https://orcid.org/0009-0003-5395-7316
Figures
Abstract
Background
Immune microenvironment is one of the essential characteristics of carotid atherosclerosis (CAS), which cannot be reversed by drug therapy alone. Thus, there is a pressing need to develop novel immunoregulatory strategies to delay this pathological process that drives cardiovascular-related diseases. This study aimed to detect changes in the immune microenvironment of vascular tissues at various stages of carotid atherosclerosis, as well as cluster and stratify vascular tissue samples based on the infiltration levels of immune cell subtypes to distinguish immune phenotypes and identify potential hub genes regulating the immune microenvironment of carotid atherosclerosis.
Materials and methods
RNA sequencing datasets for CAS vascular tissue and healthy vascular tissue (GSE43292 and GSE28829) were downloaded from the Gene Expression Omnibus (GEO) database. To begin, the immune cell subtype infiltration level of all samples in both GSE43292 and GSE28829 cohorts was assessed using the ssGSEA algorithm. Following this, consensus clustering was performed to stratify CAS samples into different clusters. Finally, hub genes were identified using the maximum neighborhood component algorithm based on the construction of interaction networks, and their diagnostic efficiency was evaluated.
Results
Compared to the controls, a higher number of immune cell subtypes were enriched in CAS samples with higher immune scores in the GSE43292 cohort. Advanced CAS was characterized by high immune cell infiltration, whereas early CAS was characterized by low immune cell infiltration in the GSE28829 cohort. Moreover, CAS progression may be related to the immune response pathway. Biological processes associated with muscle cell development may impede the progression of CAS. Finally, the hub genes PTPRC, ACTN2, ACTC1, LDB3, MYOZ2, and TPM2 had satisfactory efficacy in the diagnosis and prediction of high and low immune cell infiltration in CAS and distinguishing between early and advanced CAS samples.
Conclusion
The enrichment of immune cells in vascular tissues is a primary factor driving pathological changes in CAS. Additionally, CAS progression may be related to the immune response pathway. Biological processes linked to muscle cell development may delay the progression of CAS. PTPRC, ACTN2, ACTC1, LDB3, MYOZ2, and TPM2 may regulate the immune microenvironment of CAS and participate in the occurrence and progression of the disease.
Citation: Zhang Y, Zhang L, Jia Y, Fang J, Zhang S, Hou X (2024) Screening of potential regulatory genes in carotid atherosclerosis vascular immune microenvironment. PLoS ONE 19(12): e0307904. https://doi.org/10.1371/journal.pone.0307904
Editor: Misbahuddin Rafeeq, King Abdulaziz University Faculty of Medicine, SAUDI ARABIA
Received: January 16, 2024; Accepted: July 13, 2024; Published: December 9, 2024
Copyright: © 2024 Zhang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All RNA sequencing data files are available from the Gene Expression Omnibus (GEO) database (GEO accession number: GSE43292 and GSE28829).
Funding: This work was supported by Hebei Provincial Administration of Traditional Chinese Medicine Project, No: Z2022004. This type of funding mainly includes project funding, equipment procurement, and personnel support.
Competing interests: The authors declare that they have no competing interests.
Introduction
As is well documented, carotid atherosclerosis (CAS), the pathological basis of carotid artery stenosis, is prevalent in middle-aged and elderly individuals [1]. According to a recent epidemiological survey, up to 12% of elderly men and 5% of women suffer from asymptomatic moderate atherosclerotic neck vessel stenosis, while severe carotid artery stenosis accounted for 3% and 1% of elderly men and women, respectively [2, 3]. Indeed, it has emerged as one of the leading causes of accidental death among the elderly population worldwide [4].
The carotid artery is frequently affected by atherosclerosis and is a major contributor to ischemic stroke [5–7]. The risk of cerebral ischemia in patients with carotid atherosclerosis depends on the degree of carotid artery stenosis [8–10]. The pathological basis of atherosclerosis chiefly involves lipid deposition under the intima of the middle great artery [11–13].
The immune response exerts a significant effect on the development of carotid atherosclerosis at all stages [14, 15]. A large number of immune factors were detected in carotid atherosclerotic lesions, including immune cells and their cytokines. Pro-inflammatory cytokines, such as TNF-α, IL-1β, and IL-12, may accelerate the progression of carotid atherosclerosis [16–19]. Conversely, anti-inflammatory cytokines such as IL-10, TGF-β, and Arg-1 may delay the development of carotid atherosclerosis [20, 21].
Immune microenvironment is one of the essential characteristics of carotid atherosclerosis. Given that carotid atherosclerosis cannot be reversed by drug therapy alone, there is an urgent need to formulate new immunoregulatory strategies to prevent this pathological process that causes cardiovascular-related diseases. Therefore, this study aimed to detect changes in the immune microenvironment of vascular tissues at different stages of carotid atherosclerosis, as well as cluster and stratify vascular tissue samples based on the infiltration levels of immune cell subtypes to distinguish between different immune phenotypes and identify potential hub genes regulating the immune microenvironment of carotid atherosclerosis.
