Screening of potential regulatory genes in carotid atherosclerosis vascular immune microenvironment

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.

Fig 1. Difference in the immune microenvironment between CAS and control samples in the GSE43292 cohort.

(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.

Fig 2. Differences in the immune microenvironment between early and advanced carotid atherosclerosis samples in the GSE28829 cohort.

(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.

Fig 3. High and low immune infiltration subtypes of CAS samples in the GSE43292 cohort distinguished via consensus clustering.

(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.

Fig 4. Gene set variation analysis (GSVA) between cluster A, cluster B, and the control group.

(A) GSVA between cluster B and the control group. (B) GSVA between cluster A and the control group.

Fig 5. Enrichment analysis and hub gene screening.

(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.

Fig 6. Diagnostic efficiency evaluation of PTPRC and ACTN2 in the GSE43292 cohort.

(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.

Fig 7. Diagnostic efficiency validation of PTPRC and ACTN2 in the GSE28829 cohort.

(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.

Fig 8. Correlation between PTPRC, ACTN2, and immune cell subtypes.

(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.

Fig 9. Diagnostic efficiency evaluation of hub genes in the GSE43292 cohort.

(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.

Fig 10. Diagnostic efficiency validation of ACTC1, LDB3, MYOZ2, and TPM2 in the GSE28829 cohort.

(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.

Fig 11. Correlations between ACTC1, LDB3, MYOZ2, TPM2, and immune cell subtypes.

(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.