Plasma Levels of Bactericidal/Permeability-Increasing Protein Correlate with Systemic Inammation in Acute Coronary Syndrome

Background: Neutrophils play important roles in atherosclerosis and atherothrombosis. Bactericidal/permeability-increasing protein (BPI) is mainly expressed in the granules of human neutrophils in response to inammatory stress. This observational, cross-sectional study investigated the plasma level of BPI in patients with acute coronary syndrome (ACS) and its correlation with blood neutrophil counts and circulating inammatory biomarkers. Methods: A total of 367 patients who had acute chest pain and who were admitted to our hospital for coronary angiography (CAG) and/or percutaneous coronary intervention (PCI) from May 1, 2020 to August 31, 2020 were recruited. Among them, 256 had a cardiac troponin value above the 99th percentile upper reference limit and were diagnosed with ACS. The remaining patients (n = 111) were classied as non-ACS. The TIMI and GRACE scores were calculated at admission. The Gensini score based on CAG was used to determine atherosclerotic burden. Plasma levels of interleukin (IL)-1β, myeloperoxidase-DNA (MPO-DNA), high sensitivity C-reactive protein (hs-CRP), S100A8/A9, and BPI were measured using enzyme-linked immunosorbent assays. Correlations of plasma BPI levels with examination scores and levels of circulating inammatory biomarkers were explored. Receiver operating characteristic (ROC) curve analysis was used to determine the diagnostic ecacy of BPI for ACS and myocardial infarction. Results: Patients in the ACS group showed signicantly higher plasma BPI levels compared to the non-ACS group (46.42 ± 16.61 vs. 16.23 ± 13.19 ng/mL, p < 0.05). Plasma levels of IL-1β, MPO-DNA, hs-CRP, and S100A8/A9 in the ACS group were also signicantly higher than those in the non-ACS group (all p < 0.05). In addition, plasma BPI levels were positively correlated with the TIMI, GRACE, and Gensini scores (r = 0.176, p = 0.003; r = 0.320, p < 0.001; r = 0.263, p < 0.001, respectively) in patients with ACS. Plasma BPI levels were also positively correlated with blood neutrophil counts (r = 0.266, p < 0.001) and levels of circulating inammatory biomarkers (IL-1β, r = 0.512; MPO-DNA, r = 0.452; hs-CRP, r = 0.554; S100A8/A9, r = 0.434; all p < 0.001) in patients with ACS. ROC curve analysis revealed that the diagnostic ecacy of BPI for ACS was not inferior to that of IL-1β, MPO-DNA, hs-CRP, S100A8/A9, or blood neutrophil counts. ROC analysis also showed that the diagnostic ecacy of BPI for myocardial infarction was not inferior to that of creatine kinase (CK)-MB or cardiac troponin I. Conclusion: BPI is associated with systemic inammation in ACS and may be involved in the process of atherosclerosis and atherothrombosis. The potential of BPI as a prognostic and diagnostic biomarker for ACS should be investigated in clinical settings.