Materials and methods
Data download
RNA sequencing datasets for CAS vascular tissue and healthy vascular tissue (GSE43292 [22] and GSE28829 [23]) were retrieved from the Gene Expression Omnibus (GEO) database. Thirty-two CAS samples in the GSE43292 cohort were regarded as the CAS group, and paired 32 normal vascular tissue samples were designated as the control group. In the GSE28829 cohort, 13 early CAS samples and 16 advanced CAS samples were categorized as the early and advanced CAS groups.
Quantification of immune microenvironment
The ssGSEA algorithm was used to convert RNA sequencing data from all samples to immune cell subtype infiltration data. Then, the immune scores of all samples were calculated using the ESTIMATE algorithm.
Consensus cluster
Ward’s method was utilized for consensus clustering to divide CAS samples in the GSE43292 cohort. When K = x (when the samples were divided into x clusters), the cumulative distribution function was the flattest in the range of consensus index values ([0.1, 0.9]).
Differentially expressed genes (DEGs)
DEGs were identified using the following criteria: genes with a logarithm of fold change exceeding 1 and an adjusted P value less than 0.05.
Enrichment analysis
With human genes as the background, enrichment analysis of DEGs was conducted using R package (clusterProfiler) based on gene ontology (GO) and Kyoto encyclopedia of genes and genomes (KEGG) databases. KEGG pathways and GO terms were retained based on a screening condition of q < 0. 05.
Hub gene screening
Protein-protein interaction networks (PPI) of DEGs were generated using the STRING website (http://string-db.org). The CytoHubba plug-in of Cytoscape 3.2 software and the maximum neighborhood component algorithm were employed to screen hub genes among DEGs. The top 10 genes were retained.
Principal component analysis (PCA)
A linear function that converted original variables to new variables and removed redundant information was constructed. Principal component 1 and principal component 2 were used to construct a two-dimensional plane to characterize the phenotype of the samples.
Results
Changes in the immune microenvironment in CAS
With the exception of Type 2 T helper cells, central memory CD8 T cells, central memory CD4 T cells, and effector memory CD8 T cells, the levels of infiltration of the majority of immune cell subtypes were higher in the CAS samples compared to the control samples in the GSE43292 cohort (P < 0.001, Fig 1A). Likewise, the immune score in CAS samples was higher than that in control samples in the GSE43292 cohort (P < 0.001, Fig 1B). In the GSE43292 cohort, PCA based on immune cell subtype infiltration showed that CAS and control samples were clustered in different regions in the coordinate system composed of principal component 1 as the X-axis and principal component 2 as the Y-axis (Fig 1C). Overall, these results exposed that control and CAS samples possessed different immune microenvironment characteristics, which may play a decisive role in disease development and could be used to differentiate between control and CAS samples.
(A) Differences in immune cell subtypes between CAS and control samples. (B) Differences in immune scores between CAS and control samples. (C) Principal component analysis based on the infiltration level of immune cell subtypes in CAS and control samples. Abbreviation: CAS, carotid atherosclerosis; PC, principal component.
Similarly, the infiltration levels of the majority of immune cell subtypes were higher in the advanced CAS samples than in the early CAS samples in the GSE28829 cohort (Fig 2A). As anticipated, the immune score of advanced CAS samples was higher than that of early CAS samples in the GSE43292 cohort (P < 0.001, Fig 2B). Additionally, PCA based on immune cell subtype infiltration demonstrated that advanced and early CAS samples were clustered in different regions in the coordinate system composed of principal component 1 as the X-axis and principal component 2 as the Y-axis (Fig 2C). The immune microenvironment is a key hallmark of disease and changes throughout the disease process. Describing the phenotypes of diseases based on the immune microenvironment can delineate the heterogeneity of CAS and offer new insights into its pathogenesis.
(A) Differences in immune cell subtypes between early and advanced CAS samples. (B) Differences in immune scores between early and advanced CAS samples. (C) Principal component analysis based on the infiltration level of immune cell subtypes in early and advanced CAS samples. Abbreviation: CAS, carotid atherosclerosis; PC, principal component.
