functions (MFs) of the enriched FDEGs were
Ferroptosis-related immune cell
peroxidase activity and oxidoreductase activity (Figures 3A,B).
The crucial KEGG pathways were the TNF signaling pathway,
infiltration in the samples
alcoholic liver disease, NOD-like receptor signaling pathway,
We used the CIBERSORT package (R software) to analyze
lipid and atherosclerosis, and biosynthesis of amino acids
immune cell infiltration in the samples, and 49 AAA samples
(Figure 3C). GO terms and KEGG pathways enriched with the
and 10 normal aortic samples extracted from the GSE57691
highest significantly adjusted p values are shown in Table 1
dataset that met the standard for FRG expression were selected
and Figures 3D,E. These results suggested that responses to
for this analysis (P < 0.05). We found a significant difference
ferroptosis-related inflammation and oxidative stress might play
in the abundance of 22 types of immune cells between the
roles in AAA formation.
10 normal aortic samples and 49 AAA samples (Figure 5A).
Specifically, we found significant differences in the infiltration
Construction and analysis of the ratio of CD8+ T cells, naive CD4+ T cells, regulatory T cells
protein–protein interaction network, (Tregs), M0 macrophages, M2 macrophages and eosinophils
between the normal aortic and AAA samples (Figure 5C).
hub genes and transcription factor
Among all the differentially abundant immune cells in the AAA
predictions related to FDEGs involved and normal samples, naive CD4+ T cells and CD8+ T cells were
in AAA formation the most positively correlated (Pearson’s correlation = 0.99),
and the second strongest positive correlation was found
The correlation of FDEGs was calculated on the basis of for CD8+ T cells and Tregs (Pearson’s correlation = 0.59)
their expression in AAA aortic vessel and normal samples (Figure 5B). In addition to these infiltrated immune cell
(Figure 4A). An FDEG coexpression network including GO samples, we selected 31 unruptured AAA samples and 17
terms and KEGG pathways (cellular detoxification, positive RAAA samples extracted from GSE98278 dataset because
regulation of cold-induced thermogenesis, the TNF signaling they met the FRG expression standard, and a Pearson’s test
pathway, etc.) was generated with the Cytoscape ClueGO plug- was performed with these samples to determine correlations
in (Figure 4B). After removing isolated nodes and node pairs, (Figure 5D). The infiltration ratio of Tregs was significantly
the network included 190 edges and 35 nodes (Figure 4C). The different in the RAAA samples (Figure 5F). Specifically, plasma
maximal clique centrality (MCC) method was used (with cells were the most negatively correlated with Tregs (Pearson’s
the Cytoscape plug-in CytoHubba) to generate a subnetwork correlation = −0.84), while CD8+ T cells and Tregs were most
with 10 crucial genes (Figure 4D and Table 2). IL6, ALB, positively correlated in the RAAA and unruptured AAA samples
CAV1, PTGS2, NOX4, PRDX6, GPX4, HSPA5, HSPB1, and (Pearson’s correlation = 0.75) (Figure 5E).
NCF2 showed the highest degrees of centrality in the crucial
gene cluster. CEBPG, NFAT5, SOX10, GTF2IRD1, STAT1, and
RELA were identified TFs that may affect these crucial genes TABLE 1 Top 5 GO and KEGG pathways enriched FDEGs counts
FDEGs in AAA and normal aortic samples.
(Figure 4E).
ID Description Counts

TOP 5-GO
Expression verification of the FDEGs in
GO:0062197 Cellular response to chemical stress 14
the RAAA and unruptured AAA samples GO:0006979 Response to oxidative stress 14
GO:0071496 Cellular response to external 11
We screened 40 FDEGs that showed significantly different stimulus
expression levels in the AAA and normal aortic wall samples. GO:0001894 Tissue homeostasis 9
Thirteen FDEGs in the GSE98278 dataset (17 RAAA and 31 GO:0060249 Anatomical structure homeostasis 9
AAA samples) were differentially expressed; the expression TOP 5-KEGG
of 7 FDEGs was upregulated, and that of 6 FDEGs was hsa04668 TNF signaling pathway 5

downregulated (Table 3 and Supplementary Figure 3). Three hsa04936 Alcoholic liver disease 5

of these 13 genes exhibited similar expression patterns; hsa04621 NOD-like receptor signaling 5
pathway
the expression of GPX4, phosphatidylethanolamine-binding
hsa05417 Lipid and atherosclerosis 5
protein 1 (PEBP1), and SLC2A1 was downregulated in the
hsa01230 Biosynthesis of amino acids 4
RAAA samples compared with that in the unruptured AAA

Frontiers in Cardiovascular Medicine 07 frontiersin.org

Ren et al. 10.3389/fcvm.2022.875434

FIGURE 4
The correlation analysis, the PPI networks and TFs prediction of FDEGs in AAAs aortic vessels samples and normal samples of GSE57691. The
Pearson’s correlation analysis in FDEGs (A). The relationships of pathways and FDEGs using the Clue GO analysis (B). The function clusters of
FDEGs using the MCL of Cytoscape (C). The top 10 hub genes of FDEGs using the Cytohubba of Cytoscape (D). The TFs of the hub FDEGs
predicted by the iRegulon of Cytoscape (E).

