Diagnostic Value of Circulating Progranulin and Its Receptor EphA2 in Predicting the Atheroma Burden in Patients with Coronary Artery Disease

Background Progranulin (PGRN) and its potential receptor Eph-receptor tyrosine kinase-type A2 (EphA2) are inflammation-related molecules that present on the atherosclerotic plaques. However, the roles of circulating PGRN and EphA2 in coronary artery disease (CAD) remain unclear. Objective To study the clinical significance of circulating PGRN and EphA2 levels in Chinese patients undergoing coronary angiography. Methods Levels of circulating EphA2 fragments and PGRN were examined in 201 consecutive individuals who underwent coronary angiography for suspected CAD in our center from Jan 2020 to Oct 2020. Demographic characteristics, results of biochemical and auxiliary examinations, and other relevant information were collected. The coronary atheroma burden was quantified by the Gensini score and the existence of chronic total occlusion (CTO). Univariate analysis and multivariate logistic regression analysis were used to analyze the risk factors for acute coronary syndrome (ACS). In patients with ACS and SAP, a receiver operating characteristic (ROC) curve was generated to detect the accuracy and discriminative ability of levels of EphA2 and PGRN, the Gensini score, and cardiac injury biomarkers as surrogate endpoints for CTO. Results Circulating EphA2 levels were significantly higher in patients with ACS than in subjects with stable angina pectoris (SAP) or control subjects (p < 0.001). A positive linear correlation was verified between EphA2 levels and the Gensini score (r = 0.306, p < 0.001), and negative correlation was detected with the left ventricular ejection fraction (LVEF) (r = −0.405, p < 0.001). Both PGRN and EphA2 were positively associated with cardiac injury biomarkers (i.e., NT-proBNP, cTnT, and hs-CRP) (p < 0.05). The area under the ROC curve of PGRN and EphA2 was 0.604 and 0.686, respectively (p < 0.01). Conclusions Higher circulating EphA2 and PGRN levels were detected in patients with ACS than in patients with SAP. Circulating EphA2 and PGRN levels might be diagnostic factors for predicting the atheroma burden in patients with CAD.


Introduction
Coronary artery disease (CAD) remains the leading global cause of morbidity and mortality. Acute coronary syndrome (ACS) is an acute and dangerous type of CAD and the primary manifestation of atherosclerotic progression in the coronary artery, which is closely related to endothelial dysfunction, inflammation, and coagulation [1]. Rupture of atherosclerotic plaque and persistent thrombotic vessel occlusion at the site of plaque rupture have dominated our thinking on ACS pathophysiology. Shan et al. found that an increased baseline atheroma burden independently predicts nonculprit lesion-related major adverse cardiovascular events (MACEs) in patients with ACS after a successful culprit lesion intervention [2]. Thus, early prediction of the atheroma burden in patients with CAD plays an essential role in determining the patient prognosis. Several cardiac injury biomarkers are promising markers for atheroma burden and have been reported to predict MACEs, such as N-terminal pro B-type natriuretic peptide (NT-proBNP), cardiac troponin (cTn), and high-sensitivity C-reactive protein (hs-CRP) [3,4].
Progranulin (PGRN) is a secreted glycoprotein encoded by the GRN gene that is implicated in multiple pathological processes such as regulation of inflammation, promotion of proliferation, mediation of cell cycle progression, cell motility, and neurotropic and lysosome regulation [5,6]. Studies have found that plasma PGRN levels are related to insulin resistance and inflammatory regulation [7,8]. Kojima et al. first detected the expression of PGRN in atherosclerotic plaques [9]. In animal models, PGRN deficiency leads to severe atherosclerotic lesions and increased age-related myocardial hypertrophy [10,11].
Eph-receptor tyrosine kinase-type A2 (EphA2) is also a promising cardiovascular molecule involved in the regulation of cell-cell interactions and angiogenesis [12]. According to recent studies, EphA2 (-/-) ApoE (-/-) mice show diminished atherosclerotic plaque formation and reduced proinflammatory gene expression compared to ApoE (-/-) controls, which identifies an unrecognizable role of EphA2 in regulating both plaque inflammation and progression to advanced atherosclerotic lesions [12]. Additionally, EphA2 deficiency could exacerbate the myocardial injury and the progression of ischemic cardiomyopathy [13].
Interestingly, EphA2 is a functional receptor for PGRN that has been validated in vitro, and EphA2 silencing significantly prevents PGRN-mediated autoregulation [14,15]. Moreover, PGRN degradation into granulin peptides might exacerbate inflammation in atherosclerotic values, which contributes to the progression of atherosclerosis [16]. How-ever, the clinical roles of PGRN and EphA2 in patients with CAD remain unclear. The study sought to identify the clinical significance of EphA2 and PGRN in patients with CAD.