Consensus clustering of the CAS samples
To further elucidate the characteristics of the immune microenvironment in CAS samples and explore the underlying mechanism, consensus clustering was carried out on CAS samples in the GSE43292 cohort. When K = 2 (samples were divided into two clusters), the cumulative distribution function exhibited optimal flatness in the range of consensus index values ([0.1, 0.9], Fig 3A). Next, CAS samples in the GSE43292 cohort were divided into 2 clusters, with cluster A comprising 19 samples and cluster B consisting of 13 samples (Fig 3B). The infiltration levels of most immune cell subtypes, with the exception of Type 2 T helper cells, central memory CD8 T cells, central memory CD4 T cells, and effector memory CD8 T cells, were significantly higher in Cluster A, followed by cluster B and control samples, respectively (P < 0.001, Fig 3C). Similarly, the immune score in cluster A was significantly higher compared to cluster B and the control samples (P < 0.001, Fig 3D). PCA based on immune cell subtype infiltration uncovered that cluster A and cluster B samples were clustered in different regions of the coordinate system composed of principal component 1 as the X-axis and principal component 2 as the Y-axis (Fig 3E). Cluster A was characterized by high immune cell infiltration levels, whereas Cluster B was characterized by low immune cell infiltration levels.
(A) Cumulative distribution function curve for different K values. (B) Consensus matrix when K = 2. CAS samples were divided into cluster A and cluster B. (C) Differences in the infiltration level of immune cell subtypes between cluster A, cluster B, and the control group. (D) Differences in the infiltration level of immune cell subtypes between cluster A, cluster B, and the control group. (E) Principal component analysis based on the infiltration level of immune cell subtypes in cluster A and cluster B CAS samples. Abbreviation: CDF, cumulative distribution function; PC, principal component. Note: ***, P < 0.001; **, P < 0.01; ns, P > 0.05.
Enrichment analysis and hub gene screening
Compared to the control group, immune-related KEGG pathway gene sets, such as Toll-like receptor signaling pathways, antigen processing and presentation, and intestinal immune network for IgA production, were enriched in cluster B of the GSE43292 cohort (Fig 4A). On the other hand, immune-related KEGG pathway genes, such as antigen processing and presentation, were enriched in cluster A (Fig 4B). To further explore the enrichment mechanism of immune cells in CAS samples, DEGs in Cluster A and control samples of the GSE43292 cohort were investigated. Up-regulated DEGs in cluster A of the GSE43292 cohort were enriched in biological processes related to the activation of immune response, leukocyte migration, and leukocyte activation involved in immune response (Fig 5A). They were also enriched in cellular components such as the secretory granule membrane, external side of the plasma membrane, and tertiary granule (Fig 5A), as well as in molecular functions such as immune receptor activity (Fig 5A). Finally, they were enriched in KEGG pathways related to the immune response (Fig 5B). Conversely, up-regulated DEGs in the control group of the GSE43292 cohort were enriched in biological processes such as muscle system process, muscle contraction, and muscle tissue development (Fig 5C), in cellular components associated with contractile fiber, myofibril, and sarcomere (Fig 5C), and in molecular functions involving the structural constituent of muscle (Fig 5C). Notably, they were also enriched in KEGG pathways involving dilated cardiomyopathy, hypertrophic cardiomyopathy, and arrhythmogenic right ventricular cardiomyopathy (Fig 5D). Afterward, PPI networks were constructed based on the up-regulated and down-regulated genes of cluster A compared to the control samples. The PPI data of both up-regulated and down-regulated gene sets were imported into Cytoscape software, following which the maximum neighborhood algorithm was applied to identify the top 10 hub genes with the strongest interactions. The highest-ranking hub gene in the PPI network of the up-regulated gene set in cluster A was PTPRC (Fig 5E), while that of the down-regulated gene set was ACTN2 (Fig 5F). Interestingly, the relationship between ACTC1, LDB3, MYOZ2, TPM2, and CAS has not been described in previous literature.
(A) GSVA between cluster B and the control group. (B) GSVA between cluster A and the control group.
(A) GO enrichment analysis of up-regulated genes in cluster A compared to the control group. (B) KEGG enrichment analysis of up-regulated genes in cluster A compared to the control group. (C) GO enrichment analysis of down-regulated genes in cluster A compared to the control group. (D) KEGG enrichment analysis of down-regulated genes in cluster A compared to the control group. (E) Hub gene interaction network of up-regulated genes in cluster A compared to the control group. Red to yellow indicates that genes rank from high to low in the interaction network, with PTPRC being the highest-ranking gene identified via the maximum neighborhood component algorithm. (F) Hub gene interaction network of down-regulated genes in cluster A compared to the control group. Red to yellow indicates that genes rank from high to low in the interaction network, with ACTN2 being the highest-ranking gene identified via the maximum neighborhood component algorithm. Abbreviation: BP, Biological Process; CC, Cellular Component; MF, molecular function.