Correlations of infiltrating immune neutrophils, and plasma cells in the RAAA and unruptured
cells with GPX4, SLC2A1, and PEBP1 AAA samples, and it was negatively correlated with Tregs,
follicular helper T cells, and activated NK cells in the AAA
expression in the normal, AAA and
samples (Figures 6C,D). The correlations between SLC2A1
RAAA samples
expression and infiltrating immune cells in the single-gene
groups are shown in Supplementary Figures 4E–H. PEBP1
Correlations between GPX4 expression and immune cells
expression was positively correlated with gamma delta T cells,
in the AAA and normal aortic vessel samples (the GSE57691
CD8+ T cells, and resting memory CD4+ T cells in the AAA
dataset) and in the RAAA and unruptured AAA samples
and normal aortic vessel samples (Figure 6E) and negatively
(the GSE98278 dataset) were determined. GPX4 expression
correlated with M2 macrophages and eosinophils. PEBP1
was positively correlated with plasma cells, memory B cells,
expression was negatively correlated with neutrophils, activated
monocytes, naïve B cells, neutrophils, resting natural killer
mast cells, and resting mast cells in the RAAA and unruptured
(NK) cells, and gamma delta T cells, while resting dendritic
AAA samples (Figure 6F).
cells, Tregs, follicular helper T cells, M2 macrophages, and
eosinophils were negatively correlated with GPX4 expression
in the AAA and normal aortic vessel samples (p value < 0.05,
Figure 6A). GPX4 was negatively correlated with resting The expression of key FDEGs was
mast cells in the RAAA and unruptured AAA samples verified in Ang II-induced AAA model
(p value < 0.05, Figure 6B). The correlations between mice
GPX4 expression and infiltrating immune cells in single-gene
groups are shown in Supplementary Figures 4A–D. SLC2A1 After 28 d of Ang II treatment in mice, the mouse aortas
expression was only negatively correlated with eosinophils showed significant expansion compared with those in mice
in the AAA and normal aortic vessel samples. However, treated with saline (Figures 7A,C). HE staining revealed that
SLC2A1 expression was positively correlated with monocytes, Ang II alleviated luminal region expansion, media thinning,

Frontiers in Cardiovascular Medicine 08 frontiersin.org

Ren et al. 10.3389/fcvm.2022.875434

TABLE 2 The crucial FDEGs in AAA and normal aortic samples. a greater degree in the AAA tissues compared than in normal
Gene Protein names AAA aortic walls of ApoE−/− models, and it was downregulated
in the AAA tissues compared than in normal aortic walls of
IL6 Interleukin-6 + CD57B/6J models (Figures 7H,I).
ALB Albumin +
CAV1 Caveolin-1 –
PTGS2 Prostaglandin G/H synthase
2
+ Discussion
NOX4 NADPH oxidase 4 –
In the present study, key genes involved in ferroptosis were
PRDX6 Peroxiredoxin-6 –
identified, and the mechanisms of ferroptosis in AAA formation
GPX4 Phospholipid hydroperoxide –
glutathione peroxidase and AAA rupture were explored. Correlations between FRGs
HSPA5 Endoplasmic reticulum – and immune cells in AAA formation and AAA rupture were
chaperone BiP also determined, for the first time, in this study. Moreover,
HSPB1 Heat shock protein beta-1 – we described 40 FDEGs identified from among the FDEGs in
NCF2 Neutrophil cytosol factor 2 + both the GSE57691 and FerrDb datasets. These FDEGs included
30 with downregulated expression and 10 with upregulated
+, Up-regulated expression; –, Down-regulated expression.
expression in the AAA samples compared with that in normal
TABLE 3 The FDEGs in GSE57691 and GSE98278. aortic vessel samples. Then, pathway enrichment analysis with
these FDEGs was performed. The 5 most enriched GO terms
Gene name Protein name RAAA vs AAA vs
AAA normal were cellular response to chemical stress, response to oxidative
stress, cellular response to external stimulus, tissue homeostasis,
LPCAT3 Lysophospholipid + – and anatomical structure homeostasis. The 5 most enriched
acyltransferase 5 KEGG pathways were the TNF signaling pathway, alcoholic
ACO1 Cytoplasmic aconitate + – liver disease, NOD-like receptor signaling pathway, lipid and
hydratase
atherosclerosis, and biosynthesis of amino acids. IL6, ALB,
ULK1 Serine/threonine-protein + –
kinase ULK1 CAV1, PTGS2, NOX4, PRDX6, GPX4, HSPA5, HSPB1, and
PEBP1 Phosphatidylethanolamine- – – NCF2 were the crucial FDEGs involved in AAA formation. The
binding protein FDEGs that contributed to AAA formation were also verified
1
in the RAAA and AAA samples extracted from the GSE98278
LONP1 Lon protease homolog, + –
database. However, GPX4, SLC2A1, and PEBP1 expression
mitochondrial
was downregulated in the RAAA samples compared with that
CAPG Macrophage-capping – +
protein in the unruptured AAA samples, and their expression was
GPX4 Phospholipid – – downregulated in the AAA samples compared with that in the
hydroperoxide glutathione normal aortic wall samples. In addition, among the immune
peroxidase
cells that infiltrated the AAA samples (compared with normal
HSPA5 Endoplasmic reticulum + –
chaperone BiP samples) and RAAA samples (compared with AAA samples)
CBS Cystathionine + – with FDEG expression that met the criteria for selection, the
beta-synthase numbers of CD8+ T cells, naive CD4+ T cells, Tregs, M0
PRDX6 Peroxiredoxin-6 + – macrophages, M2 macrophages and eosinophils were higher
TMBIM4 Protein lifeguard 4 – – in the AAA samples than in the normal samples, and fewer
DDIT4 DNA damage-inducible + – Tregs were found in the RAAA sample than in the AAA
transcript 4 protein sample. Single-gene correlation analysis was performed with
SLC2A1 Solute carrier family 2 – – the FDEG expression subsets extracted from the GSE57691
or GSE98278 datasets. GPX4 expression was highly correlated
with immune cell abundances in the AAA and normal samples,
and adventitia thickening (Figure 7B). SLC2A1 expression was while SLC2A1 expression was more closely related to RAAA.
upregulated to a greater degree in the AAA tissues compared This study may provide a useful reference for studying
than in normal aortic walls of ApoE−/− models, and it the pathological mechanisms of AAA and RAAA from the
was downregulated in the AAA tissues compared than in perspective of bioinformatics.
normal aortic walls of CD57B/6J models (Figures 7D,E). GPX4 Ferroptosis is characterized by the intracellular abundance
expression was downregulated to a greater degree in the AAA of lipid ROS, which are related and result in the lipid
tissues than in normal aortic walls in ApoE−/− and CD57B/6J oxidation, leading to cell membrane injury and cell death (12,
models (Figures 7F,G). PEBP1 expression was upregulated to 13, 28). Ferroptosis is associated with various pathological