Introduction
Advances in the sensitivity and precision of the cardiac troponin assay in recent years has improved the diagnostic accuracy of acute chest pain [1,2]. However, a considerable proportion of patients fall in an ambiguous category due to myocardial injury or the absence of coronary plaque rupture/erosion/dissection following coronary angiography (CAG) [2,3]. Thus, identi cation of novel biomarkers for differentiating these indeterminate patients is of great importance for the early diagnosis of acute coronary syndrome (ACS).
ACS is associated with unstable coronary atherosclerotic plaques (i.e. plaques with rupture and erosion), which may indicate exacerbation of vascular in ammation in both adaptive and innate immunity [4,5].
The roles of these circulating in ammatory biomarkers in ACS have been explored in recent years.
Bactericidal/permeability-increasing protein (BPI) is a pluripotent protein (~ 55 kDa) mainly expressed in the granules of human neutrophils in response to in ammatory stress. BPI plays a key role in the host defense against Gram-negative bacteria [20,21]. In addition to its bactericidal activity, BPI has been identi ed as a target of anti-neutrophil cytoplasmic antibody (ANCA) in a variety of diseases, such as cystic brosis [22], in ammatory bowel disease (IBD) [23,24], reactive arthritis [25], chronic obstructive pulmonary disease (COPD) [26,27], and vasculitis [28,29]. BPI participates in lipid metabolism as its structure is similar to some lipid metabolic proteins [30]. It is also closely related to diabetes, an independent risk factor of atherosclerotic disease. A previous study revealed that plasma BPI levels were signi cantly associated with insulin sensitivity, glucose metabolism, central obesity, and components of metabolic syndrome in patients with glucose intolerance [31]. It was also found that plasma BPI concentrations positively correlated with levels of high-density lipoprotein cholesterol (HDL-C), total cholesterol (TC), and low-density lipoprotein cholesterol (LDL-C), as well as endothelium-dependent vasodilatation [32]. A proteomics study reported that the BPI level was markedly decreased in patients with total coronary atherosclerotic occlusion, indicating that BPI may be a biomarker for severe atherosclerotic coronary stenosis [33]. BPI has also been shown to inhibit angiogenesis by promoting apoptosis of vascular endothelial cells both in vitro and in vivo [34]. Angiogenesis is an important event in the pathogenesis of coronary heart disease (CHD), especially myocardial infarction (MI). BPI may be a prognostic marker and a therapeutic target of CHD.
Based on the evidence in previous literature, we hypothesized that BPI may play a regulatory role in atherosclerosis-related in ammatory responses and atherothrombosis. In the present observational study, we rst con rmed the in ammatory characteristics of BPI by determining if plasma BPI levels correlated with levels of hs-CRP, IL-1β, MPO-DNA, and S100A8/A9 in patients with ACS. We then evaluated the association between plasma BPI levels and coronary atherosclerotic burden using the Gensini score.
Lastly, we compared the diagnostic e cacy of BPI with other circulating in ammatory biomarkers for ACS and MI, and determined the prognostic value of BPI based on the TIMI and GRACE scores. The ndings presented here highlight the potential of BPI as a diagnostic and prognostic marker for ACS.

Study population
Subjects who had acute chest pain and were clinically suspected of having ACS and who were admitted to our hospital for coronary angiography (CAG) and/or percutaneous coronary intervention (PCI) between May 1, 2020, and August 31, 2020 were screened. Patients who had one of the following characteristics were excluded: cardiogenic shock, acute infection, blood disorders, severe hepatic, or severe renal failure (plasma creatine level > 3 mg/dl), as well as ongoing chronic in ammatory, autoimmune, or malignant diseases. Patients who had acute chest pain and a cardiac troponin I value above the 99th percentile upper reference limit were classi ed as ACS. The remaining patients were classi ed as non-ACS. The diagnosis of MI was in accordance with the criteria of the ESC/ACCF/AHA/WHF Fourth Universal De nition of Myocardial Infarction (2018) [35]. All patients had clinical indications for CAG. Intravascular ultrasound (IVUS) was used to identify coronary plaque rupture/erosion/dissection in ACS or non-ACS patients whose cardiac troponin I values were above the 99th percentile upper reference limit in the following test. In total, 256 ACS patients and 111 non-ACS patients were included in this study. All patients ranged in age between 30 and 87 years. Among the ACS patients, 159 were diagnosed with MI based on the CAG results, dynamic changes in cardiac troponin I levels, and the IVUS results. Their demographic and clinical records (i.e. age, gender, BMI, smoking history, hyperlipidemia, diabetes mellitus, atrial brillation, hypertension, stroke, and medications), as well as biochemical and hematological data (i.e. glucose, creatinine, uric acid, lipid pro le, creatine kinase isoenzyme, platelets, neutrophils, leukocytes, etc.), were recorded for statistical analysis. BMI was calculated as weight (kg) divided by height (cm) in squared meters. This study was approved by the Ethics Committee of The First A liated Hospital of Bengbu Medical College (Bengbu, China). All participants gave written informed consent.
CAG and assessment of the severity of coronary atherosclerotic burden CAG was performed by experienced cardiologists according to the conventional Judkins method. Rapamycin-coated stents and radial artery puncture were used for all patients. During the procedure, 1 mL of 1% lidocaine was used for local anesthesia. After the artery sheath was implanted, 3,000 U heparin and 200 µg nitroglycerin were injected through the sheath. The CAG results were interpreted by experienced physicians following current guidelines [36]. The decision regarding subsequent performance of PCI was made by the same physicians according to current PCI guidelines [36].
Periprocedural medication and management were administered in accordance with current PCI guidelines [37]. Dual antiplatelet therapies and statins were administered to all patients before the procedure. The criteria for successful PCI were in accordance with international practice guidelines: TIMI grade 3 blood ow and residual stenosis ≤ 20% after the procedure. The Gensini score was used to quantitatively assess the severity of coronary atherosclerotic burden according to the distributions and degree of stenosis shown in CAG [38].