Diagnostic efficacy of hub genes
The expression level of PTPRC was higher in the cluster A samples compared to the cluster B and control samples (Fig 6A), whereas that of ACTN2 was lower (Fig 6B). The area under the ROC curve of ACTN2 and PTPRC was 0.844 and 0.787 for distinguishing between CAS and control samples, respectively (Fig 6C and 6D). Moreover, the area under the ROC curve of ACTN2 and PTPRC was 0.986 and 0.939 for distinguishing between cluster A and cluster B, respectively (Fig 6E and 6F). The expression level of PTPRC was higher in advanced CAS samples compared to early CAS samples (Fig 7A). In contrast, the expression level of ACTN2 was lower in advanced CAS samples compared with early CAS samples (Fig 7B). The area under the ROC curve of ACTN2 and PTPRC was 0.913 and 0.909 for differentiating between advanced CAS samples and early CAS samples, respectively (Fig 7C and 7D). In CAS samples of the GSE43292 and GSE28829 cohorts, ACTN2 expression was negatively correlated with the infiltration level of most immune cell subtypes (Fig 8A and 8B), whereas PTPRC expression was positively correlated with the infiltration level of most immune cell subtypes (Fig 8A and 8B). Notably, the expression levels of ACTC1, LDB3, MYOZ2, and TPM2 were lower in cluster A samples compared with cluster B and control samples (Fig 9A–9D). The area under the ROC curve of ACTC1, LDB3, MYOZ2, and TPM2 for distinguishing CAS samples from control samples was 0.790, 0.807, 0.831, and 0.834, respectively (Fig 9E–9H). The area under the ROC curve of ACTC1, LDB3, MYOZ2, and TPM2 for distinguishing between cluster A and cluster B was 0.927, 0.988, 0.960 and 0.992, respectively (Fig 9I–9L). Of note, the expression levels of ACTC1 (Fig 10A), MYOZ2 (Fig 10C), and TPM2 (Fig 10D) were lower in advanced CAS samples compared with early CAS samples. However, LDB3 expression was comparable between advanced and early CAS samples (Fig 10B). The area under the ROC curve of ACTC1, LDB3, MYOZ2, and TPM2 for differentiating advanced CAS samples from early CAS samples was 0.721, 0.683, 0.764, and 0.865, respectively (Fig 10E–10H). In the CAS samples of the GSE43292 (Fig 11A) and GSE28829 cohorts (Fig 11B), the expression of ACTC1, LDB3, MYOZ2, and TPM2 was negatively correlated with infiltration levels of most immune cell subtypes.
(A and B) Differences in PTPRC and ACTN2 expression between cluster A, cluster B, and the control group. (C and D) ACTN2 and PTPRC diagnostic efficiency on distinguishing the carotid atherosclerosis and control samples. (E and F) The accuracy of ACTN2 and PTPRC in distinguishing cluster A and cluster B samples. Abbreviation: CAS, carotid atherosclerosis; PC, principal component; AUC, area under curve; CI, confidence interval. Note: ***, P < 0.001; *, P < 0.05; ns, P > 0.05.
(A and B) Difference in PTPRC and ACTN2 expression between advanced and early CAS samples. (C and D) The accuracy of ACTN2 and PTPRC for distinguishing between advanced and early CAS samples. Abbreviation: CAS, carotid atherosclerosis; PC, principal component; AUC, area under curve; CI, confidence interval. Note: ***, P < 0.001; *, P < 0.05; ns, P > 0.05.
(A) Correlation between PTPRC, ACTN2, and immune cell subtypes in CAS samples in the GSE43292 cohort. (B) Correlation between PTPRC, ACTN2, and immune cell subtypes in CAS samples in the GSE28829 cohort. Note: ***, P < 0.001; **, P < 0.01; *, P < 0.05.
(A-D) Differences in ACTC1, LDB3, MYOZ2, and TPM2 expression between cluster A, cluster B, and the control group. (E-H) The accuracy of ACTC1, LDB3, MYOZ2, and TPM2 for distinguishing between CAS and control samples. (I-L) The accuracy of ACTC1, LDB3, MYOZ2, and TPM2 for distinguishing between cluster A and cluster B samples. Abbreviation: CAS, carotid atherosclerosis; AUC, the area under the curve; CI, confidence interval. Note: ***, P < 0.001; *, P < 0.05; ns, P > 0.05.
(A-D) Differences in ACTC1, LDB3, MYOZ2, and TPM2 expression between advanced and early CAS samples. (E-H) The accuracy of ACTC1, LDB3, MYOZ2, and TPM2 for distinguishing advanced from early CAS samples. Abbreviation: CAS, carotid atherosclerosis; AUC, the area under the curve; CI, confidence interval. Note: ***, P < 0.001; *, P < 0.05; ns, P > 0.05.
(A) Correlation between ACTC1, LDB3, MYOZ2, TPM2, and immune cell subtypes in CAS samples in the GSE43292 cohort. (B) Correlation between ACTC1, LDB3, MYOZ2, TPM2, and immune cell subtypes in CAS samples in the GSE28829 cohort. Note: ***, P < 0.001; **, P < 0.01; *, P < 0.05.