Frontiers in Cardiovascular Medicine 09 frontiersin.org

Ren et al. 10.3389/fcvm.2022.875434

FIGURE 5
Ferroptosis-related infiltration of immune cells in the FEGs subsets of GSE57691 and GSE98278. The infiltration of immune cells in FEGs subsets
of GSE57691 (A). The correlation of immune cells in FEGs subsets of GSE57691 (B). The difference infiltration of immune cells in FEGs subsets of
GSE57691 (C). The infiltration of immune cells in FEGs subsets of GSE98278 (D). The correlation of immune cells in FEGs subsets of GSE98278
(E). The difference infiltration of immune cells in FEGs subsets of GSE98278 (F).

Frontiers in Cardiovascular Medicine 10 frontiersin.org

Ren et al. 10.3389/fcvm.2022.875434

FIGURE 6
The correlation of key FDEGs and infiltration of immune cells in the FRGs subsets of GSE57691 and GSE98278. The correlation of GPX4 and
infiltration of immune cells in AAA and normal samples of GSE57691 (A). The correlation of GPX4 and infiltration of immune cells in RAAA and
AAA samples of GSE98278 (B). The correlation of SLC2A1 and infiltration of immune cells in AAA and normal samples of GSE57691 (C). The
correlation of SLC2A1 and infiltration of immune cells in RAAA and AAA samples of GSE98278 (D). The correlation of PEBP1 and infiltration of
immune cells in AAA and normal samples of GSE57691 (E). The correlation of PEBP1 and infiltration of immune cells in RAAA and AAA samples
of GSE98278 (F).

Frontiers in Cardiovascular Medicine 11 frontiersin.org

Ren et al. 10.3389/fcvm.2022.875434

FIGURE 7
The reverified of key FDEGs with Ang II induced-AAA model in ApoE−/− and CD57B/6J mice. The drams of abdominal aorta in groups (A). The
HE staining of abdominal aorta in each group (B). The abdominal aorta diameters increased rate of AAA compared with control samples [(C),
n = 5]. The IF of SLC2A1 on abdominal aorta in groups (D) and mean area of SLC2A1 in groups (E). The IF of GPX4 on abdominal aorta in groups
(F) and mean area of GPX4 in groups (G). The IF of PEBP1 on abdominal aorta in groups (H) and mean area of PEBP1 in groups (I) (n: p > 0.05;
*p < 0.05; **p < 0.01; ***p < 0.001).