Blood Samples
Blood samples (5 mL) were collected from each subject through the median cubital vein and tested within 30 mins after admission. These samples were collected in tubes with heparin sodium or EDTA. Samples with heparin sodium were used for a series of blood tests following a standard protocol from the Department of Clinical Laboratory at our hospital, which included routine blood analysis, blood glucose, coagulation function, lipid pro le, uric acid, renal function, creatine kinase isoenzyme, and cardiac troponin I. Plasma samples were collected by centrifugation at 2,500 × g for 15 min and stored at -80˚C until subsequent analysis. Additional morning fasting blood samples were collected within 24 h of admission for monitoring fasting blood glucose, creatine kinase isoenzyme, and cardiac troponin I.

Enzyme-linked Immunosorbent Assay (Elisa)
The MPO-bound DNA complexes in plasma samples were detected using ELISA as previously described [39][40][41]. Brie y, a 96-well microtiter plate (Corning) was coated with anti-MPO antibody (4 µg/mL, Bio-Rad) at 4°C overnight, followed by 2-h incubation with 1% bovine plasma albumin in phosphate buffered saline (PBS) at room temperature. Plasma samples were added to the plate and incubated overnight at 4°C. For detection, anti-dsDNA-HRP antibody (1:100 dilution, Roche Diagnostic) was added to the samples and incubated at room temperature for 2 h. The reaction was developed with 3,3′,5,5′tetramethylbenzidine (BD Biosciences) and terminated by the addition of 2 N sulfuric acid. Absorbance was detected at 450 nm using a plate reader (Synergy, BioTek). The plasma levels of BPI, S100A8/A9, hs-CRP, and IL-1β were also measured using ELISA kits according to the manufacturer's instructions (Cusabio Biotech, China). The absorbance was read at 450 nm in a microplate reader. The means were used to calculate the levels of BPI, S100A8/A9, hs-CRP, and IL-1β in corresponding samples based on established standard curves. ELISA experiments were performed in triplicate.

Statistical analysis
Numerical data are shown as means ± standard deviation (SD). Student's t-test or one-way analysis of variance followed by Bonferroni test was used to determine statistical signi cance. Pearson's or Spearman's correlation analysis was used to determine the correlation between parameters. Linear regression analysis was performed to identify independent in uencing factors of BPI. A p-value < 0.05 was considered statistically signi cant. Receiver operating characteristic (ROC) curve analysis was used to evaluate the diagnostic e cacy of plasma BPI levels for ACS and MI. All data analyses were performed using SPSS 22.0 software (IBM Corp.).

Demographic and clinical characteristics of participants at baseline
There were no signi cant differences in age, gender, BMI, smoking history, hypercholesterolemia, diabetes mellitus, atrial brillation, stroke, hypertension, and medications between the ACS and non-ACS groups (Table 1). Data are presented as mean±SD.
*p<0.05 was statistically signi cant when comparisons were made among the non-ACS and ACS group.  Figure 1a).
Levels of Hs-CRP, IL-1β, MPO-DNA, and S100A8/A9 in the ACS and control groups Plasma levels of hs-CRP, IL-1β, MPO-DNA, and S100A8/A9 in the ACS group were signi cantly higher compared with the non-ACS group (all p < 0.05) ( Table 2).

TIMI, GRACE, and Gensini scores in the ACS and control groups
The TIMI, GRACE, and Gensini scores of the ACS group were signi cantly higher compared with the non-ACS group (all p < 0.05) ( Table 2).