Discussion
At present, advances in microarray have facilitated the comprehensive analysis of mRNA expression profiles across the global genome. The GSE43292 dataset contained the RNA sequencing data of CAS vascular tissue and adjacent healthy vascular tissue pairs, whilst the GSE28829 comprised early and advanced CAS vascular tissue. These two sets of samples, including control vascular tissue, early diseased vascular tissue, and advanced diseased vascular tissue, enabled the characterization of changes in the immune microenvironment throughout disease progression to determine trends and statuses of the immune microenvironment during pathological transformation associated with the disease.
The results of this study unveiled that compared with healthy vascular tissue, immune cells were enriched in CSA samples throughout disease progression. Specifically, the level of immune cell infiltration was higher in advanced CAS samples compared to early CAS samples. Previous studies reported that various immune cell subtypes play a pivotal role in the pathogenesis of CAS [24–26]. Macrophages drive atherosclerosis and associated complications, playing distinct roles across all stages of atherosclerosis, and are also the most abundant type of inflammatory cells in plaques [27]. In addition to the high level of infiltration of macrophages, their plasticity and heterogeneity play a vital role in the immune response to atherosclerosis [28]. Indeed, macrophages can shift their phenotype according to their microenvironment to allow them to fulfill specific roles [28, 29]. For instance, M1 macrophages promote and maintain inflammatory responses and thus exert atherogenic effects [30]. In contrast, type 2 T helper cells stimulate the differentiation of M2 macrophages [31, 32], which express high levels of anti-inflammatory factors, balance the activity of M1 macrophages, alleviate inflammatory responses, initiate tissue repair, promote tissue remodeling and angiogenesis, and exert phagocytic effects to mediate the inflammatory response, thereby exerting anti-atherosclerotic effects [33, 34]. Herein, the level of type 2 T helper cells remained unchanged in both early and advanced CAS samples. Dendritic cells play an essential role in CAS by stimulating chemokine and cytokine secretion and participating in antigen presentation and lipid uptake, which may induce inflammation or promote immune tolerance [35–37]. B cells may be implicated in systemic and local immune responses to atherosclerosis by modulating cellular immune responses through intercellular contact, antigen presentation, and cytokine production [38, 39]. T cells are the second largest group of immune cells in carotid atherosclerotic plaques after macrophages. Th1 cells were the dominant form in atherosclerotic plaques, forming a mutually stimulating loop with macrophages conducing to disease progression [40, 41].
Furthermore, PCA demonstrated that the level of immune cell infiltration could be used to distinguish healthy vascular tissue from CAS samples, both early and late stages. The immune microenvironment is one of the essential characteristics of disease and can be used to describe different disease phenotypes. The enrichment of immune cells in vascular tissues was identified as a principal driving force behind pathological changes in diseases. Characterizing the immune microenvironment characteristics of CAS samples and exploring the enrichment mechanism of immune cell subtypes in CAS samples may potentially assist in elucidating disease heterogeneity and the molecular mechanism underlying disease progression. Hub genes that regulate the immune microenvironment or are significantly associated with the infiltration of immune cell subtypes may be potential diagnostic markers and therapeutic targets.
To delineate the heterogeneity of CAS samples and identify the enrichment mechanism of immune cells, CAS samples were divided into cluster A, hallmarked by a high infiltration level of immune cell subtypes, and cluster B, featuring a low infiltration level of immune cell subtypes via consensus clustering. In order to detect changes in the immune microenvironment, differentially expressed genes were identified between high immune infiltration clusters and healthy vascular tissues. Importantly, up-regulated genes in cluster A were significantly enriched in biological processes related to immune response, whereas down-regulated genes were significantly enriched in biological processes related to muscle development. The pathogenesis of atherosclerotic lesions has been hypothesized to be an exaggerated response to damage of endothelial cells and smooth muscle cells in the arterial wall [42, 43]. According to prior investigations, smooth muscle cell proliferation, smooth muscle cell synthesis, and secretion of connective tissue components are the critical pathological changes observed in atherosclerosis [44, 45]. The results of the enrichment analysis validated the representativeness of the selected DEGs and the accuracy of sample classification.
To identify hub genes that regulate or are significantly associated with the immune microenvironment, PPI networks were generated, and the MNC algorithm was applied to identify the highest-ranking hub genes, namely PTPRC and ACTN2. These genes were effective in disease diagnosis and differentiation between early and advanced stages of the disease. Specifically, PTPRC expression was positively correlated with most immune cell subtypes, with its expression level increasing with disease progression, implying that it promoted the development of the disease.
In comparison, ACTN2 expression was negatively correlated with most immune cell subtypes, and its expression level decreased with disease progression, suggesting that it impeded disease progression.