Frontiers in Cardiovascular Medicine 12 frontiersin.org

Ren et al. 10.3389/fcvm.2022.875434

conditions, including acute tissue injury, infection, cancer, and samples. T cells constitute the main leukocyte subset in AAA,
neurodegeneration (29). Ferroptosis also plays an important and they aggregate in the highest level in perivascular adipose
role in cardiovascular diseases, such as atherosclerosis, coronary tissue. CD4+ and CD8+ T cell populations have been shown to
heart disease, and cardiac hypertrophy (17, 30, 31). Moreover, be highly activated, and CD4+ T cells have been shown to be
ferroptosis is closely related to the inflammatory response in the highest activation state in AAA wall (45). In the present
and oxidative stress in vascular endothelial cells, SMCs, and study, Tregs were more abundant in unruptured AAA samples
macrophages (17, 32, 33). In this study, we found that FDEGs than in RAAA samples. Tregs may be a potential therapeutic
were mostly enriched in the response to oxidative stress, which target for inhibiting AAA progression (46). Several studies have
includes IL6, PTGS2, PRDX6, GPX4, and HSPB1. As in previous found that Tregs can inhibit AAA formation (47, 48). Tregs
studies, ferroptosis was found to be closely related to ROS, cells could suppress the COX-2 expression (49). Simvastatin
and ROS are crucial to AAA (14, 34). Ferroptosis-inducing treatment prevented the development of Ang II induced
factors can directly or indirectly affect glutathione peroxidase AAA in ApoE(−/−) mice, which may be partially due to the
through different pathways, resulting in decreased intracellular induction of Treg accumulation (50). In this study, we examined
antioxidant capacity and lipid ROS accumulation, which the correlation and differential enrichment of immune cell
ultimately leads to oxidative stress-related cell death (29). ROS infiltration with AAA, rAAA, by extracting data subsets of
and oxidative stress promote the degradation of the extracellular ferroptosis related genes, which was quietly differential with
matrix (ECM) and apoptosis of VSMCs and promote the previous study and indicated the relationships in ferroptosis
formation and progression of AAA (5, 35). However, few studies related genes, immune cells infiltration and AAA formation. We
have reported correlations between lipid peroxidation and AAA, found Tregs was higher in AAA samples compared with health
which might become a new research direction. After oxidative samples, while it was higher in AAA compared with rAAA
stress, some signaling pathways were found to be enriched in ferroptosis related genes subsets. And GPX4 was negatively
with FDEGs, such as the TNF pathway and NOD-like receptor related with Tregs in AAA and healthy samples, which was
signaling pathway. Cigarette smoke extract has been shown to needed more researches.
upregulate TNF-α, MMP-2, and MMP-9 expression in SMCs, Glutathione peroxidase 4 (GPX4), an essential antioxidant
and this effect was inhibited by Fer-1 (a ferroptosis inhibitor) peroxidase that directly reduces phospholipid hydroperoxides
(33). TNF-α signaling-induced oxidative stress and apoptosis even when they are incorporated into membranes and
have been implicated in AAA pathogenesis (36, 37). The core lipoproteins, reduces fatty acid hydroperoxides, cholesterol
role of the Nod-like receptor (NLR) protein family in the hydroperoxides, and thymine hydroperoxides. It also plays a key
innate immune response involves the signal transduction ATP role in protecting cells from oxidative damage by preventing
enzyme, which affects atherosclerosis and AAA formation by membrane lipid peroxidation. GPX4 may be required to prevent
modulating inflammation that can cause autoimmunity (31, cells from undergoing ferroptosis, a non-apoptotic cell death
38, 39). NLRs also play roles in ferroptosis by regulating caused by iron-dependent accumulation of lipid ROS (51), and
oxidative stress (40). Therefore, FRGs may be involved in AAA may protect cells from the toxicity induced by ingested lipid
formation by regulating the TNF and NLR signaling pathways hydroperoxides. GPX4 plays a role in primary T cell responses
and regulating the oxidative stress and inflammatory reaction to viral and parasitic infections by protecting T cells from
in the cells of the aortic wall, such as endothelial cells, SMCs, ferroptosis and supporting T cell expansion. It also plays a
and macrophages. Regulating FRG expression might reduce role in arachidonic acid metabolism in platelets (52). Iron-
oxidative stress and the inflammatory response associated with catalyzed disorder of intracellular lipid peroxide metabolism
AAA, thereby reducing AAA formation. leads to GPX4 inactivation, destruction of the redox balance,
Immune cells in the blood infiltrate tissues, and these and ferroptosis; these findings show that GPX4 plays an
immune cells can be isolated from the tissues they infiltrate. important role in the pathophysiology of cardiovascular diseases
Notably, immune cell infiltration is closely related to clinical such as atherosclerosis, cardiac hypertrophy, cardiomyopathy,
outcomes of carcinomas (41, 42). In recent decades, studies and AAA (18, 53). GPX4 is the essential antioxidant enzyme
have shown that many inflammatory cells, including T cells, that protects against lipid peroxidation and is a regulator
macrophages, dendritic cells, neutrophils, B cells and mast cells, of iron concentration in vascular endothelial cells (32, 54).
infiltrate the aortic wall, suggesting that they may play key In our study, we found that GPX4 was involved in AAA
roles in AAA formation (43). In a previous study, a higher formation by regulating immune cell infiltration, and GPX4
proportion of macrophages, CD8+ T cells, and resting mast cells expression was most positively correlated with plasma cells and
was found in AAA tissues than in normal tissues (44). In our negatively correlated with eosinophils. GPX4 expression was
study, we found ferroptosis-related differences in the infiltration also negatively correlated with resting mast cells in RAAA.
of immune cells, and Tregs, CD8+ T cells, M0 macrophages, M2 Solute carrier family 2 member 1 (SLC2A1), also known as
macrophages, eosinophils and naive CD4+ T cells were found in glucose transporter 1 (GLUT1), enables glucose (a hydrophilic
higher proportions in the AAA tissue than in the normal tissue molecule) to move through the cell membrane, and it can

