Unraveling the proteomic signatures of coronary artery disease and hypercholesterolemia

Abstract

Atherosclerosis, a leading cause of coronary artery disease (CAD), is heavily influenced by hypercholesterolemia (HC). Proteomics research has shown promise in identifying biological markers for CAD diagnosis and prognosis. This cross-sectional study aimed to identify novel biomarkers for detecting HC and CAD. Through the analysis of proteome data from healthy controls (n = 45) and patients diagnosed with HC (n = 51) or CAD (n = 32), distinct protein patterns associated with each condition were identified. Significant alterations in protein levels were identified with a false discovery rate (FDR)-corrected q-value of <0.05. Subsequent receiver operating characteristic (ROC) analysis, with an area under the curve (AUC) greater than 0.75, was conducted. CAD patients exhibited significantly increased levels of the cholesterol-metabolizing protein proprotein convertase subtilisin/kexin type 9 (PCSK9) and varied levels of the angiogenesis-related protein stromal-cell-derived factor-1 (SDF-1) compared to controls. In pairwise comparisons among the study groups, 65 proteins showed significant differential expression. Notably, 14 of these proteins had significant correlations with blood cholesterol levels. Additionally, 22 of the identified proteins were associated with CAD or HC pathways, with nine proteins being common to both conditions (APO E, APO E3, MMP-3, PCSK9, SDF-1, APO B, PAFAH, 60 kDa heat shock protein, and TGF-beta-activated kinase 1 and MAP3K7-binding protein 1 fusion). Nevertheless, this is an exploratory study, and validation studies are needed to confirm these potential protein biomarkers for CAD in the context of HC.

Figure 1.

Separation of study groups and differential protein expressions among Con, HC, and CAD groups. (A) The graph illustrates the OPLS-DA separation of Con (blue), HC (green), and CAD (red); (B) The diagram summarizes the identification of significant proteins with statistical analyses; (C–E) Volcano plots depict differentially expressed proteins in multiple comparisons of CAD vs Con (C), HC vs Con (D), and CAD vs HC (E). Log₂ FC indicates a fold increase (+) or decrease (−) between groups. Proteins with increased expressions in the CAD group (red), increased expressions in the HC group (green), and upregulated proteins in the control group (blue) are shown. Con: Control; HC: Hypercholesterolemia; CAD: Coronary artery disease; OPLS-DA: Orthogonal partial least square discriminant analysis; FC: Fold change.

Figure 2.

Significantly differentially expressed proteins and their diagnostic values in Con, HC, and CAD groups. Box plots represent differential expressions of proteins, and ROC curves show AUC values of differentially expressed proteins in pairwise comparisons of (A–D) CAD vs Con, (E–L) HC vs Con, or (M–AP) CAD vs HC. In CAD vs Con comparison, box plots of proteins are as in (A) SDF-1, (C) PCSK9. ROC curves are in B and D, respectively. In HC vs Con comparison, box plots of proteins are as in (E) QORL1, (G) LRP1B, (I) Transferrin, and (K) APO B. ROC curves are in F, H, J, and L, respectively. In CAD vs HC comparison, box plots of proteins are as in (M) APO B, (O) SHC1, (Q) VAV, (S) LRP1B, (U) FYN, (W) WNK3, (Y) PDPK1, (AA) FER, (AC) CAMK2B, (AE) PCSK9, (AG) PAFAH, (AI) PDE3A, (AK) HSP 60, (AM) TAK1-TAB1, and (AO) CSRP3. ROC curves are N, P, R, T, V, X, Z, AB, AD, AF, AH, AJ, AL, AN, and AP, respectively. *P < 0.05, ** P < 0.01, *** P < 0.001 **** P < 0.0001. All proteins in represented comparisons also have significant q values (<0.05). Con: Control; HC: Hypercholesterolemia; CAD: Coronary artery disease; ROC: Receiver operating characteristic; AUC: Area under the curve; SDF-1: Stromal-cell-derived factor-1; PCSK9: Proprotein convertase subtilisin/kexin type 9; LRP1B: LDL receptor-related protein 1B; VAV: Proto-oncogene vav; PDE3A: Phosphodiesterase 3A; HSP 60: 60 kDa heat shock protein; TAK1-TAB1: TGF-beta-activated kinase 1 and MAP3K7-binding protein 1 fusion; CSRP3: Cysteine and glycine-rich protein 3; QORL1: Quinone oxidoreductase-like protein 1; APO B: APOlipoprotein B; SHC1: SHC-transforming protein 1; FYN: Tyrosine-protein kinase Fyn; WNK3: Serine/threonine-protein kinase WNK3; CAMK2B: Calcium/calmodulin-dependent protein kinase type II subunit beta.

Figure 3.

Positive or negative correlation between protein levels and lipid profiles. The heat map represents the r-values of each Pearson correlation analysis between lipid parameters and proteomes. r values are between +1 and −1. Statistically significant correlations were *P < 0.05, **P < 0.01, ***P < 0.001 ****P < 0.0001.

Figure 4.

Protein expression patterns and related pathways. The figure shows significant protein up or down regulations among the study groups and their relation to disease pathways.