Correlation of plasma BPI levels with traditional risk factors and circulating in ammatory biomarkers
To investigate the clinical relevance of high plasma BPI levels in ACS, we examined the correlations of BPI levels with 14 clinical characteristics, 15 blood parameters, four in ammation biomarkers, and ve CAG indexes in ACS patients. We adjusted p-values using the LD-adjusted Bonferroni correction to decrease the probability of Type I errors. The signi cance level was set as p < 0.005 after correction.
We found that plasma BPI levels positively correlated with the TIMI and GRACE scores (r = 0.176, p = 0.003; r = 0.320, p < 0.001) in patients with ACS, and this correlation was more signi cant in the whole cohort (r = 0.486, p < 0.001; r = 0.384, p < 0.001). After correction, the correlations of plasma BPI levels with the TIMI and GRACE scores were con rmed in patients with ACS (all p < 0.001) (Tables 3 and 4).  No signi cant correlations were found between plasma BPI levels and blood platelet counts, although recent RNA-seq data analysis reported the upregulation of BPI in the platelets of patients with STEMI/NSTEMI [42]. Plasma BPI levels positively correlated with blood neutrophil counts (r = 0.266, p < 0.001) in patients with ACS, and this correlation was more signi cant in the whole cohort (r = 0.316, p < 0.001). After correction, the correlations between plasma BPI levels and blood neutrophil counts were con rmed in patients with ACS (p < 0.001) (Tables 3 and 4, Figure 2).
In addition, we found that plasma BPI levels positively correlated with glucose levels (r = 0.138, p = 0.0027) in patients with ACS, and this correlation was also signi cant in the whole cohort (r = 0.135, p = 0.007). After correction, the correlations between plasma BPI levels and glucose levels were con rmed in patients with ACS (p < 0.001) (Tables 3 and 4).
We next examined the correlations of plasma BPI levels with the levels of hs-CRP, IL-1β, MPO-DNA, and S100A8/A9. Plasma BPI levels positively correlated with the levels of hs-CRP, IL-1β, MPO-DNA, and S100A8/A9 (r = 0.746; r = 0.741; r = 0.611; r = 0.529, all p < 0.001) in patients with ACS (Table 3). To identify factors independently associated with BPI, we performed linear regression analysis of plasma BPI levels with the concentrations of hs-CRP, IL-1β, MPO-DNA, and S100A8/A9 using Spearman's correlation analysis. The p-values were corrected using the LD-adjusted Bonferroni correction. After correction, the correlations of BPI levels with the concentrations of hs-CRP, IL-1β, MPO-DNA, and S100A8/A9 persisted in patients with ACS (all p < 0.001). These results indicate that plasma BPI levels positively correlate with the concentrations of hs-CRP, IL-1β, MPO-DNA, and S100A8/A9 (Tables 3 and 4, Figure 3).

Correlation Of Plasma Bpi Level With Cag Results
We further investigated the correlations of BPI levels with CAG results, including "the number of diseased coronary arteries", "calci ed lesions, chronic total occlusion", "in-stent restenosis", and "total number of stents in coronary". We found that plasma BPI levels positively and signi cantly correlated with "the number of diseased coronary arteries", "calci ed lesions, chronic total occlusion", and "total number of stents in coronary" in the whole cohort (r = 0.489, p < 0.001; r = 0.128, p = 0.011; r = 0.264, p < 0.001; r = 0.277, p < 0.001), but not in patients with ACS. However, after LD-adjusted Bonferroni correction, the correlations of plasma BPI levels with the above indexes were not signi cant (Tables 3 and 4, Figure 4).
The severity of coronary atherosclerosis was assessed using the Gensini scoring system based on the CAG results. The Gensini scoring system is a well-recognized and widely used system that evaluates the severity of coronary atherosclerosis in the clinic [38, 43,44]. We also evaluated the correlation between BPI levels and the Gensini score and found that higher plasma BPI levels were associated with higher Gensini scores in patients with ACS (r = 0.263, p < 0.001), as well as in the whole cohort (r = 0.605, p < 0.001) ( Table 3). After correction, the correlation between plasma BPI levels and the Gensini score was con rmed in patients with ACS (p < 0.001) (Tables 3 and 4, Figure 4).
Using the same method, we found that the optimal cut-off of plasma BPI levels for MI was 38.71 ng/ml.
The ROC curves for BPI, CK-MB, and cardiac troponin I (TnI) are compared in Figure 7.