Frontiers in Cardiovascular Medicine 13 frontiersin.org

Ren et al. 10.3389/fcvm.2022.875434

promote glycolysis and endothelial cell proliferation and inhibit control samples of CD57B/6J mice. This suggests that the
atherosclerotic lesion formation in hyperlipidemic mice (55). CD57B/6J mouse model, might be better choices for studying
SLC2A1-mediated glucose uptake promotes glycolysis, thereby the mechanism of ferroptosis involved in AAA formation. This
promoting pyruvate oxidation and the three-ring acid cycle, may be related to the knockout of ApoE gene, and further
stimulating fatty acid synthesis, and ultimately promoting research is needed to prove its specific mechanism.
lipid peroxidation-dependent ferroptosis (56). The relationship Some limitations of this study need to be noted. First, the key
between SLC2A1 expression and AAA is not clear. In our study, FDEGs we identified were validated with animal models, but the
we found that SLC2A1 expression was closely correlated with specific pathways were not confirmed with in vitro studies or
the infiltration of immune cells, especially in RAAA tissues. other functional studies, which will be important experiments
In FRG subsets extracted from the GSE57691 dataset, SLC2A1 in future research. Second, we found that the expression of these
expression was most negatively related with eosinophils. key genes was significantly different in RAAA, but because of a
However, SLC2A1 expression was most positively correlated lack of animal models and clinical samples, we could not verify
with monocytes and most negatively correlated with activated these FDEGs. Further study may lead to explanations of the
NK cells in FRG subsets of GSE98278. specific mechanisms of these genes in AAA rupture. However,
PEBP1 (phosphatidylethanolamine-binding protein 1), our results can be enlightening, to some extent, in subsequent
also called Raf kinase inhibitor protein (RKIP), inhibits mechanistic studies.
RAF1 kinase activity by inhibiting RAF1 activation, causing
RAF1/MEK complex dissociation and competitively inhibiting
MEK phosphorylation (57). RKIP negatively regulates NLRP1 Conclusion
activation and NLRP3 and NLRC4-induced inflammatory body
formation and plays an important role in the pathogenesis of Our study was the first to show that ferroptosis plays a
type 2 diabetes (58). Overexpression of CUG triplet repeat- crucial role in AAA formation and rupture and that these
binding protein 1 (CELF1) negatively regulates the protein processes are related to TNF and the NOD-like signaling
expression of PEBP1, and knocking down PEBP1 expression pathway regulation of inflammatory and oxidative responses.
partially enhances ROS production and the apoptosis rate Notably, differential expression of FRGs leads to differences in
of neonatal cardiomyocytes (59). PEBP1 is involved in immune cell infiltration in AAA formation and rupture. The key
regulating endothelial cell autophagy and affects atherosclerosis FRGs, GPX4 may be novel therapeutic target in AAA treatment.
(60). However, the impact of PEBP1 on AAA has not been
clearly reported. We found that PEBP1 expression was
downregulated to a greater extent in AAA tissues than in Data availability statement
normal samples and downregulated in RAAA to a greater
extent than in AAA samples. These results indicate that PEBP1 The datasets presented in this study can be found in
is a crucial factor that negatively regulates AAA formation and online repositories. The names of the repository/repositories
suppresses AAA rupture. and accession number(s) can be found below: https://www.
Most important, among the types of cell death, ferroptosis ncbi.nlm.nih.gov/geo/, GSE57691; https://www.ncbi.nlm.nih.
is more extensively involved in AAA formation and AAA gov/geo/, GSE98278.
rupture. Inflammation and the oxidative response are crucial
biological processes in AAA formation and AAA rupture, and
these processes are highly correlated with ferroptosis. FDEGs Ethics statement
were enriched in TNF- and NOD-like signaling pathways
in AAA formation. GPX4, SLC2A1 and PEBP1 expression The animal study was reviewed and approved by the
was downregulated in both the AAA and RAAA samples. Institutional Animal Care and Use Committee of Peking Union
Differential expression of FRGs resulted in differences in Medical College Hospital.
immune cell infiltration in AAA formation and AAA rupture.
According to the correlation analysis of single-gene and immune
cell infiltration in the FRG subsets, GPX4 and PEBP1 was a Author contributions
more important contributor to AAA formation, and SLC2A1
was a more important contributor to RAAA. According to JR, YL, and YZ conceived the focus of the study, wrote, and
the reverified with Ang II- induced AAA models, the key edited the manuscript. JR, YL, LW, and SC contributed expertise
FDEG, GPX4, was downregulated in AAA compared to control with preparation of the manuscript and the figures. CXL, DY,
groups of ApoE−/− and CD57B/6J mice. However, the SLC2A1 FL, and CZL provided the critical revision with insightful and
and PEBP1 expression were upregulated in AAA models constructive comments to improve the manuscript. JR and YL
compared with control samples in ApoE−/− mice, while they contributed equally to this study and manuscript. All authors
were downregulated expression in AAA models compared with read and approved the manuscript.

Frontiers in Cardiovascular Medicine 14 frontiersin.org

Ren et al. 10.3389/fcvm.2022.875434

Funding Publisher’s note

This work was supported by grants from the Natural All claims expressed in this article are solely those of the
Science Foundation of China (grant number 82070492), the authors and do not necessarily represent those of their affiliated
Chinese Academy of Medical Sciences, the Innovation Fund organizations, or those of the publisher, the editors and the
for Medical Sciences (CIFMS 2021-I2M-1-008), and the Youth reviewers. Any product that may be evaluated in this article, or
Research Foundation of Peking Union Medical College Hospital claim that may be made by its manufacturer, is not guaranteed
(Pumch201911865). or endorsed by the publisher.