PTPRC is highly implicated in immune responses. It can promote T cell proliferation and differentiation, B cell proliferation, participate in T cell and B cell signal transduction pathways, positively regulate antigen receptor-mediated signal transduction pathways and protein kinases, release calcium ions into the cytoplasm, defend against responsive viruses, and regulate the cell cycle [46–48]. ACTN2 is highly expressed in the cytoskeleton and actin filaments and can bind with actin and calcium ions to participate in the formation of muscle structure. Besides, it is a key molecule that maintains the physiological morphology and function of muscles. Down-regulating the expression of ACTN2 may accelerate the loss of vascular smooth muscle function and elasticity. Nevertheless, studies examining the role of these two genes in carotid atherosclerosis are scarce, warranting further exploration [49, 50].
The relationship between ACTC1, LDB3, MYOZ2, TPM2, and CAS remains elusive. The expression levels of ACTC1, LDB3, MYOZ2, and TPM2 were negatively correlated with the infiltration level of most immune cell subtypes, positioning them as potential protective factors against CAS. In addition, while these genes are associated with muscle development and physiological processes, their functions in CAS remain to be elucidated.
It is worthwhile recognizing that previous studies have used the same dataset as ours and identified hub genes that, while partially overlapping, are not completely consistent. This may be ascribed to inconsistencies in the algorithms used. For example, Liu et al. identified RBM47, HCK, CD53, TYROBP, and HAVCR2 as hub genes in advanced atherosclerotic plaques via network-based analysis [51]. They first used WGCNA to screen key modules, then constructed protein interaction networks to screen hub genes, all of which were pathogenic genes. Our study initially distinguished the immune phenotypes of the samples using consensus clustering and subsequently screened differentially expressed genes using the maximum neighborhood component algorithm. These hub genes may play an instrumental role in the differentiation of immune phenotypes among samples. Herein, hub genes were screened from two perspectives of high and low immune cell infiltration, including potential pathogenic and protective genes.
However, some limitations of our study should not be overlooked. To begin, the screening models for diagnostic genes are diverse and complex, and the diagnostic efficacy of the screened genes remains unknown in other datasets. Thus, high-quality cohorts with sufficient sample sizes are necessary to validate our results. Secondly, regularization techniques and cross-validation were not utilized in the present study. Applying these methods through various machine learning algorithms can enhance the accuracy and reliability of results. Considering the limited number of genes screened using the maximum neighborhood component algorithm, some genes might have been validated by previous studies. Therefore, instead of further narrowing our scope with regularization techniques and cross-validation, the diagnostic efficacy of all unpublished genes in the downloaded datasets was validated.
Conclusion
The enrichment of immune cells in vascular tissues promoted pathological changes in CAS. Advanced CAS was characterized by high immune cell infiltration levels, whereas early CAS was hallmarked by low immune cell infiltration levels. Moreover, CAS progression may be related to the immune response pathway. Biological processes related to muscle cell development may delay CAS progression. Meanwhile, the expression of the hub genes CAS. PTPRC and ACTN2, ACTC1, LDB3, MYOZ2, and TPM2, identified in the interaction network of differentially expressed genes between cluster A and healthy vascular tissues, play an essential role in regulating the immune microenvironment of CAS and participate in its occurrence and progression. Finally, PTPRC and ACTN2, ACTC1, LDB3, MYOZ2, and TPM2 demonstrated favorable efficacy for distinguishing between high and low immune cell infiltration CAS samples, as well as for differentiating early from advanced CAS stages.
References
- 1. Agabiti-Rosei E, Muiesan M L. Carotid atherosclerosis, arterial stiffness and stroke events[J]. Adv Cardiol, 2007,44:173–186.
- 2. Demirkol S, Balta S, Celik T, et al. Carotid Intima Media Thickness and Its Association With Total Bilirubin Levels in Patients With Coronary Artery Ectasia[J]. Angiology, 2020,71(5):425–430.
- 3. Gui Y K, Li Q, Liu L, et al. Plasma levels of ceramides relate to ischemic stroke risk and clinical severity[J]. Brain Res Bull, 2020,158:122–127.
- 4. Dossabhoy S, Arya S. Epidemiology of atherosclerotic carotid artery disease[J]. Semin Vasc Surg, 2021,34(1):3–9.
- 5. Moroni F, Ammirati E, Magnoni M, et al. Carotid atherosclerosis, silent ischemic brain damage and brain atrophy: A systematic review and meta-analysis[J]. Int J Cardiol, 2016,223:681–687.
- 6. Chaturvedi S. Diagnosis and Management of Large Artery Atherosclerosis[J]. Continuum (Minneap Minn), 2023,29(2):486–500.
- 7. Spence J D, Azarpazhooh M R, Larsson S C, et al. Stroke Prevention in Older Adults: Recent Advances[J]. Stroke, 2020,51(12):3770–3777.
- 8. Heck D, Jost A. Carotid stenosis, stroke, and carotid artery revascularization[J]. Prog Cardiovasc Dis, 2021,65:49–54.