Discussion
In the present study, we measured plasma BPI levels in patients with ACS and assessed their correlation with clinical characteristics and circulating in ammatory biomarkers. The major ndings of this study are as follows: 1) the ACS group had signi cantly higher plasma BPI levels than the non-ACS group; 2) plasma BPI levels positively correlated with hs-CRP, IL-1β, MPO-DNA, and S100A8/A9 levels, as well as blood neutrophil counts in patients with ACS; 3) plasma BPI levels positively correlated with the TIMI and GRACE scores, as well as the Gensini score, which indicated the severity of coronary atherosclerosis in patients with ACS; 4) the diagnostic e cacy of plasma BPI levels for ACS was not inferior to that of hs-CRP, IL-1β, MPO-DNA, S100A8/A9, and blood neutrophil counts, and also not inferior to CK-MB and TnI for the diagnosis of MI. Therefore, plasma BPI levels may serve as a biomarker for ACS.
Atherosclerosis is a chronic in ammatory disease of the vessel wall and a major cause of death worldwide. In ammation substantially contributes to the initiation, progression, and destabilization of atherosclerosis. Although the pathogenic mechanisms underlying atherosclerosis vary across patients, neutrophils are considered the major immune cell that drives the development of this disease [45].
Neutrophils are activated in response to a variety of stimuli, such as in ammatory cytokines, and initiate subsequent effector functions. S100A8/A9 [7,8] and IL-1β [9][10][11] are important in ammatory mediators in the process of atherosclerosis, and hs-CRP is a canonical biomarker well studied in CHD.
In the current study, we found that patients with ACS had signi cantly higher plasma BPI levels compared with non-ACS cases. We then sought to determine if plasma BPI levels correlated with the levels of hs-CRP, IL-1β, and S100A8/A9, as well as blood neutrophil counts. We found that plasma BPI levels positively correlated with the concentrations of circulating in ammatory biomarkers and blood neutrophil counts in ACS (Tables 3 and 4). These data indicate that BPI is involved in the pathogenesis of atherosclerosis, but to better understand how BPI affects atherosclerotic plaques we assessed the severity of coronary atherosclerosis in ACS using the Gensini scoring system [43]. We further examined the correlation between plasma BPI levels and the severity of coronary atherosclerosis and found that higher plasma BPI levels correlated with more severe coronary atherosclerosis in ACS, which was consistent with the results of the whole cohort. All the above ndings support that BPI is an in ammatory marker involved in the pathogenesis of atherosclerosis.
Further analysis revealed that plasma BPI levels were positively correlated with the TIMI and GRACE scores (r = 0.176, p = 0.003; r = 0.320, p < 0.001) in patients with ACS and the whole cohort (r = 0.486, p < 0.001; r = 0.384, p < 0.001), indicating that BPI was correlated with traditional risk factors and may predict the prognosis of ACS. BPI has been shown to participate in lipid metabolism [32,46] owing to its lipid protein structure [47]. Plasma BPI concentrations were also positively associated with the levels of TC, LDL-C, and HDL-C [32]. It has also been suggested that BPI is related to insulin sensitivity and glucose tolerance [31]. In our study, plasma BPI levels were positively correlated with glucose levels (r = 0.138, p = 0.0027) in patients with ACS and the whole cohort (r = 0.135, p = 0.007), which was consistent with the ndings by Gubern et al. [31]. Future investigation of the causal relationship between BPI and traditional risk factors is needed.
When unstable coronary atherosclerotic plaque is accompanied by rupture/erosion in ACS or an acute coronary event, consequences may vary depending on the severity of concomitant in ammation and segmenting coronary artery thrombosis, namely, the interactions among injured endothelium, activated neutrophils, and platelets [48,49]. NETs are strongly implicated in thrombosis by inducing platelet activation and coagulation [50], as well as thrombus formation [51]. During in ammatory stress, NETs are released by neutrophils as structures composed of DNA bers lined with histones and granule proteins, such as neutrophil elastase and MPO. The net-like structures have been detected in atherosclerotic lesions and arterial thrombi in humans and mice. Functionally, NETs induce the activation of endothelial cells, antigen-presenting cells, and platelets [ 52], resulting in proin ammatory immune response. Overall, these ndings suggest that NETs are not only present in plaques and thrombi, but also play an important role in triggering atherosclerotic plaque formation and arterial thrombosis [53]. Previous studies have shown that MPO-DNA is a component of coronary thrombi in acute MI [54], suggesting a prothrombotic state [18], and is correlated with the prognosis of STEMI [19], severity of cardiovascular disease [55,56], and coronary atherosclerosis [57]. S100A8/A9 is bound to NETs both in vitro and in vivo [7,8,58] and is implicated in the pathogenesis of MI. Increased expression of S100A8/A9 was observed in the atherosclerotic plaques of unstable angina pectoris patients [15,59]. In line with these ndings, we found that the ACS group exhibited signi cantly higher levels of MPO-DNA and S100A8/A9 compared with the non-ACS group. Moreover, plasma BPI levels were positively correlated with the levels of MPO-DNA and S100A8/A9. Although MPO-DNA and S100A8/A9 play important roles in NETs formation, the role of BPI in the formation of NETs warrants further exploration.
In recent years, advances in the sensitivity and precision of the cardiac troponin assay have improved the diagnosis of acute chest pain. However, some patients with ACS have elevated cardiac troponin levels due to myocardial injury rather than atherosclerotic plaque rupture/erosion with thrombosis. The combined used of biomarkers related to the pathogenesis of ACS and those for myocardial necrosis may contribute to the early identi cation of these patients for appropriate treatment. In the present study, ROC analysis showed that the diagnostic e cacy of BPI for MI was not inferior to that of CK-MB or TnI. BPI is not a marker of myocardial necrosis, and an elevation in plasma BPI levels only indicates increased aseptic in ammation in patients with MI. Consistently, we found that BPI, hs-CRP, IL-1β, MPO-DNA, S100A8/A9, and blood neutrophil counts all had diagnostic e cacy for ACS, and the diagnostic value of BPI for ACS was superior to that of circulating in ammatory biomarkers. Early measurement of hemostatic plasma markers in patients with ACS may provide pathophysiological information of an acute coronary event [60]. In this study, the subgroup analysis showed that plasma BPI levels were positively correlated with the hemostatic markers in MI, such as D-dimer and brinogen. This may be consistent with the concept that some ruptured/erosional coronary atherosclerotic plaques do not evolve into MI due to lower degrees of in ammation and thrombosis. Hence, BPI may be used to differentiate MI from myocardial injury in ACS.
Several limitations to this study should be addressed. First, this study had a relatively small sample size.
Second, although we excluded patients with acute infection, ongoing chronic in ammatory, autoimmune, or malignant diseases in this observational, cross-sectional study, the increase in plasma BPI levels in patients with ACS needs to be cautiously interpreted. Finally, we did not compare BPI with other risk factors for predicting the prognosis of ACS because we did not analyze major adverse cardiovascular events during follow-up. Future studies are needed to validate our ndings of BPI as a biomarker in ACS and to elucidate the mechanisms underlying BPI-induced in ammation and thrombosis.