Conflict of interest Supplementary material

The authors declare that the research was conducted in the The Supplementary Material for this article can be
absence of any commercial or financial relationships that could found online at: https://www.frontiersin.org/articles/10.3389/
be construed as a potential conflict of interest. fcvm.2022.875434/full#supplementary-material

References
1. Norman PE, Powell JT. Abdominal aortic aneurysm: The prognosis in 15. Kobayashi M, Suhara T, Baba Y, Kawasaki NK, Higa JK, Matsui T.
women is worse than in men. Circulation. (2007) 115:2865–9. doi: 10.1161/ Pathological roles of iron in cardiovascular disease. Curr Drug Targets. (2018)
CIRCULATIONAHA.106.671859 19:1068–76. doi: 10.2174/1389450119666180605112235
2. Sakalihasan N, Michel JB, Katsargyris A, Kuivaniemi H, Defraigne JO, Nchimi 16. Chen X, Li X, Xu X, Li L, Liang N, Zhang L, et al. Ferroptosis and
A, et al. Abdominal aortic aneurysms. Nat Rev Dis Primers. (2018) 4:34. doi: 10. cardiovascular disease: Role of free radical-induced lipid peroxidation. Free Radic
1038/s41572-018-0030-7 Res. (2021) 55:405–15. doi: 10.1080/10715762.2021.1876856
3. Nordon IM, Hinchliffe RJ, Loftus IM, Thompson MM. Pathophysiology and 17. Ouyang S, You J, Zhi C, Li P, Lin X, Tan X, et al. Ferroptosis: The potential
epidemiology of abdominal aortic aneurysms. Nat Rev Cardiol (2011) 8:92–102. value target in atherosclerosis. Cell Death Dis. (2021) 12:782. doi: 10.1038/s41419-
doi: 10.1038/nrcardio.2010.180 021-04054-3
4. Chen Q, Wang Q, Zhu J, Xiao Q, Zhang L. Reactive oxygen species: Key 18. Zhang Y, Xin L, Xiang M, Shang C, Wang Y, Wang Y, et al. The
regulators in vascular health and diseases. Br J Pharmacol. (2018) 175:1279–92. molecular mechanisms of ferroptosis and its role in cardiovascular disease. Biomed
doi: 10.1111/bph.13828 Pharmacother. (2022) 145:112423. doi: 10.1016/j.biopha.2021.112423
5. Mccormick ML, Gavrila D, Weintraub NL. Role of oxidative stress in the 19. Zou HX, Qiu BQ, Lai SQ, Huang H, Zhou XL, Gong CW, et al. Role of
pathogenesis of abdominal aortic aneurysms. Arterioscler Thromb Vasc Biol (2007) ferroptosis-related genes in Stanford type a aortic dissection and identification
27:461–9. doi: 10.1161/01.ATV.0000257552.94483.14 of key genes: New insights from bioinformatic analysis. Bioengineered. (2021)
12:9976–90. doi: 10.1080/21655979.2021.1988840
6. Richard S, Hotchkiss AS, Mcdunn JE, Swanson PE. Cell death. N Engl J Med.
(2009) 361:1570–83. doi: 10.1056/NEJMra0901217 20. Chen Y, He Y, Wei X, Jiang DS. Targeting regulated cell death in aortic
aneurysm and dissection therapy. Pharmacol Res. (2022) 176:106048. doi: 10.1016/
7. Rayner KJ. Cell death in the vessel wall: The good, the bad, the ugly.
j.phrs.2021.106048
Arterioscler Thromb Vasc Biol. (2017) 37:e75–81. doi: 10.1161/ATVBAHA.117.
309229 21. Chen Y, Yi X, Huo B, He Y, Guo X, Zhang Z, et al. BRD4770 functions as
a novel ferroptosis inhibitor to protect against aortic dissection. Pharmacol Res.
8. Chai H, Tao Z, Qi Y, Qi H, Chen W, Xu Y, et al. IKK epsilon deficiency
(2022) 177:106122. doi: 10.1016/j.phrs.2022.106122
attenuates angiotensin II-induced abdominal aortic aneurysm formation in mice
by inhibiting inflammation, oxidative stress, and apoptosis. Oxid Med Cell Longev. 22. Biros E, Gabel G, Moran CS, Schreurs C, Lindeman JH, Walker PJ, et al.
(2020) 2020:3602824. doi: 10.1155/2020/3602824 Differential gene expression in human abdominal aortic aneurysm and aortic
9. Zhou T, Deroo E, Yang H, Stranz A, Wang Q, Ginnan R, et al. MLKL and occlusive disease. Oncotarget. (2015) 6:12984–96. doi: 10.18632/oncotarget.3848
CaMKII are involved in RIPK3-mediated smooth muscle cell necroptosis. Cells. 23. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, et al. limma powers
(2021) 10:2397. doi: 10.3390/cells10092397 differential expression analyses for RNA-sequencing and microarray studies.
10. Qin Y, Zheng B, Yang GS, Yang HJ, Zhou J, Yang Z, et al. Salvia miltiorrhiza- Nucleic Acids Res. (2015) 43:e47. doi: 10.1093/nar/gkv007
derived sal-miR-58 induces autophagy and attenuates inflammation in vascular 24. Yu G, Wang LG, Han Y, He QY. clusterProfiler: An R package for comparing
smooth muscle cells. Mol Ther Nucleic Acids. (2020) 21:492–511. doi: 10.1016/j. biological themes among gene clusters. OMICS. (2012) 16:284–7. doi: 10.1089/omi.
omtn.2020.06.015 2011.0118
11. Zhang X, Li F, Wang W, Ji L, Sun B, Xiao X, et al. Macrophage pyroptosis
25. Chen HZ, Wang F, Gao P, Pei JF, Liu Y, Xu TT, et al. Age-associated
is mediated by immunoproteasome subunit β5i (LMP7) in abdominal aortic
sirtuin 1 reduction in vascular smooth muscle links vascular senescence and
aneurysm. Biochem Biophys Res Commun. (2020) 533:1012–20. doi: 10.1016/j.bbrc.
inflammation to abdominal aortic aneurysm. Circ Res. (2016) 119:1076–88. doi:
2020.09.082
10.1161/CIRCRESAHA.116.308895
12. Hadian K, Stockwell BR. SnapShot: Ferroptosis. Cell. (2020) 181:1188–
26. Newman AM, Liu CL, Green MR, Gentles AJ, Feng W, Xu Y, et al. Robust
1188e1. doi: 10.1016/j.cell.2020.04.039
enumeration of cell subsets from tissue expression profiles. Nat Methods. (2015)
13. Tang D, Kroemer G. Ferroptosis. Curr Biol. (2020) 30:R1292–7. doi: 10.1016/ 12:453–7. doi: 10.1038/nmeth.3337
j.cub.2020.09.068
27. Li FD, Nie H, Tian C, Wang HX, Sun BH, Ren HL, et al. Ablation
14. Yu Y, Yan Y, Niu F, Wang Y, Chen X, Su G, et al. Ferroptosis: A cell death and inhibition of the immunoproteasome catalytic subunit LMP7 attenuate
connecting oxidative stress, inflammation and cardiovascular diseases. Cell Death experimental abdominal aortic aneurysm formation in mice. J Immunol. (2019)
Discov. (2021) 7:193. doi: 10.1038/s41420-021-00579-w 202:1176–85. doi: 10.4049/jimmunol.1800197