- 9. Arasu R, Arasu A, Muller J. Carotid artery stenosis: An approach to its diagnosis and management[J]. Aust J Gen Pract, 2021,50(11):821–825.
- 10. Dharmakidari S, Bhattacharya P, Chaturvedi S. Carotid Artery Stenosis: Medical Therapy, Surgery, and Stenting[J]. Curr Neurol Neurosci Rep, 2017,17(10):77.
- 11. Saba L, Nardi V, Cau R, et al. Carotid Artery Plaque Calcifications: Lessons From Histopathology to Diagnostic Imaging[J]. Stroke, 2022,53(1):290–297.
- 12. Jashari F, Ibrahimi P, Nicoll R, et al. Coronary and carotid atherosclerosis: similarities and differences[J]. Atherosclerosis, 2013,227(2):193–200.
- 13. Thapar A, Jenkins I H, Mehta A, et al. Diagnosis and management of carotid atherosclerosis[J]. BMJ, 2013,346: f1485.
- 14. Kubatova H, Poledne R, Pitha J. Immune cells in carotid artery plaques: what can we learn from endarterectomy specimens?[J]. Int Angiol, 2020,39(1):37–49.
- 15. Wijeratne T, Menon R, Sales C, et al. Carotid artery stenosis and inflammatory biomarkers: the role of inflammation-induced immunological responses affecting the vascular systems[J]. Ann Transl Med, 2020,8(19):1276.
- 16. Ji E, Lee S. Antibody-Based Therapeutics for Atherosclerosis and Cardiovascular Diseases[J]. Int J Mol Sci, 2021,22(11).
- 17. Ferrario C M, Strawn W B. Role of the renin-angiotensin-aldosterone system and proinflammatory mediators in cardiovascular disease[J]. Am J Cardiol, 2006,98(1):121–128.
- 18. Niyonzima N, Halvorsen B, Sporsheim B, et al. Complement activation by cholesterol crystals triggers a subsequent cytokine response[J]. Mol Immunol, 2017,84:43–50.
- 19. Ye J, Wang Y, Wang Z, et al. The Expression of IL-12 Family Members in Patients with Hypertension and Its Association with the Occurrence of Carotid Atherosclerosis[J]. Mediators Inflamm, 2020,2020:2369279.
- 20. Verma S K, Garikipati V N, Krishnamurthy P, et al. IL-10 Accelerates Re-Endothelialization and Inhibits Post-Injury Intimal Hyperplasia following Carotid Artery Denudation[J]. PLoS One, 2016,11(1): e147615.
- 21. Zeng Z, Guo R, Wang Z, et al. Circulating Monocytes Act as a Common Trigger for the Calcification Paradox of Osteoporosis and Carotid Atherosclerosis via TGFB1-SP1 and TNFSF10-NFKB1 Axis[J]. Front Endocrinol (Lausanne), 2022,13:944751.
- 22. Ayari H, Bricca G. Identification of two genes potentially associated in iron-heme homeostasis in human carotid plaque using microarray analysis[J]. J Biosci, 2013,38(2):311–315.
- 23. Doring Y, Manthey H D, Drechsler M, et al. Auto-antigenic protein-DNA complexes stimulate plasmacytoid dendritic cells to promote atherosclerosis[J]. Circulation, 2012,125(13):1673–1683.
- 24. Kotfis K, Biernawska J, Zegan-Baranska M, et al. Characteristics of peripheral immune cell subsets in patients with carotid atherosclerosis undergoing carotid endarterectomy[J]. Arch Med Sci Atheroscler Dis, 2018,3: e129–e136.
- 25. Gao J, Shi L, Gu J, et al. Difference of immune cell infiltration between stable and unstable carotid artery atherosclerosis[J]. J Cell Mol Med, 2021,25(23):10973–10979.
- 26. Wang L, Gao B, Wu M, et al. Profiles of Immune Cell Infiltration in Carotid Artery Atherosclerosis Based on Gene Expression Data[J]. Front Immunol, 2021,12:599512.
- 27. Chou E L, Lino C C, Chaffin M, et al. Vascular smooth muscle cell phenotype switching in carotid atherosclerosis[J]. JVS Vasc Sci, 2022,3:41–47.
- 28. de Gaetano M, Crean D, Barry M, et al. M1- and M2-Type Macrophage Responses Are Predictive of Adverse Outcomes in Human Atherosclerosis[J]. Front Immunol, 2016,7:275.
- 29. Wolf D, Ley K. Immunity and Inflammation in Atherosclerosis[J]. Circ Res, 2019,124(2):315–327.
- 30. Carbone F, Satta N, Burger F, et al. Vitamin D receptor is expressed within human carotid plaques and correlates with pro-inflammatory M1 macrophages[J]. Vascul Pharmacol, 2016,85:57–65.