Study Population.
The study recruited adult patients scheduled for coronary angiography at the Department of Cardiology in Zhongshan Hospital Fudan University between Jan 2020 and Oct 2020. Patients were included if they were between 40 and 85 y of age. The exclusion criteria were severe infectious diseases, severe liver or kidney insufficiency, and malignant diseases. All participants provided written informed consent, and the experimental scheme was approved by the Zhongshan Hospital Institutional Ethics Committee. The study was performed in accordance with the Declaration of Helsinki.

Sample Collection and Measurement.
Preangiography blood samples (4 mL) were obtained from the right jugular vein of patients upon admission and rapidly centrifuged at 3,000 rpm for 10 min at 4°C. Finally, approximately 2 mL of supernatants was transferred to RNase-free tubes and stored at a temperature of -80°C until subsequent assays. In all patients, the levels of EphA2 and PGRN were measured using commercial high-sensitivity enzyme-linked immunosorbent   , and hs-CRP, were tested by the laboratory of Zhongshan Hospital Affiliated to Fudan University using an automatic biochemistry analyzer or automatic electrochemiluminescence immunoassay analyzer.

Groups and Assessment of the Atheroma Burden.
The results were compared between the control group, the stable angina pectoris (SAP) group, and the ACS group. The control group was defined as coronary angiography showing patency or significant stenosis of the major coronary artery (including the left anterior descending artery, circumflex artery, and right coronary artery) of less than 50%. In the case of SAP, the angina symptom should have been stable for at least 6 months with normal myocardial enzyme levels at admission and ≥50% luminal narrowing in at least one major coronary artery. ACS was defined as patients with non-STsegment elevation myocardial infarction (NSTEMI) or STsegment elevation myocardial infarction (STEMI) and unstable angina (UA) [17]. The Gensini scores were calculated to identify the severity and complexity of coronary atherosclerotic lesions. The Gensini score equals the sum of all segment scores (each segment score is equal to the segment weighting factor multiplied by a severity score) [18]. The segment weighting factors were 0.5, 1, 1.5, 2.5, and 5.0, according to the vascular site. Severity scores were used to reflect the specific percentage of luminal stenosis by layering 32, 16, 8, 4, 2, and 1, for 100%, 99%, 90%, 75%, 50%, and 25%, respectively. Chronic total occlusion (CTO) was defined as anterior flow TIMI = 0 in the occluded segment, and the occlusion time was at least 3 months.
2.5. Statistical Analysis. Continuous data are presented as the mean and standard deviation (SD) or median (interquartile ranges). The Shapiro-Wilk test was performed to evaluate normality. Differences between groups were calculated using the one-way ANOVA or the Kruskal-Wallis test. For multiple comparisons between groups, Tukey's HSD post hoc test and the Wilcoxon rank-sum test were used. Categorical variables are presented as numbers and percentages and were compared using the chi-square test or Fisher's exact test. Spearman's rank correlation coefficients were calculated to analyze the correlation between two variables. Univariate analysis and multivariate logistic regression analysis were performed to analyze the risk factors for ACS. The forward Wald method was used to analyze the risk factors. The results are presented as odds ratios (ORs) and 95% confidence intervals (95% CIs). In patients with ACS and SAP, a receiver operating characteristic (ROC) curve was generated to detect the accuracy and discriminative ability of levels of EphA2 and PGRN, the Gensini score, and cardiac injury biomarkers as surrogate endpoints for CTO. All statistics were calculated using bilateral tests; p < 0:05 was considered statistically significant. All statistical analyses were performed with SPSS software (IBM SPSS Statistics 22.0).

Results
3.1. Baseline Characteristics of Subjects in Each Group. Two hundred one patients were enrolled in the study. The baseline characteristics of the study subjects are presented in Table 1. No significant differences in demographic characteristics or risk factors were observed between groups (p > 0:05). However, patients with ACS showed significantly increased lowdensity lipoprotein (LDL) and hemoglobin A1c levels compared to control subjects (p < 0:05). In particular, the levels of hs-CRP, NT-proBNP, cTnT, and CK-MB were elevated in patients with ACS compared with patients with SAP or control subjects (p < 0:001). Circulating EphA2 levels, the Gensini score, and left ventricular ejection fraction (LVEF) were significantly higher in patients with ACS than in patients with SAP and control subjects (p < 0:001) (Figures 1(a), 1(c), and 1(d)). Higher PGRN levels were detected in the ACS group than in patients with SAP (p < 0:05) (Figure 1(b)).

Features of Circulating EphA2 and PGRN Levels in
Patients with ACS or SAP. One hundred fifty-nine patients were diagnosed with ACS or SAP. The median Gensini score was 41.0. All patients were divided into a group with a lower atheroma burden (patients with a Gensini score < 41:0, n = 80) and greater atheroma burden (patients with a Gensini NT-proBNP, cTnT, CK-MB, and hs-CRP levels were higher in the CTO group than in the non-CTO group (p < 0:05).