Frontiers in Cardiovascular Medicine 15 frontiersin.org

Ren et al. 10.3389/fcvm.2022.875434

28. Zhou B, Liu J, Kang R, Klionsky DJ, Kroemer G, Tang D. Ferroptosis is 45. Sagan A, Mikolajczyk TP, Mrowiecki W, Macritchie N, Daly K, Meldrum A,
a type of autophagy-dependent cell death. Semin Cancer Biol. (2020) 66:89–100. et al. T cells are dominant population in human abdominal aortic aneurysms and
doi: 10.1016/j.semcancer.2019.03.002 their infiltration in the perivascular tissue correlates with disease severity. Front
Immunol. (2019) 10:1979. doi: 10.3389/fimmu.2019.01979
29. Li J, Cao F, Yin HL, Huang ZJ, Lin ZT, Mao N, et al. Ferroptosis: Past,
present and future. Cell Death Dis. (2020) 11:88. doi: 10.1038/s41419-020-2 46. Suh MK, Batra R, Carson JS, Xiong W, Dale MA, Meisinger T, et al. Ex vivo
298-2 expansion of regulatory T cells from abdominal aortic aneurysm patients inhibits
aneurysm in humanized murine model. J Vasc Surg. (2020) 72:1087–1096e1081.
30. Zhang X, Zheng C, Gao Z, Chen H, Li K, Wang L, et al. SLC7A11/xCT
doi: 10.1016/j.jvs.2019.08.285
Prevents Cardiac Hypertrophy by Inhibiting Ferroptosis. Cardiovasc Drugs Ther.
(2021) 36:437–47. doi: 10.1007/s10557-021-07220-z 47. Meng X, Yang J, Zhang K, An G, Kong J, Jiang F, et al. Regulatory
T cells prevent angiotensin II-induced abdominal aortic aneurysm
31. Zhou Y, Zhou H, Hua L, Hou C, Jia Q, Chen J, et al. Verification of ferroptosis in apolipoprotein E knockout mice. Hypertension. (2014) 64:875–82.
and pyroptosis and identification of PTGS2 as the hub gene in human coronary doi: 10.1161/HYPERTENSIONAHA.114.03950
artery atherosclerosis. Free Radic Biol Med. (2021) 171:55–68. doi: 10.1016/j.
freeradbiomed.2021.05.009 48. Zhou Y, Wu W, Lindholt JS, Sukhova GK, Libby P, Yu X, et al. Regulatory
T cells in human and angiotensin II-induced mouse abdominal aortic aneurysms.
32. Sakai O, Yasuzawa T, Sumikawa Y, Ueta T, Imai H, Sawabe A, et al. Role Cardiovasc Res. (2015) 107:98–107. doi: 10.1093/cvr/cvv119
of GPx4 in human vascular endothelial cells, and the compensatory activity of
brown rice on GPx4 ablation condition. Pathophysiology. (2017) 24:9–15. doi: 49. Liu B, Kong J, An G, Zhang K, Qin W, Meng X. Regulatory T cells protected
10.1016/j.pathophys.2016.11.002 against abdominal aortic aneurysm by suppression of the COX-2 expression. J Cell
Mol Med. (2019) 23:6766–74. doi: 10.1111/jcmm.14554
33. Sampilvanjil A, Karasawa T, Yamada N, Komada T, Higashi T, Baatarjav C,
et al. Cigarette smoke extract induces ferroptosis in vascular smooth muscle cells. 50. Meng L, Lu Y, Wang X, Sui W, Ge X, Zhong M, et al. Statin therapy protects
Am J Physiol Heart Circ Physiol. (2020) 318:H508–18. doi: 10.1152/ajpheart.00559. against abdominal aortic aneurysms by inducing the accumulation of regulatory T
2019 cells in ApoE(–/–) mice. J Mol Med. (2022) 100:1057–70. doi: 10.1007/s00109-022-
02213-3
34. Li Z, Kong W. Cellular signaling in abdominal aortic aneurysm. Cell Signal.
51. Yang WS, Sriramaratnam R, Welsch ME, Shimada K, Skouta R, Viswanathan
(2020) 70:109575. doi: 10.1016/j.cellsig.2020.109575
VS, et al. Regulation of ferroptotic cancer cell death by GPX4. Cell. (2014) 156:317–
35. Emeto TI, Moxon JV, Au M, Golledge J. Oxidative stress and abdominal aortic 31. doi: 10.1016/j.cell.2013.12.010
aneurysm: Potential treatment targets. Clin Sci. (2016) 130:301–15. doi: 10.1042/
52. Sutherland M, Shankaranarayanan P, Schewe T, Nigam S. Evidence for
CS20150547
the presence of phospholipid hydroperoxide glutathione peroxidase in human
36. Golledge AL, Walker P, Norman PE, Golledge J. A systematic review of platelets: Implications for its involvement in the regulatory network of the 12-