- 31. Kimura S, Noguchi H, Nanbu U, et al. Relationship between CCL22 Expression by Vascular Smooth Muscle Cells and Macrophage Histamine Receptors in Atherosclerosis[J]. J Atheroscler Thromb, 2018,25(12):1240–1254.
- 32. Barros M H, Hauck F, Dreyer J H, et al. Macrophage polarisation: an immunohistochemical approach for identifying M1 and M2 macrophages[J]. PLoS One, 2013,8(11): e80908.
- 33. Burger F, Baptista D, Roth A, et al. NLRP3 Inflammasome Activation Controls Vascular Smooth Muscle Cells Phenotypic Switch in Atherosclerosis[J]. Int J Mol Sci, 2021,23(1).
- 34. Jing Y, Gao B, Han Z, et al. HOXA5 induces M2 macrophage polarization to attenuate carotid atherosclerosis by activating MED1[J]. IUBMB Life, 2021,73(9):1142–1152.
- 35. Zernecke A, Erhard F, Weinberger T, et al. Integrated single-cell analysis based classification of vascular mononuclear phagocytes in mouse and human atherosclerosis[J]. Cardiovasc Res, 2022.
- 36. Wang Y K, Wang J, Hua F, et al. TREM-1 Modulates Dendritic Cells Maturation and Dendritic Cell-Mediated T-Cell Activation Induced by ox-LDL[J]. Oxid Med Cell Longev, 2022,2022:3951686.
- 37. Bacci S, Pieri L, Buccoliero A M, et al. Smooth muscle cells, dendritic cells and mast cells are sources of TNFalpha and nitric oxide in human carotid artery atherosclerosis[J]. Thromb Res, 2008,122(5):657–667.
- 38. Kojima Y, Volkmer J P, McKenna K, et al. CD47-blocking antibodies restore phagocytosis and prevent atherosclerosis[J]. Nature, 2016,536(7614):86–90.
- 39. Douna H, de Mol J, Amersfoort J, et al. IFNgamma-Stimulated B Cells Inhibit T Follicular Helper Cells and Protect Against Atherosclerosis[J]. Front Cardiovasc Med, 2022,9:781436. pmid:35187121
- 40. Chaudhari S M, Sluimer J C, Koch M, et al. Deficiency of HIF1alpha in Antigen-Presenting Cells Aggravates Atherosclerosis and Type 1 T-Helper Cell Responses in Mice[J]. Arterioscler Thromb Vasc Biol, 2015,35(11):2316–2325. pmid:26404487
- 41. Smirnova N F, Gayral S, Pedros C, et al. Targeting PI3Kgamma activity decreases vascular trauma-induced intimal hyperplasia through modulation of the Th1 response[J]. J Exp Med, 2014,211(9):1779–1792.
- 42. Shi J, Yang Y, Cheng A, et al. Metabolism of vascular smooth muscle cells in vascular diseases[J]. Am J Physiol Heart Circ Physiol, 2020,319(3): H613–H631.
- 43. Sorokin V, Vickneson K, Kofidis T, et al. Role of Vascular Smooth Muscle Cell Plasticity and Interactions in Vessel Wall Inflammation[J]. Front Immunol, 2020,11:599415.
- 44. Durham A L, Speer M Y, Scatena M, et al. Role of smooth muscle cells in vascular calcification: implications in atherosclerosis and arterial stiffness[J]. Cardiovasc Res, 2018,114(4):590–600. pmid:29514202
- 45. Bennett M R, Sinha S, Owens G K. Vascular Smooth Muscle Cells in Atherosclerosis[J]. Circ Res, 2016,118(4):692–702.
- 46. Wei J, Fang D, Zhou W. CCR2 and PTPRC are regulators of tumor microenvironment and potential prognostic biomarkers of lung adenocarcinoma[J]. Ann Transl Med, 2021,9(18):1419.
- 47. Al B M, Ali A, McMullin M F, et al. Protein tyrosine phosphatase receptor type C (PTPRC or CD45)[J]. J Clin Pathol, 2021,74(9):548–552.
- 48. Guo G, Li B, Li Q, et al. PTPRC Overexpression Predicts Poor Prognosis and Correlates with Immune Cell Infiltration in Pediatric Acute Myeloid Leukemia[J]. Clin Lab, 2022,68(7).
- 49. Ranta-Aho J, Olive M, Vandroux M, et al. Mutation update for the ACTN2 gene[J]. Hum Mutat, 2022,43(12):1745–1756.
- 50. Lek M, Quinlan K G, North K N. The evolution of skeletal muscle performance: gene duplication and divergence of human sarcomeric alpha-actinins[J]. Bioessays, 2010,32(1):17–25.
- 51. Liu C, Zhang H, Chen Y, et al. Identifying RBM47, HCK, CD53, TYROBP, and HAVCR2 as Hub Genes in Advanced Atherosclerotic Plaques by Network-Based Analysis and Validation[J]. Front Genet, 2020,11:602908.