Conclusions
In conclusion, this is the rst study that assessed plasma BPI levels in patients with ACS. We showed that circulating BPI levels were increased in patients with ACS and positively correlated with blood neutrophil counts and levels of hs-CRP, IL-1β, MPO-DNA, and S100A8/A9. ROC curve analysis shows that BPI has diagnostic value for both ACS and MI. Also, plasma BPI levels were positively correlated with the severity of coronary atherosclerosis in ACS. Therefore, BPI was associated with systemic in ammation in ACS and may be involved in NET-and neutrophil-mediated atherosclerosis and atherothrombosis. The potential role of BPI as a diagnostic and prognostic biomarker for ACS warrants further investigations.

Declarations
The authors declared that they have no con icts of interest to this work.

Con ict of interest
The authors declared that they have no con icts of interest in this work.

Availability of data and material
The datasets generated and analyzed during the present study are available from the corresponding author upon reasonable request.  Correlations of plasma BPI levels with blood neutrophil and platelet counts.  Correlations of plasma BPI levels with the levels of hs-CRP, IL-1β, MPO-DNA, and S100A8/A9.

Figure 4
Correlations of plasma BPI levels with the TIMI, GRACE, and Gensini scores. ROC analysis of the diagnostic e cacy of plasma BPI, Hs-CRP, and blood neutrophil counts for ACS. Figure 6 ROC analysis of the diagnostic e cacy of plasma BPI, IL-1β, MPO-DNA, and S100A8/A9 for ACS.