studies examining inflammation associated cytokines in human abdominal aortic lipoxygenase pathway of arachidonic acid metabolism. Biochem J. (2001) 353:91–
aneurysm samples. Dis Markers. (2009) 26:181–8. doi: 10.1155/2009/352319 100. doi: 10.1042/bj3530091
37. Li J, Krishna SM, Golledge J. The Potential role of kallistatin in the 53. Strauss E, Tomczak J, Staniszewski R, Oszkinis G. Associations and
development of abdominal aortic aneurysm. Int J Mol Sci. (2016) 17:1312. doi: interactions between variants in selenoprotein genes, selenoprotein levels and the
10.3390/ijms17081312 development of abdominal aortic aneurysm, peripheral arterial disease, and heart
failure. PLoS One. (2018) 13:e0203350. doi: 10.1371/journal.pone.0203350
38. Proell M, Riedl SJ, Fritz JH, Rojas AM, Schwarzenbacher R. The Nod-Like
Receptor (NLR) family: A tale of similarities and differences. PLoS One. (2008) 54. Xie M, Huang P, Wu T, Chen L, Guo R. Inhibition of miR-214-3p protects
3:e2119. doi: 10.1371/journal.pone.0002119 endothelial cells from ox-LDL-induced damage by targeting GPX4. Biomed Res Int.
(2021) 2021:9919729. doi: 10.1155/2021/9919729
39. Maitiseyiti A, Ci H, Fang Q, Guan S, Shawuti A, Wang H, et al. Identification
of novel long noncoding RNAs and their role in abdominal aortic aneurysm. 55. Yang Q, Xu J, Ma Q, Liu Z, Sudhahar V, Cao Y, et al. PRKAA1/AMPKalpha1-
Biomed Res Int. (2020) 2020:3502518. doi: 10.1155/2020/3502518 driven glycolysis in endothelial cells exposed to disturbed flow protects against
atherosclerosis. Nat Commun. (2018) 9:4667. doi: 10.1038/s41467-018-07132-x
40. Zhu T, Shi L, Yu C, Dong Y, Qiu F, Shen L, et al. Ferroptosis promotes
photodynamic therapy: Supramolecular photosensitizer-inducer nanodrug for 56. Song X, Liu J, Kuang F, Chen X, Zeh HJ III, Kang R, et al. PDK4 dictates
enhanced cancer treatment. Theranostics. (2019) 9:3293–307. doi: 10.7150/thno. metabolic resistance to ferroptosis by suppressing pyruvate oxidation and fatty acid
32867 synthesis. Cell Rep. (2021) 34:108767. doi: 10.1016/j.celrep.2021.108767
41. Tang R, Xu J, Zhang B, Liu J, Liang C, Hua J, et al. Ferroptosis, necroptosis, 57. Keller ET. Metastasis suppressor genes: A role for raf kinase inhibitor protein
and pyroptosis in anticancer immunity. J Hematol Oncol. (2020) 13:110. doi: 10. (RKIP). Anticancer Drugs. (2004) 15:663–9. doi: 10.1097/01.cad.0000136877.89057.
1186/s13045-020-00946-7 b9
42. Yee PP, Li W. Tumor necrosis: A synergistic consequence of metabolic 58. Qin Q, Liu H, Shou J, Jiang Y, Yu H, Wang X. The inhibitor effect of
stress and inflammation. Bioessays. (2021) 43:e2100029. doi: 10.1002/bies.20210 RKIP on inflammasome activation and inflammasome-dependent diseases. Cell
0029 Mol Immunol. (2021) 18:992–1004. doi: 10.1038/s41423-020-00525-3
43. Yuan Z, Lu Y, Wei J, Wu J, Yang J, Cai Z. Abdominal aortic aneurysm: Roles 59. Hu X, Wu P, Liu B, Lang Y, Li T. RNA-binding protein CELF1 promotes
of inflammatory cells. Front Immunol. (2020) 11:609161. doi: 10.3389/fimmu.2020. cardiac hypertrophy via interaction with PEBP1 in cardiomyocytes. Cell Tissue Res.
609161 (2021) 387:111–21. doi: 10.1007/s00441-021-03541-5
44. Nie H, Qiu J, Wen S, Zhou W. Combining bioinformatics techniques to study 60. Wang L, Li H, Zhang J, Lu W, Zhao J, Su L, et al. Phosphatidylethanolamine
the key immune-related genes in abdominal aortic aneurysm. Front Genet. (2020) binding protein 1 in vacular endothelial cell autophagy and atherosclerosis. J
11:579215. doi: 10.3389/fgene.2020.579215 Physiol. (2013) 591:5005–15. doi: 10.1113/jphysiol.2013.262667

Frontiers in Cardiovascular Medicine 16 frontiersin.org