Diagnostic Efficiency of Circulating EphA2 and PGRN
Levels to Predict CTO. We analyzed the diagnostic efficiency of circulating EphA2 and PGRN levels, the Gensini score, and cardiac injury biomarkers (NT-proBNP, cTnT, CK-MB, and hs-CRP) in predicting CTO in patients with ACS or SAP. As shown in Figure 3, the Gensini score showed a good discriminative ability to predict the presence of CTO (AUC of 0.911, 95% CI 0.864-0.958). Levels of EphA2, PGRN, and cardiac injury biomarkers significantly predicted the presence of CTO with low to moderate diagnostic efficiency (AUC > 0:6) ( Figure 3 and Table 5). The optimal cutoff for EphA2 was 63.6 pg/mL (sensitivity at the optimal cutoff of 89.1% and specificity at the optimal cutoff of 46.3%). Moreover, the optimal cutoff for PGRN was 38.2 ng/mL (sensitivity at the optimal cutoff of 78.3% and specificity at the optimal cutoff of 44.3%) ( Table 5).

Discussion
Our study first determined the features of circulating PGRN and EphA2 levels in Chinese patients with CAD and analyzed the association between PGRN and EphA2 levels and the atheroma burden in patients undergoing coronary angiography. Collectively, we reported three important findings [1]. Circulating EphA2 and PGRN levels are significantly

Disease Markers
higher in patients with ACS than in patients with SAP [2]. Levels of soluble N-terminal EphA2 fragments in blood were correlated with PGRN concentrations, the Gensini score, and the cardiac injury biomarkers in study subjects [3]. EphA2 and PGRN levels could statistically predict the presence of CTO in patients with CAD.    To our knowledge, increased PGRN levels might be novel biomarkers of chronic inflammatory diseases, such as chronic periodontitis and community-acquired pneumonia [20,21]. Moreover, serum PGRN levels are reported to be associated with systemic inflammatory markers [22]. According to previous studies, mouse models deficient in PGRN and ApoE exhibit more severe atherosclerotic lesions than PGRN (+/+) ApoE (-/-) mice [11]. However, the exact role of PGRN in atherosclerosis remains unclear [23].
In our study, patients with ACS had higher Gensini scores than non-ACS subjects (p < 0:001). We identified significant associations between PGRN levels, LVEF, CTO, and the levels of cardiac markers. Unlike the modest negative correlation between PGRN and high-density lipoprotein levels in Korean adults with CAD [23], we found that PGRN levels were not correlated with indices of glucose and lipid metabolism.
EphA2 is a member of the Eph receptor kinase family. Remarkably, the EphA2 gene is located in the region of human chromosome 1 (1p36) 16 associated with early myocardial infarction and the region of chromosome 4 associated with increased susceptibility to atherosclerosis (athsq1 locus) in mice [24,25]. Circulating EphA2 is an effective biomarker to diagnose malignant tumors [26]. Recent studies have revealed the potential role of EphA2 in regulating plaque inflammation and progression [27]. Therefore, we detected levels of the soluble EphA2 fragment in the blood of patients with CAD. As expected, circulating EphA2 concentrations  were positively correlated with the levels of cardiac injury markers (i.e., cTnT, NT-proBNP, hs-CRP, and CK series). Additionally, EphA2 levels were positively correlated with the Gensini score (r = 0:306, p < 0:001) but negatively correlated with the LVEF (r = −0:405, p < 0:001), suggesting that EphA2 may be a potential diagnostic biomarker for CAD.
Considering the potential significance of EphA2 and PGRN in inflammatory regulation and the progression of atherosclerosis formation, we applied a correlation analysis to assess the relationship between EphA2 and PGRN levels and found that EphA2 levels were moderately positively correlated with PGRN levels in all study subjects (r = 0:407, p < 0:001). In the prediction of ACS, we divided EphA2 levels into two groups based on a median of 116.32 pg/mL and included it in the multivariate logistic regression analysis with sex, levels of LDL, hemoglobinA1, and NT-proBNP, the Gensini score, and other factors. The OR value for EphA2 levels ≥ 116:32 pg/mL was 10.715 (95% CI: 3.259-35.227, p < 0:001).
Finally, we observed higher levels of circulating EphA2 and cardiac injury markers in patients with a greater atheroma burden than in those with a lower atheroma burden grouped by either the Gensini score or the existence of CTO. In addition, levels of EphA2 and PGRN significantly predicted CTO (AUC > 0:6, p < 0:05), suggesting the potential diagnostic efficiency of circulating EphA2 and PGRN levels in evaluating the atheroma burden in patients with CAD, which further confirms the roles of EphA2 and PGRN in the progression of atherosclerosis.
However, our study has some limitations. This study represented a small observational finding from a single center. Second, temporal changes (such as three months or 12 months) in EphA2 and PGRN levels were not considered before the study, which would make the conclusion more convincing. Finally, the present study illustrated but did not test the phenomenon of the synergistic mechanism of EphA2 and PGRN. More detailed study designs are needed to explore the specific molecular mechanism of the PGRN/E-phA2 axis in CAD.

Conclusions
In summary, circulating EphA2 levels are significantly higher in patients with ACS. Our findings suggested the diagnostic value of circulating PGRN and EphA2 levels in predicting the atheroma burden in patients with CAD.